Hanover Pitch Book Andover Demand

We are going to build a pitch book for Hanover, one slide at a time. The analysis you perform will build to the final deliverable, so wait until I tell you to build the entire deck. We will start with the opening analysis, that will eventually form the first content slide:

For ZIP 01810 (Andover, MA), build the demand section of an equity-raise deck for a 334-unit market-rate community at 300 Minuteman Road: median household income, renter household income, share of households earning $150K or more, net worth tier versus the Boston MSA, and the median household income of people moving into 01810 versus moving out. Write three investment-highlight bullets from it.

Sep 29

Command

30 lines

mkdir -p hanover && cat > hanover/demand_calcs_01810.py <<'EOF'
# Demand-section calcs: ZIP 01810 vs Boston MSA (14460). Inputs from RealAI Datamart (SuperCensus, as of 2026-08-15)
zip_ = dict(hhi_med=180672, hhi_avg=211050, renter=95223, owner=216283, msa_pct=82,
  br={'lt150':[0,0,0,.003973,.008035,.050592,.076373,.192831], '150_200':.061363,'200_350':.184090,'350_500':.144358,'500p':.278386},
  nw={'0_25':.0496,'25_50':.0182,'50_75':.0128,'75_100':.0112,'100_150':.0210,'150_250':.0333,'250_500':.0697,'500_750':.0629,'750_1m':.0842,'1m_2m':.1442,'2m_3m':.1824,'3m':.3105},
  mig_in=297090, mig_out=251631, n_in=1163, n_out=1284, age_in=41, age_out=52, nw_in=6.1, nw_out=8.1, net=-121, net_pct=-.004821)
msa = dict(hhi_med=124803, hhi_avg=196182, renter=74832,
  br={'lt150':[.00106,.003668,.013787,.019218,.046793,.115057,.121731,.213173], '150_200':.051479,'200_350':.180997,'350_500':.113903,'500p':.119133},
  nw={'0_25':.1180,'25_50':.0378,'50_75':.0277,'75_100':.0248,'100_150':.0475,'150_250':.0714,'250_500':.1294,'500_750':.0890,'750_1m':.0919,'1m_2m':.1144,'2m_3m':.1059,'3m':.1421},
  mig_in=147010, mig_out=166331)
def s150(d): b=d['br']; return b['150_200']+b['200_350']+b['350_500']+b['500p']
def nw1m(d): n=d['nw']; return n['1m_2m']+n['2m_3m']+n['3m']
for k,d in [('01810',zip_),('Boston MSA',msa)]:
    print(k, f"150K+ share={s150(d):.1%}  200K+={s150(d)-d['br']['150_200']:.1%}  NW $1M+={nw1m(d):.1%}  NW<$100K={sum(list(d['nw'].values())[:4]):.1%}")
print(f"Median HHI premium vs MSA: {zip_['hhi_med']/msa['hhi_med']-1:.1%} (${zip_['hhi_med']-msa['hhi_med']:,})")
print(f"Renter median premium vs MSA: {zip_['renter']/msa['renter']-1:.1%} (${zip_['renter']-msa['renter']:,})")
print(f"Renter/owner ratio 01810: {zip_['renter']/zip_['owner']:.1%}")
print(f"150K+ share gap vs MSA: {(s150(zip_)-s150(msa))*100:.1f} pts")
print(f"NW $1M+ gap vs MSA: {(nw1m(zip_)-nw1m(msa))*100:.1f} pts")
print(f"Migration: in ${zip_['mig_in']:,} vs out ${zip_['mig_out']:,}; delta ${zip_['mig_in']-zip_['mig_out']:,} ({zip_['mig_in']/zip_['mig_out']-1:.1%})")
print(f"MSA migration delta ${msa['mig_in']-msa['mig_out']:,}")
print(f"In-mover median vs resident median: {zip_['mig_in']/zip_['hhi_med']-1:.1%}")
# 30% rent-to-income affordability at renter median
print(f"Renter median supports rent at 30% RTI: ${zip_['renter']*.30/12:,.0f}/mo; in-mover median: ${zip_['mig_in']*.30/12:,.0f}/mo")
# Consistency check: bracket distribution vs reported median/avg
cum=sum(zip_['br']['lt150']); print(f"Check 01810: share below $150K={cum:.1%}, below $200K={cum+zip_['br']['150_200']:.1%} -> bracket-implied median lies in $200K-$350K band, reported median ${zip_['hhi_med']:,}")
lb = 150e3*zip_['br']['150_200']+200e3*zip_['br']['200_350']+350e3*zip_['br']['350_500']+500e3*zip_['br']['500p']
print(f"Check 01810: lower-bound mean from $150K+ brackets alone=${lb:,.0f} vs reported avg ${zip_['hhi_avg']:,}")
EOF
python3 hanover/demand_calcs_01810.py

Andover 01810's demand story is newcomer income, not population growth. People moving in earn $45K more than people moving out, while the Boston MSA as a whole is losing income through migration.

Demand scorecard: ZIP 01810 vs. Boston MSA

Demand Profile — ZIP 01810 (Andover, MA) vs. Boston MSA

Metric			01810 Advantage
Median household income	$180,672	$124,803	+44.8%
Renter median household income	$95,223	$74,832	+27.2%
Households earning $150K+ (SuperCensus brackets)	66.8%1	46.6%	+20.3 pts
Households earning $200K+ (ACS 2024 5-yr)	44%	—	—
Median net worth tier	Tier 9 ($1M–$2M)	Tier 7 ($500K–$750K)	Two tiers higher
Households with net worth $1M+	63.7%	36.2%	+27.5 pts
In-mover median HHI	$297,0902	$147,0102	—
Out-mover median HHI	$251,631	$166,331	—
Migration income delta (in minus out)	+$45,459	–$19,321	01810 gains income; MSA loses it

What the numbers say

Income: The ZIP's median household income is $180,672 , 44.8% above the Boston MSA's $124,803 . That puts it in the 82nd percentile of ZIPs in the metro .
Renter income: Renters earn a median $95,223 , 27.2% more than MSA renters .
Wealth: The median household sits in the $1M–$2M net-worth band , two tiers above the MSA's $500K–$750K . That ranks in the 84th percentile of the metro .
Migration: Nearly 1,200 movers were tracked in each direction, so the sample is solid . In-movers are younger (median age 41 vs. 52) . They also have less accumulated wealth: $250K–$500K against $750K–$1M for people leaving . High income without much built-up wealth is a typical renter-by-choice profile, which fits new market-rate rental product.

Draft investment highlights

Wealthy, high-earning submarket. Median household income of $180.7K is about 45% above the Boston MSA, and the median household's net worth is $1M–$2M, two tiers above the metro . Nearly two-thirds of households have a net worth of $1M or more, vs. about one-third across the MSA .
Newcomers are raising the income base. People moving into 01810 have a median household income of $297K, $45K (18%) more than people leaving . The Boston MSA runs the other way, with a $19K income loss through migration . In-movers are 11 years younger and still building wealth, the renter-by-choice profile a new market-rate community is built for .
Well-off renters. Andover's renters earn a median $95K, 27% above metro renters . Renters are only 20% of households , and 44% of households earn over $200K (ACS) .

Before these go on a slide

Use a lower figure for the $150K+ share. The Datamart bracket split (66.8% at $150K+) doesn't match the ZIP's own median and average income. The ACS count of households under $50K (13%) is also far higher than the Datamart's roughly 1% . I'd cite the ACS "44% earn $200K+" on the slide, or about 59% at $150K+ (my estimate from the ACS ranges), rather than 66.8%.
Don't pitch this as a growth market. Net migration is slightly negative (–121 people, –0.5%) . Median income rose only 0.65% over 12 months, vs. 2.1% for the MSA . The deck should sell the quality of new residents, not how many are arriving.
Rent levels depend on the higher earners. At 30% of income, the typical existing renter can afford about $2,381 a month . Rents above that rely on the higher-earning newcomers, so the rent slide needs to address this directly.

When you're ready, the next slide could cover supply and rent comps.

yes plan slide 2 as a rent comp slide for a new 334-unit community at 300 Minuteman Road, Andover, MA: multifamily properties of 120+ units in the Andover submarket built 2016 or later. For each, show units, year built, rent type, in-place rent by bedroom and per square foot, and occupancy, against the Andover submarket's in-place rent by bedroom. Then count how many 120+ unit properties have delivered in the submarket since 2016

slide 3 would cover For ZIP 01810 (Andover, MA), chart median asking rent over the last 12 months and show current occupancy, average new-lease tradeout, and median days on market for leases signed in the past 30 days

Sep 29

Now the calculations.

Command

33 lines

mkdir -p hanover && cat > hanover/slides2_3_calcs.py <<'EOF'
# Slide 2 comp premiums vs Andover submarket; Slide 3 ZIP 01810 trends. Source: RealAI Rent Index as of 2026-09-26
sub = {'1':2437.07,'2':2764.10,'3':3617.50,'avg':2798.95,'psf':2.66}
comps = {  # name: units, yr, 1BR,2BR,3BR, avg, psf, occ
 'The Slate at Andover':(225,2016,2408.32,2962.80,4171.46,2719.28,2.81,.9556),
 'Berry Farms':(196,2016,2515.42,2991.56,3736.07,2848.11,2.98,.9337),
 'The Point at Merrimack River*':(248,2018,2561.99,3156.15,3886.40,2834.95,2.94,.9677),
 'Haven North Andover*':(192,2019,2512.29,3176.32,None,2836.58,2.71,.9531),
 'Avalon North Andover*':(221,2022,2475.56,2992.63,4051.36,2763.34,2.95,.9502),
 '505 Sutton*':(136,2022,2799.66,3248.26,None,2854.92,3.26,.8676)}
def prem(v,k): return None if v is None else v/sub[k]-1
for n,c in comps.items():
    p=[prem(c[2],'1'),prem(c[3],'2'),prem(c[4],'3')]
    print(f"{n:32s} 1BR {p[0]:+.1%} | 2BR {p[1]:+.1%} | 3BR {'n/a' if p[2] is None else f'{p[2]:+.1%}'} | avg {c[5]/sub['avg']-1:+.1%} | psf {c[6]/sub['psf']-1:+.1%}")
def summ(names,label):
    sel=[comps[n] for n in names]; U=sum(s[0] for s in sel)
    out=[label, f"units={U}", f"wtd occ={sum(s[0]*s[7] for s in sel)/U:.1%}", f"wtd avg in-place=${sum(s[0]*s[5] for s in sel)/U:,.0f}"]
    for i,k in [(2,'1'),(3,'2'),(4,'3')]:
        v=[s[i] for s in sel if s[i]]; a=sum(v)/len(v); out.append(f"{k}BR simple avg ${a:,.0f} ({a/sub[k]-1:+.1%} vs sub, n={len(v)})")
    print(' | '.join(out))
summ(['The Slate at Andover','Berry Farms'],'Strict (Andover submarket)')
summ(list(comps),'Expanded (6 props)')
# Slide 3
ts = {'2025-09':2732,'2025-10':2750,'2025-11':2620,'2025-12':2590,'2026-01':2565,'2026-02':2628,'2026-03':2640.5,'2026-04':2640,'2026-05':2800,'2026-06':2824,'2026-07':2700,'2026-08':2621}
v=list(ts.values())
print(f"Sep25->Aug26 median asking: {v[-1]/v[0]-1:+.1%}; trough ${min(v):,.0f} Jan-26, peak ${max(v):,.0f} Jun-26 ({max(v)/min(v)-1:+.1%})")
print(f"Current (9/26/26) median asking $2,820 vs in-place median $2,559: spread {2820/2559-1:+.1%}")
tr={'2026-03':(.0305,13),'2026-04':(.004858,26),'2026-05':(.032766,23),'2026-06':(.051478,28),'2026-07':(.035834,18),'2026-08':(.059155,23)}
print(f"Mar-Aug 2026 lease-weighted new-lease tradeout: {sum(p*n for p,n in tr.values())/sum(n for _,n in tr.values()):+.1%} on {sum(n for _,n in tr.values())} leases")
tr2={'2025-09':(-.002938,11),'2025-10':(.04985,10),'2025-11':(-.021016,14),'2025-12':(-.003027,160),'2026-01':(.000813,20),'2026-02':(-.010955,14)}
print(f"Sep25-Feb26 lease-weighted tradeout: {sum(p*n for p,n in tr2.values())/sum(n for _,n in tr2.values()):+.1%} on {sum(n for _,n in tr2.values())} leases")
EOF
python3 hanover/slides2_3_calcs.py

Slide 2: Rent comps

Under the Datamart's "Andover" submarket, only 2 properties with 120+ units have been built since 2016, and both opened in 2016 . That count is too narrow for the slide. The Datamart puts four more 2018–2022 deliveries in the neighboring "Lawrence" submarket , including The Point at Merrimack River, which is in ZIP 01810. Counting them, 6 properties (1,218 units) have delivered across Andover and North Andover since 2016, with none completed since 2022 . The comp set below shows all six, with the four reassigned properties flagged.

Rent Comps — 120+ Unit Communities Built 2016+ vs. Andover Submarket In-Place Rent

Property	City / ZIP	Datamart Submarket	Units	Year Built	Rent Type	1BR In-Place	2BR In-Place	3BR In-Place	Avg In-Place $/SF	Occupancy
Andover submarket (strict criteria)
	Andover 01810	Andover	225	2016	Market & Affordable	$2,408 ($2.94/SF)	$2,963 ($2.58/SF)	$4,171 ($2.76/SF)	$2.81	95.6%
	North Andover 01845	Andover	196	2016	Market	$2,515 ($3.18)	$2,992 ($2.80)	$3,736 ($3.05)	$2.98	93.4%
Same towns, Datamart assigns to Lawrence submarket
	Andover 01810	Lawrence	248	2018	Market & Affordable	$2,562 ($3.12)	$3,156 ($2.64)	$3,886 ($2.66)	$2.94	96.8%
	North Andover 01845	Lawrence	192	2019	Market	$2,512 ($2.90)	$3,176 ($2.51)	—2	$2.71	95.3%
	North Andover 01845	Lawrence	221	2022	Market	$2,476 ($3.17)	$2,993 ($2.58)	$4,051 ($2.72)	$2.953	95.0%
	North Andover 01845	Lawrence	136	2022	Market	$2,800 ($3.09)	$3,248 ($2.94)	—2	$3.263	86.8%
Submarket Benchmark
Andover Submarket (All MF)4	—	—	—	—	—	$2,437 ($3.03)	$2,764 ($2.49)	$3,618 ($2.72)	$2.66	93.2%

Rent Comps: Andover & Lawrence Submarkets

What the comps show

The premium for new product is in 2- and 3-bedrooms, not 1-bedrooms.
- Against submarket in-place rent, the comps' 2BRs average +7.7% (strict set) / +11.7% (all six). 3BRs average about +9.5% .
- The 1BR premium is only +1.0% / +4.5% . New 1BRs rent for about the same as the existing stock.
Average rents fall in a tight range. Across all six, the unit-weighted average in-place rent is $2,805 and occupancy is 94.3% . Every property's average is between $2,719 and $2,855.
The newest, priciest property is struggling to fill. 505 Sutton gets the set's highest rent per SF at $3.26 but is only 86.8% occupied . That is the clearest sign of where rents top out.
This ties back to slide 1. The average comp rent ($2,805) is above the $2,381 a typical existing Andover renter can afford . The subject's lease-up depends on the higher-earning newcomers.
The supply case is strong, but label it correctly. Andover itself permitted only 9 new homes in 2023 . North Andover's 2022–23 deliveries were zoning-enabled corridor projects . On the slide, say "no 120+ unit deliveries since 2022 across Andover and North Andover." Don't say "2 since 2016," which depends on how the Datamart draws the submarket.

Slide 3: ZIP 01810 leasing momentum

ZIP 01810 — Median Asking vs. In-Place Rent, Trailing 12 Months

ZIP 01810 Leasing Snapshot (as of 2026-09-26)

Metric	Value	Context
Median asking rent	$2,8201	+0.8% year over year; 10.2% above the $2,559 median in-place rent
Occupancy	96.7%	97.1% a year ago
Avg new-lease tradeout	+5.85% (+$159)	Last 30 days, small sample
Median days on market, leases signed past 30 days	59 days	36 leases signed

What the numbers show

Asking rents move with the season rather than trending. The current median is $2,820, up 0.8% from a year earlier . On the monthly series, rents fell 4.1% from September 2025 to August 2026, bottoming at $2,565 in January and peaking at $2,824 in June . Monthly medians are based on as few as 69 listings , so the chart is noisy.
Occupancy is tight. It is 96.7% now vs. 97.1% a year ago .
The +5.85% tradeout is real, but the sample is small. It's based on only 23 leases in August . The steadier figure: new leases have run +3.6% over the prior tenant's rent since March (131 leases), vs. –0.2% over the six months before (229 leases) . New-lease pricing has turned up.
Units take about two months to lease (59 days) despite the high occupancy . Renters in 01810 aren't competing for units. The 10.2% gap between asking and in-place rents is being closed gradually, not taken at once .
December 2025 looks like a data problem. It shows 166 leases signed and 217 median days on market, likely a batch catch-up . I'd leave it off any leasing-velocity chart.

Decisions before the deck build

Comp set: I recommend using all six on the slide, with the strict two-property result in a footnote. Otherwise it looks like we left out Avalon and The Point.
Headline tradeout: I'd lead with the +3.6% six-month figure and show +5.85% as the latest reading.

yes, let's use all 6, and include the footnotes, and lead with the 6-month figure

Slide 4 is Sponsor track record: Hanover commands a premium (Atlanta)

For Hanover Edgewood (Atlanta, GA) and Hanover Midtown (Atlanta, GA), show in-place rent by bedroom and per square foot versus each property's submarket in-place rent by bedroom, and median days on market for leases signed in the past 30 days versus the submarket.

Sep 29

Noted for slides 2 and 3: all six comps with footnotes, and the +3.6% six-month tradeout as the headline figure.

Slide 4: Hanover track record in Atlanta

The "commands a premium" headline holds for Hanover Midtown and for leasing speed at both properties. It does not hold on rent for Hanover Edgewood, and both properties are less occupied than their submarkets. As written, the slide would claim more than the data supports.

Hanover Atlanta Portfolio vs. Submarket — In-Place Rent and Leasing Velocity

Metric			Premium			Premium
Property
Units / Year built	422 / 2022	—	—	421 / 2023	—	—
In-place rent by bedroom
Studio	$1,418 ($2.60/SF)	$1,437 ($2.68/SF)	–1.3%	$1,897 ($3.38/SF)	$1,808 ($3.42/SF)	+4.9%
1BR	$1,806 ($2.24)	$1,608 ($2.15)	+12.3%	$2,481 ($2.92)	$2,293 ($2.87)	+8.2%
2BR	$2,339 ($2.04)	$1,858 ($1.73)	+25.9%	$3,396 ($2.51)	$3,022 ($2.44)	+12.4%
3BR	$2,870 ($1.96)	$2,177 ($1.60)	+31.8%	$4,202 ($2.63)	$4,025 ($2.74)	+4.4%
Average	$1,9901	$1,744	+14.1%	$2,7401	$2,496	+9.8%
Average $/SF	$2.20	$1.97	+11.7%	$2.86	$2.81	+1.8%
Leasing
Median days on market, leases signed past 30 days	352	63	28 days faster	68	82	14 days faster
Leases signed past 30 days	23	197	—	46	188	—
Occupancy	89.8%3	94.0%	–4.1 pts	80.5%3	85.3%	–4.8 pts

What holds up

Hanover leases faster at both properties. Edgewood's units lease in a median 35 days vs. 63 for its submarket. Midtown's lease in 68 days vs. 82 . This is the point to lead the slide with.
Midtown's rent premium is real. Average rent is $2,740, 9.8% above the Midtown submarket, with a premium in every bedroom type from +4.4% to +12.4% . But rent per SF is only 1.8% above the submarket . Most of the premium comes from larger units (1,066 SF average) , not higher pricing per foot.

What investors will push back on

Edgewood's premium depends on which benchmark you use.
- Against its Datamart submarket, Chandler-McAfee/West Belvedere Park, Edgewood is +14.1% overall and +26–32% on 2–3BRs .
- Against its own ZIP, 30307, it rents 2.6% below average . That ZIP sits in a different Datamart submarket than the one assigned to the property .
- An investor who checks Edgewood-area rents will find the ZIP comparison. I'd show both, or present Edgewood as a leasing-speed story only.
Both properties are less occupied than their submarkets.
- Edgewood is at 89.8% and Midtown at 80.5%, about 4–5 points below their submarkets .
- Short days on market alongside low occupancy can mean a lease-up that is still under way. It can also mean rent is being set with occupancy given up to get it. Either way, "commands a premium" invites the question.
- Adjusted for occupancy, the lead shrinks: revenue per unit is +9.1% for Edgewood and +3.6% for Midtown against their submarkets .
Ownership check. The Datamart shows only the fee-owner entities: "HANIMPJV EASTSIDE GA LLC" (with Toll Southeast) at Edgewood and "USHAN 1230 WEST PEACHTREE" at Midtown . Confirm Hanover's role (developer, owner or manager) before the slide calls these the sponsor's track record.

Suggested headline: "Hanover product leases 14–28 days faster than its submarket and earns a 10% rent premium in Midtown."

The occupancy gap and Edgewood's ZIP comparison would go in the footnotes. Ideally we'd add each property's lease-up history, since stabilization date is what explains the occupancy. I can pull the monthly occupancy series for both before the deck build if you want it.

Not included in this share

great, let's put this together in the pitch book. Here is Hanover's logo, and use whatever branding/colors you found on the website and research to help you out.

Sep 29

Content

311 lines

"""Hanover | 300 Minuteman Road pitch book. All figures from RealAI Datamart pulls (as of 2026-09-26 rent index;
2026-08-15 SuperCensus) and sandbox calcs in demand_calcs_01810.py, slides2_3_calcs.py, slide4_calcs.py."""
from pptx import Presentation
from pptx.util import Inches, Pt, Emu
from pptx.dml.color import RGBColor
from pptx.enum.text import PP_ALIGN, MSO_ANCHOR
from pptx.chart.data import CategoryChartData
from pptx.enum.chart import XL_CHART_TYPE, XL_LEGEND_POSITION
from pptx.enum.shapes import MSO_SHAPE

GREEN = RGBColor(0x11, 0x57, 0x40)
DARK = RGBColor(0x23, 0x1F, 0x20)
GRAY = RGBColor(0x6B, 0x6B, 0x6B)
LIGHT = RGBColor(0xE8, 0xF0, 0xEC)
MID = RGBColor(0x7F, 0xA8, 0x98)
WHITE = RGBColor(0xFF, 0xFF, 0xFF)
RED = RGBColor(0xA3, 0x3B, 0x2E)
FONT = "Arial"

prs = Presentation()
prs.slide_width, prs.slide_height = Inches(13.333), Inches(7.5)
BLANK = prs.slide_layouts[6]
W = prs.slide_width


def tb(slide, x, y, w, h, text="", size=12, bold=False, color=DARK, align=PP_ALIGN.LEFT, anchor=MSO_ANCHOR.TOP, italic=False):
    box = slide.shapes.add_textbox(x, y, w, h)
    tf = box.text_frame
    tf.word_wrap = True
    tf.margin_left = tf.margin_right = Inches(0.05)
    tf.margin_top = tf.margin_bottom = Inches(0.03)
    tf.vertical_anchor = anchor
    lines = text if isinstance(text, list) else [text]
    for i, ln in enumerate(lines):
        p = tf.paragraphs[0] if i == 0 else tf.add_paragraph()
        p.alignment = align
        runs = ln if isinstance(ln, list) else [(ln, bold, color)]
        for r in runs:
            t, b, c = (r + (color,))[:3] if len(r) == 2 else r
            run = p.add_run()
            run.text = t
            run.font.name = FONT
            run.font.size = Pt(size)
            run.font.bold = b
            run.font.italic = italic
            run.font.color.rgb = c
    return box


def rect(slide, x, y, w, h, fill, line=None):
    s = slide.shapes.add_shape(MSO_SHAPE.RECTANGLE, x, y, w, h)
    s.fill.solid()
    s.fill.fore_color.rgb = fill
    if line is None:
        s.line.fill.background()
    else:
        s.line.color.rgb = line
    s.shadow.inherit = False
    return s


def header(slide, kicker, title, num):
    rect(slide, 0, 0, W, Inches(0.12), GREEN)
    tb(slide, Inches(0.5), Inches(0.3), Inches(10.5), Inches(0.3), kicker.upper(), 11, True, GREEN)
    tb(slide, Inches(0.5), Inches(0.58), Inches(11.2), Inches(0.9), title, 22, True, DARK)
    slide.shapes.add_picture("logo.png", Inches(11.35), Inches(0.3), height=Inches(0.55))
    ln = slide.shapes.add_connector(1, Inches(0.5), Inches(1.45), W - Inches(0.5), Inches(1.45))
    ln.line.color.rgb = MID
    ln.line.width = Pt(0.75)
    # footer
    tb(slide, Inches(0.5), Inches(7.1), Inches(9), Inches(0.3),
       "300 Minuteman Road  |  Andover, MA  |  Confidential – For Discussion Purposes Only", 8, False, GRAY)
    tb(slide, W - Inches(1.2), Inches(7.1), Inches(0.7), Inches(0.3), str(num), 8, True, GREEN, PP_ALIGN.RIGHT)


def table(slide, x, y, w, rows, col_w, font=10, header_fill=GREEN, row_h=0.32, bold_rows=(), shade_rows=(), section_rows=()):
    nr, nc = len(rows), len(rows[0])
    shp = slide.shapes.add_table(nr, nc, x, y, w, Inches(row_h * nr))
    t = shp.table
    tblPr = shp._element.graphic.graphicData.tbl.tblPr
    # remove default style banding look
    tblPr.set("bandRow", "0"); tblPr.set("firstRow", "0")
    for j, cw in enumerate(col_w):
        t.columns[j].width = Inches(cw)
    for i, r in enumerate(rows):
        t.rows[i].height = Inches(row_h)
        for j, val in enumerate(r):
            c = t.cell(i, j)
            c.margin_left = c.margin_right = Inches(0.06)
            c.margin_top = c.margin_bottom = Inches(0.02)
            c.vertical_anchor = MSO_ANCHOR.MIDDLE
            tf = c.text_frame
            tf.word_wrap = True
            p = tf.paragraphs[0]
            p.alignment = PP_ALIGN.LEFT if j == 0 else PP_ALIGN.CENTER
            run = p.add_run()
            run.text = str(val)
            run.font.name = FONT
            run.font.size = Pt(font)
            c.fill.solid()
            if i == 0:
                c.fill.fore_color.rgb = header_fill
                run.font.bold = True
                run.font.color.rgb = WHITE
            elif i in section_rows:
                c.fill.fore_color.rgb = MID
                run.font.bold = True
                run.font.color.rgb = WHITE
            elif i in shade_rows:
                c.fill.fore_color.rgb = LIGHT
                run.font.bold = True
                run.font.color.rgb = DARK
            else:
                c.fill.fore_color.rgb = WHITE if i % 2 else RGBColor(0xF6, 0xF8, 0xF7)
                run.font.color.rgb = DARK
                run.font.bold = i in bold_rows
    if section_rows:
        for i in section_rows:
            t.cell(i, 0).merge(t.cell(i, nc - 1))
    return t


def kpi(slide, x, y, w, h, value, label, sub=""):
    rect(slide, x, y, w, h, LIGHT)
    rect(slide, x, y, Inches(0.08), h, GREEN)
    tb(slide, x + Inches(0.2), y + Inches(0.1), w - Inches(0.3), Inches(0.6), value, 24, True, GREEN)
    tb(slide, x + Inches(0.2), y + Inches(0.68), w - Inches(0.3), Inches(0.35), label, 11, True, DARK)
    if sub:
        tb(slide, x + Inches(0.2), y + Inches(1.0), w - Inches(0.3), h - Inches(1.05), sub, 9, False, GRAY)


def bullets(slide, x, y, w, h, items, size=12, gap=6):
    box = slide.shapes.add_textbox(x, y, w, h)
    tf = box.text_frame
    tf.word_wrap = True
    for i, (head, body) in enumerate(items):
        p = tf.paragraphs[0] if i == 0 else tf.add_paragraph()
        p.space_after = Pt(gap)
        r0 = p.add_run(); r0.text = "■  "; r0.font.size = Pt(size - 3); r0.font.color.rgb = GREEN; r0.font.name = FONT
        r1 = p.add_run(); r1.text = head + " "; r1.font.bold = True; r1.font.size = Pt(size); r1.font.color.rgb = DARK; r1.font.name = FONT
        r2 = p.add_run(); r2.text = body; r2.font.size = Pt(size); r2.font.color.rgb = DARK; r2.font.name = FONT
    return box


def footnote(slide, lines, y=6.35, h=0.75):
    tb(slide, Inches(0.5), Inches(y), Inches(12.3), Inches(h), lines, 7.5, False, GRAY)


# ---------------- Slide 1: Cover ----------------
s = prs.slides.add_slide(BLANK)
rect(s, 0, 0, W, prs.slide_height, GREEN)
rect(s, 0, Inches(5.9), W, Inches(1.6), DARK)
s.shapes.add_picture("logo_white.png", Inches(0.8), Inches(0.8), height=Inches(1.1))
tb(s, Inches(0.8), Inches(2.55), Inches(11), Inches(0.4), "EQUITY OFFERING  |  INVESTOR PRESENTATION", 14, True, MID)
tb(s, Inches(0.8), Inches(3.0), Inches(11.5), Inches(1.0), "300 Minuteman Road", 44, True, WHITE)
tb(s, Inches(0.8), Inches(3.9), Inches(11.5), Inches(0.6), "334-Unit Market-Rate Community  |  Andover, Massachusetts", 22, False, WHITE)
tb(s, Inches(0.8), Inches(6.2), Inches(8), Inches(0.4), "September 2026", 14, True, WHITE)
tb(s, Inches(0.8), Inches(6.6), Inches(11.5), Inches(0.5),
   "Confidential – For discussion purposes only. Not an offer to sell or a solicitation of an offer to buy securities.", 9, False, MID)

# ---------------- Slide 2: Demand ----------------
s = prs.slides.add_slide(BLANK)
header(s, "Market Demand  |  ZIP 01810 (Andover, MA)",
       "An affluent, highly educated submarket whose new arrivals out-earn those leaving by $45K", 2)
rows = [["Metric", "ZIP 01810", "Boston MSA", "01810 Advantage"],
        ["Median household income", "$180,672", "$124,803", "+44.8%"],
        ["Renter median household income", "$95,223", "$74,832", "+27.2%"],
        ["Households earning $200K+ (ACS)¹", "44%", "—", "—"],
        ["Median household net worth", "$1M – $2M", "$500K – $750K", "Two tiers higher"],
        ["Households with net worth $1M+", "63.7%", "36.2%", "+27.5 pts"],
        ["In-mover median HHI", "$297,090", "$147,010", "—"],
        ["Out-mover median HHI", "$251,631", "$166,331", "—"],
        ["Migration income delta (in – out)²", "+$45,459", "–$19,321", "01810 gains; MSA loses"]]
table(s, Inches(0.5), Inches(1.7), Inches(7.0), rows, [2.7, 1.3, 1.35, 1.65], font=10.5, row_h=0.46, shade_rows=(8,))
tb(s, Inches(7.9), Inches(1.65), Inches(4.9), Inches(0.4), "INVESTMENT HIGHLIGHTS", 12, True, GREEN)
bullets(s, Inches(7.9), Inches(2.05), Inches(4.95), Inches(4.3), [
    ("Wealthy, high-earning submarket.",
     "Median household income of $180.7K is ~45% above the Boston MSA, and the median household's net worth "
     "is $1M–$2M, two tiers above the metro. Nearly two-thirds of households have a net worth of $1M or more, vs. about one-third across the MSA."),
    ("Newcomers are raising the income base.",
     "In-movers earn a median $297K, $45K (18%) more than out-movers, while the Boston MSA loses $19K of income through migration. "
     "In-movers are 11 years younger (41 vs. 52) and still building wealth, the renter-by-choice profile a new market-rate community is built for."),
    ("Well-off renters.",
     "Andover renters earn a median $95K, 27% above metro renters. Renters are only 20% of households, "
     "and 44% of all households earn over $200K."),
], size=11, gap=9)
footnote(s, ["¹ ACS 2024 5-year via Census Reporter. The ACS figure is shown instead of the Datamart income brackets, which don't match the ZIP's reported median.  "
             "² Median household income of households tracked moving in vs. out, Aug 2024 – Jul 2026 (1,163 in / 1,284 out).",
             "Note: Net migration is roughly flat (–121 people, –0.5%); the story is the quality of new residents, not how many are arriving.  "
             "Source: RealAI SuperCensus (as of Aug 15, 2026); US Census ACS."], y=6.4)

# ---------------- Slide 3: Rent comps ----------------
s = prs.slides.add_slide(BLANK)
header(s, "Competitive Set  |  120+ Unit Communities Built 2016+",
       "Six communities built in nine years, none since 2022 – new product earns a premium on 2BRs and 3BRs", 3)
rows = [["Property", "City / ZIP", "Units", "Built", "Rent Type", "1BR", "2BR", "3BR", "Avg $/SF", "Occ."],
        ["Andover submarket", "", "", "", "", "", "", "", "", ""],
        ["The Slate at Andover", "Andover 01810", "225", "2016", "Mkt & Aff.", "$2,408", "$2,963", "$4,171", "$2.81", "95.6%"],
        ["Berry Farms", "N. Andover 01845", "196", "2016", "Market", "$2,515", "$2,992", "$3,736", "$2.98", "93.4%"],
        ["Same towns, Datamart \"Lawrence\" submarket¹", "", "", "", "", "", "", "", "", ""],
        ["The Point at Merrimack River", "Andover 01810", "248", "2018", "Mkt & Aff.", "$2,562", "$3,156", "$3,886", "$2.94", "96.8%"],
        ["Haven North Andover", "N. Andover 01845", "192", "2019", "Market", "$2,512", "$3,176", "—", "$2.71", "95.3%"],
        ["Avalon North Andover", "N. Andover 01845", "221", "2022", "Market", "$2,476", "$2,993", "$4,051", "$2.95", "95.0%"],
        ["505 Sutton", "N. Andover 01845", "136", "2022", "Market", "$2,800", "$3,248", "—", "$3.26", "86.8%"],
        ["Comp set (6)²", "", "1,218", "", "", "$2,546", "$3,088", "$3,961", "", "94.3%"],
        ["Andover submarket in-place", "", "", "", "", "$2,437", "$2,764", "$3,618", "$2.66", "93.2%"],
        ["Comp premium vs. submarket", "", "", "", "", "+4.5%", "+11.7%", "+9.5%", "", ""]]
table(s, Inches(0.5), Inches(1.65), Inches(8.6), rows,
      [2.35, 1.3, 0.55, 0.5, 0.85, 0.65, 0.65, 0.65, 0.6, 0.5], font=9, row_h=0.36,
      section_rows=(1, 4), shade_rows=(9, 10, 11))
tb(s, Inches(9.4), Inches(1.6), Inches(3.5), Inches(0.4), "KEY TAKEAWAYS", 12, True, GREEN)
bullets(s, Inches(9.4), Inches(2.0), Inches(3.45), Inches(4.4), [
    ("Supply is limited.", "No 120+ unit community has delivered in Andover or North Andover since 2022. Andover permitted only 9 new homes in 2023."),
    ("The premium is in larger units.", "Compared with submarket in-place rents, comp 2BRs run +11.7% and 3BRs +9.5%. 1BRs run only +4.5%."),
    ("Rents cluster tightly.", "Comp average in-place rents range from $2,719 to $2,855. The unit-weighted average is $2,805, with 94.3% occupancy."),
    ("Where rents top out.", "505 Sutton gets the highest rent at $3.26/SF but is only 86.8% occupied."),
], size=10.5, gap=8)
footnote(s, ["¹ Under strict criteria (Datamart \"Andover\" submarket, 120+ units, built 2016+), only The Slate and Berry Farms qualify. The Datamart assigns four "
             "same-town 2018–2022 deliveries, including The Point in ZIP 01810, to the adjacent Lawrence submarket; they are included here as direct competitors.  "
             "² Rent by bedroom is a simple average of the properties reporting that unit type (3BR n = 4); occupancy is weighted by units.",
             "In-place rent = average rent on leased units. Studios omitted (reported only by Avalon, $2,123, and 505 Sutton, $2,225). "
             "The subject is not yet in the dataset.  Source: RealAI Rent Index (as of Sep 26, 2026); Andover MA News; Town of North Andover."],
         y=6.35)

# ---------------- Slide 4: Leasing momentum ----------------
s = prs.slides.add_slide(BLANK)
header(s, "Leasing Momentum  |  ZIP 01810",
       "Tight occupancy and new-lease pricing turning upward – new leases up 3.6% over prior rents since March", 4)
cd = CategoryChartData()
cd.categories = ["Sep-25", "Oct-25", "Nov-25", "Dec-25", "Jan-26", "Feb-26", "Mar-26", "Apr-26", "May-26", "Jun-26", "Jul-26", "Aug-26"]
cd.add_series("Median asking rent", (2732, 2750, 2620, 2590, 2565, 2628, 2640.5, 2640, 2800, 2824, 2700, 2621))
cd.add_series("Median in-place rent", (2550, 2545, 2544, 2540, 2540.5, 2541, 2545, 2530, 2541, 2545, 2550, 2559))
gf = s.shapes.add_chart(XL_CHART_TYPE.LINE_MARKERS, Inches(0.5), Inches(1.95), Inches(7.6), Inches(4.2), cd)
ch = gf.chart
ch.has_legend = True
ch.legend.position = XL_LEGEND_POSITION.BOTTOM
ch.legend.include_in_layout = False
ch.legend.font.size = Pt(10); ch.legend.font.name = FONT
va = ch.value_axis
va.minimum_scale, va.maximum_scale, va.major_unit = 2400, 2900, 100
va.tick_labels.font.size = Pt(9); va.tick_labels.number_format = '"$"#,##0'; va.tick_labels.number_format_is_linked = False
va.major_gridlines.format.line.color.rgb = RGBColor(0xDD, 0xDD, 0xDD)
va.format.line.fill.background()
ca = ch.category_axis
ca.tick_labels.font.size = Pt(9)
for ser, col in zip(ch.plots[0].series, [GREEN, GRAY]):
    ser.format.line.color.rgb = col
    ser.format.line.width = Pt(2.5)
    ser.smooth = False
    ser.marker.format.fill.solid(); ser.marker.format.fill.fore_color.rgb = col
    ser.marker.format.line.color.rgb = col
tb(s, Inches(0.5), Inches(1.6), Inches(7.6), Inches(0.35), "Median asking vs. in-place rent, trailing 12 months (monthly)", 11, True, DARK)
kx, kw, kh = Inches(8.5), Inches(2.1), Inches(1.95)
kpi(s, kx, Inches(1.65), kw, kh, "$2,820", "Median asking rent", "Current; +0.8% year over year; 10.2% above the $2,559 in-place median")
kpi(s, kx + Inches(2.25), Inches(1.65), kw, kh, "96.7%", "Occupancy", "97.1% a year ago")
kpi(s, kx, Inches(3.75), kw, kh, "+3.6%", "New-lease tradeout", "Mar–Aug 2026, 131 leases (–0.2% in the prior 6 mo.). Latest 30 days: +5.85%")
kpi(s, kx + Inches(2.25), Inches(3.75), kw, kh, "59 days", "Median days on market", "Leases signed in the past 30 days (36 leases)")
tb(s, Inches(8.5), Inches(5.85), Inches(4.35), Inches(0.5),
   "Rents are seasonal: lowest in January ($2,565), highest in June ($2,824).", 9.5, True, GREEN)
footnote(s, ["Tradeout = change in rent on a newly signed lease vs. the prior tenant's lease on the same unit, weighted by leases signed each month. "
             "The latest 30-day +5.85% (+$159) rests on 23 leases. Monthly asking medians rest on 69–121 listed units and are directionally reliable.",
             "Coverage: ~1,213 tracked multifamily units in ZIP 01810.  Source: RealAI Rent Index (current snapshot as of Sep 26, 2026; monthly series through Aug 2026)."],
         y=6.45, h=0.6)

# ---------------- Slide 5: Sponsor track record ----------------
s = prs.slides.add_slide(BLANK)
header(s, "Sponsor Track Record  |  Atlanta, GA",
       "Hanover product leases 14–28 days faster than its submarket and earns a 10% rent premium in Midtown", 5)
rows = [["Metric", "Hanover Edgewood", "Submarket¹", "Premium", "Hanover Midtown", "Midtown Submarket", "Premium"],
        ["Units / Year built", "422 / 2022", "—", "—", "421 / 2023", "—", "—"],
        ["Studio", "$1,418  ($2.60/SF)", "$1,437  ($2.68)", "–1.3%", "$1,897  ($3.38/SF)", "$1,808  ($3.42)", "+4.9%"],
        ["1BR", "$1,806  ($2.24)", "$1,608  ($2.15)", "+12.3%", "$2,481  ($2.92)", "$2,293  ($2.87)", "+8.2%"],
        ["2BR", "$2,339  ($2.04)", "$1,858  ($1.73)", "+25.9%", "$3,396  ($2.51)", "$3,022  ($2.44)", "+12.4%"],
        ["3BR", "$2,870  ($1.96)", "$2,177  ($1.60)", "+31.8%", "$4,202  ($2.63)", "$4,025  ($2.74)", "+4.4%"],
        ["Average in-place rent", "$1,990", "$1,744", "+14.1%", "$2,740", "$2,496", "+9.8%"],
        ["Average $/SF", "$2.20", "$1.97", "+11.7%", "$2.86", "$2.81", "+1.8%"],
        ["Median days on market²", "35 days", "63 days", "28 days faster", "68 days", "82 days", "14 days faster"],
        ["Occupancy", "89.8%", "94.0%", "–4.1 pts", "80.5%", "85.3%", "–4.8 pts"]]
table(s, Inches(0.5), Inches(1.65), Inches(8.9), rows, [1.85, 1.35, 1.2, 1.05, 1.35, 1.2, 0.9], font=9.5, row_h=0.4, shade_rows=(8,))
tb(s, Inches(9.7), Inches(1.6), Inches(3.2), Inches(0.4), "WHAT IT SHOWS", 12, True, GREEN)
bullets(s, Inches(9.7), Inches(2.0), Inches(3.15), Inches(4.3), [
    ("Faster leasing.", "Units at both communities lease 17–44% faster than their submarkets, a consistent result across two different neighborhoods."),
    ("Midtown premium.", "Rents run 9.8% above the Midtown submarket and higher in every unit type. Larger units (1,066 SF average) account for most of it; $/SF is +1.8%."),
    ("Edgewood 2BR/3BR premium.", "+26% to +32% vs. the assigned submarket; see footnote 1 for the ZIP comparison."),
], size=10.5, gap=9)
footnote(s, ["¹ Edgewood's Datamart-assigned submarket is Chandler-McAfee/West Belvedere Park. Against its own ZIP (30307), Edgewood rents 2.6% below average ($1,990 vs. $2,042).  "
             "² Median days on market for leases signed in the past 30 days (Edgewood: 23 leases; Midtown: 46).",
             "Both communities are less occupied than their submarkets (see table); adjusted for occupancy, revenue per unit is +9.1% (Edgewood) and +3.6% (Midtown) vs. submarket. "
             "In-place rent = average rent on leased units; figures in parentheses are $/SF.  Source: RealAI Rent Index (as of Sep 26, 2026)."],
         y=6.3, h=0.8)

# ---------------- Slide 6: Sources & disclaimer ----------------
s = prs.slides.add_slide(BLANK)
header(s, "Appendix", "Sources, Definitions & Important Information", 6)
bullets(s, Inches(0.5), Inches(1.75), Inches(12.3), Inches(4.8), [
    ("RealAI SuperCensus (Aug 15, 2026):", "Household income (median, renter, owner), net worth distribution and tiers, and migration cohorts "
     "(tracked households moving into / out of ZIP 01810, Aug 2024 – Jul 2026; 1,163 in / 1,284 out)."),
    ("RealAI Rent Index (Sep 26, 2026):", "Property and geography in-place and asking rents by bedroom and per SF, physical occupancy, new-lease tradeout, "
     "and days on market. Monthly series through Aug 2026."),
    ("US Census ACS 2024 5-year (via Census Reporter):", "Household income distribution for ZCTA 01810 (44% of households earn over $200K; median $172,606)."),
    ("Supply context:", "Andover MA News (Sep 16, 2025) on Andover housing starts; Town of North Andover Master Plan Implementation (housing) on 2023 corridor deliveries."),
    ("Definitions:", "In-place rent = average rent on leased units. Asking rent = rent on units currently listed. Tradeout = change from the prior tenant's rent to the new lease on the same unit. "
     "Premium = property ÷ benchmark – 1. Submarkets follow RealAI Datamart boundaries."),
    ("Important information:", "This presentation is for discussion purposes only and does not constitute an offer to sell or a solicitation of an offer to buy any security. "
     "Market data are third-party estimates and may be revised; past performance of Hanover communities is not indicative of future results. "
     "Ownership entities shown in public records for Hanover Edgewood and Hanover Midtown should be confirmed against Hanover's own records."),
], size=11, gap=10)

prs.save("Hanover_300_Minuteman_Road_Pitch_Book.pptx")
print("saved")

Not included in this share

The Hanover pitch book is built: six slides in Hanover's branding, with the decisions from our earlier rounds applied.

#	Slide	Headline
1	Cover	300 Minuteman Road: 334-unit market-rate community, Andover, MA
2	Market demand	An affluent, highly educated submarket whose new arrivals out-earn those leaving by $45K
3	Rent comps	Six communities built in nine years, none since 2022; new product earns a premium on 2- and 3-bedrooms
4	Leasing momentum	Tight occupancy and new-lease pricing turning upward; new leases up 3.6% over prior rents since March
5	Sponsor track record	Hanover product leases 14–28 days faster than its submarket and earns a 10% rent premium in Midtown
6	Appendix	Sources, definitions and the offering disclaimer

Branding. The green (#115740) and near-black come straight from your logo file. The logo is full-color on content slides and white on the green cover.

What changed from our drafts

Slide 2 uses the Census figure for high earners. The table and third bullet say 44% of households earn over $200K, not the 66.8% at $150K+ from the Datamart. The footnote explains why .
Slide 3 lists all six comps, with the footnote explaining that four of them sit in the Datamart's "Lawrence" submarket .
Slide 4 leads with the +3.6% six-month tradeout and shows the latest +5.85% as a secondary figure .
Slide 5 keeps the occupancy gap and Edgewood's ZIP comparison in footnotes. The appendix disclaimer asks for the Edgewood and Midtown ownership entities to be confirmed.

Before it goes to investors

The deck has no rents or returns for 300 Minuteman Road itself. It needs a slide on the project's proposed rents and returns before it can support an equity raise.
Consider a firm-overview slide. Hanover's website cites 28,038 units built . I haven't added that because the site gives no date for the figure.
Give it a visual pass in PowerPoint. I checked the rendered PDF programmatically for text running off the page or overlapping, but I haven't looked at the slides themselves.

Not included in this share

Attached is an actual pitch book from Hanover, for a property called Sugarloaf. I would like you to recreate the presentation for 300 Minuteman in Hanover's style. You will need to go out to the web to source photos of the property and conduct further research on what makes the surrounding area appealing. Generate a map with areas of interest and any other visuals you can source from the web. Leave a placeholder for cost assumptions and operational assumptions, as you do not have this in-hand. Ask me any questions you have.

7:21 PM

Before I start the rebuild, a quick search turned up facts about the project that change the deck. The biggest: the earlier slides call this a market-rate community, and Hanover's public filings say it isn't.

About a quarter of the units are planned as affordable. Hanover is applying through the state's Local Initiative Program (a "friendly 40B"). It proposes reserving 84 of the 334 units (25%) for households earning up to 80% of area median income . That leaves about 250 market-rate units. The cover, the rent comps and the affordability point on slide 2 should be reframed around that.
Hanover already built in Andover. It developed The Point at Merrimack River in 2018 (248 units, also a 40B) . That property is already in our rent comp set, at 96.8% occupancy and $2.94/SF. It makes a better track-record story than Atlanta.
Site facts I can use:
- Vacant, 20.24 acres, bought for $2.5M from Spear Street Capital in May 2026 .
- About two acres reserved for future commercial use .
- Three buildings of four and five stories, with 515 parking spaces .
- The site was previously approved for a 224,500 SF pharmaceutical plant and sits next to the MBTA multifamily overlay district .
Entitlement is the live risk. The Select Board has taken no vote, and its second reading is tentatively planned for Oct. 26. Board members raised traffic, sewer capacity and archaeology on the former Shattuck Farm . In the Sugarloaf format this belongs in the Executive Summary.

I've copied the Sugarloaf book to the saved workspace (hanover/ref/sugarloaf.pdf, file 01a10d80-4a50-7570-abe3-93d9af7f1cfc) and studied its layout. The deck will follow it: white pages, navy serif headings, a navy title page, a map-led site aerial, financial and assumptions pages, points of interest, demographics, supply pipeline, rent comps with photos, sale comps and key contacts. A few questions before I build:

Affordability structure to present

LIP / friendly 40B: 250 market + 84 affordable at 80% AMI (as presented to the Select Board)

Project name for the cover and footers

Hanover Minuteman Park

Visuals to use

Web photos of the site, Minuteman Park and nearby attractions (credited)Renderings of similar Hanover projects from hanoverco.com, labeled "similar project" as in Sugarloaf

Sugarloaf sections to include beyond the ones we've built

Supply pipeline (map + delivery schedule)Asset sale comps1/3/5-mile radius demographics + target-income chartEntitlement status / timeline

Deliverable format

Editable PowerPoint, letter landscape like Sugarloaf (plus a PDF export)

Instructions

GOAL: Build the "Why Andover / Minuteman Park" and "Nearby Points of Interest" content for an institutional multifamily development investment book (Hanover Company style) for "Hanover Minuteman Park", a proposed 334-unit apartment community (250 market-rate + 84 affordable at 80% AMI under a Local Initiative Program / friendly 40B) on a vacant 20.24-acre site at 300 Minuteman Road, Andover, MA 01810, inside Minuteman Park (a ~1M SF office/lab park off River Road near I-93 and I-495). Investors need to understand what makes the location appealing: employers, highway/transit access, retail/amenities, recreation, schools, and Boston proximity.

TASKS

Geocode the subject (300 Minuteman Road, Andover MA) with google_places_search_text (bias lat 42.66, lon -71.20). Record lat/lon.
Identify 14–20 points of interest within ~10 miles that a renter or investor would care about, with lat/lon for each (use google_places_search_text) and drive distance/time estimate from the subject (state "approx." and how you estimated; straight-line × 1.3 is acceptable if labeled). Categories and suggestions to check (verify each, drop anything not real):
- Major employers / job nodes: Minuteman Park tenants (e.g., Alexandria lab campus, 150 Minuteman Rd lab), Pfizer Andover, Philips (Andover), RTX/Raytheon Andover, Schneider Electric (Andover), MKS Instruments, Vicor, Mercury Systems, Wilmington/Burlington Route 128 tech corridor, Lowell (UMass Lowell). Find employee counts where publicly sourced.
- Highways/transit: I-93 interchange (River Rd / Dascomb Rd), I-495 interchange, MBTA Commuter Rail stations (Andover, Ballardvale, North Wilmington), drive time to downtown Boston and Logan Airport.
- Retail/amenities: grocery (Market Basket, Whole Foods Andover, Wegmans Burlington?, Trader Joe's), downtown Andover (Main Street), The Loop (Methuen), Tewksbury Market Basket plaza, restaurants.
- Recreation/culture: Harold Parker State Forest, Merrimack River, Shawsheen River trails, Andover Country Club, Phillips Academy & Addison Gallery of American Art.
- Schools: Andover Public Schools rankings (use nearby_school_ratings at the subject for ELEMENTARY/MIDDLE/HIGH within 4 miles) and Phillips Academy.
Write a 3-paragraph "Why Minuteman Park / Why Andover" narrative (like an OM's "Why Sugarloaf?" page): economic base, access, quality-of-life. Every factual claim must carry a source URL. Include 4–6 headline stat callouts (e.g., "X employees within Y miles", "Andover High rated 10/10", "25 miles to Boston").
Find any info on the Minuteman Park master-plan vision (housing + grocery-anchored retail + roads), and the MBTA Communities multifamily overlay district adjacency. Source URLs: andovermanews.com, andoverledger.com, eagletribune.com, bisnow.com, andoverma.gov.
Download 6–8 high-quality photos of POIs (e.g., downtown Andover Main Street, Phillips Academy/Addison Gallery, Harold Parker State Forest, Merrimack River, Andover commuter rail, Minuteman Park/River Rd office-lab buildings). STRONGLY prefer Wikimedia Commons (freely licensed) — fetch the file page to record license + author, then download the original or a ≥1600px version with curl (use a browser User-Agent header). Save as JPG under subagents/area/photos/ with descriptive filenames. Verify each file with identify (must be a real image, ≥1200px wide). Record credits.

RULES: Use web_search/web_fetch for facts; never invent numbers, employee counts, or distances — if unverifiable, say so. Flag conflicts between sources. Don't use paywalled content you can't read.

OUTPUT (write to subagents/area/):

poi.json: list of {name, category, lat, lon, distance_mi, drive_min_est, distance_method, one_line_description, source_url}. Include the subject as the first entry with category "subject".
narrative.md: the 3-paragraph narrative with inline source URLs, plus the stat callouts each with source URL.
photos/ folder + photo_credits.json: {filename, subject, source_page_url, author, license, width, height}.
schools.json from the school ratings call. Return a compact report: subject lat/lon, the POI list (name, category, distance), the stat callouts with sources, photo filenames with licenses, and any gaps.

Results

All files are in subagents/area/: poi.json, narrative.md, schools.json, photo_credits.json, photos/ (7 JPGs). Scripts: build_poi.py (distances) and fetch_photos.py (photo download).

Subject. 300 Minuteman Rd, Andover MA 01810 geocodes to lat 42.69317, lon -71.20855 (Google Places ID ChIJN6N0CxGo44kRm-1IqiHNW-I).

POI list (22 plus the subject).

Distance is approx. drive miles: haversine straight-line × 1.3. Drive minutes assume 25, 35 or 45 mph depending on distance, with no traffic. Both are labeled in poi.json.
Employers:
- Philips, 3000 Minuteman Rd, 0.7 mi
- Mercury Systems, 0.7 mi
- Schneider Electric R&D, 1.5 mi
- Raytheon, 350 Lowell St, 4.7 mi
- Vicor, 6.1 mi
- Pfizer, Burtt Rd, 7.5 mi
- UMass Lowell, 8.4 mi
Transit:
- Andover MBTA station, 5.3 mi
- Ballardvale MBTA station, 6.7 mi
- North Station, 30.9 mi (~41 min, modeled)
- Logan Airport, 32.2 mi (~43 min, modeled)
Retail:
- Market Basket, Methuen, 1.2 mi
- Market Basket, North Andover, 4.8 mi
- Whole Foods, Andover, 5.1 mi
- The Loop, Methuen, 5.6 mi
- Downtown Main Street, 5.7 mi
- Burlington Mall, 18.9 mi
Recreation and culture:
- Merrimack River Trail, 0.4 mi
- Andover Country Club, 3.3 mi
- Harold Parker State Forest, 10.4 mi
- Phillips Academy and Addison Gallery, 6.4 mi each
Schools: Andover High School, 4.8 mi.

Stat callouts (sources; details in narrative.md).

Raytheon has 10,000 Andover employees (2024), about a third of private-sector jobs: https://andovermanews.com/andover-by-the-numbers-employers/. Another article estimates 10,450 from state data: https://andovermanews.com/andover-by-the-numbers-top-taxpayers/.
Minuteman Park is 25 miles north of Boston, directly off I-93, five miles south of the NH border: Newmark, https://www.nmrk.com/insights/press-releases/newmark-completes-341-million-sale-of-eight-assets-within-minuteman-park-in-andover-massachusetts.
Andover High ranks #38 in MA and #765 nationally, with a 97% graduation rate: U.S. News, https://www.usnews.com/education/best-high-schools/massachusetts/districts/andover/andover-high-school-9263. RealAI's school tool rates it 10/10.
Minuteman Park is about 1M SF per Bisnow (https://www.bisnow.com/news/boston/deal-sheet/houston-developer-acquires-andover-site-for-housing-development-boston-deal-sheet-134732). Andover News describes a 127-acre park (top-taxpayers URL above). Eight assets traded for $341M in 2022 (Newmark URL above).
Andover has 2 MBTA Commuter Rail stations (https://storymaps.arcgis.com/stories/a0735bc4131446b897892ed8ec50d62c). A sample inbound train runs Andover 4:37 to North Station in about 52 min (https://transitapp.com/en/region/boston/mbta/commuter-rail-haverhill).
Hanover estimates $1.5M/yr in property tax versus about $150K today: https://andoverledger.com/articles/andover-weighs-334-unit-apartment-complex-at-minuteman-park-mu4dwlit.

Master-plan and overlay findings.

Hanover showed the Select Board a preliminary vision for Minuteman Park with more housing, new roads and a grocery-anchored retail area. None of it is before the board for approval (https://andovermanews.com/andover-select-board-reviews-334-unit-minuteman-park-proposal).
The project includes about 2 acres reserved for future retail. It is 3 buildings of 4–5 stories: 201 one-bed, 110 two-bed and 23 three-bed units, with 515 parking spaces (Eagle-Tribune, https://www.eagletribune.com/news/merrimack_valley/334-unit-housing-complex-proposed-at-minuteman-park/article_dc99915b-55a1-467d-ad1c-1b2dd1d60baa.html).
The Eagle-Tribune says the parcel is "adjacent to the MBTA Multifamily Overlay District." Andover voters created that district on April 30, 2024 (search snippet of https://andoverma.gov/1069/Multifamily-Overlay-District).
The LIP path: the Select Board took no vote and a second reading is targeted for Oct 26. Board members raised traffic, sewer capacity and archaeology concerns. The parcel was previously approved in 2023 for a 224,500 SF cGMP pharmaceutical manufacturing facility, which did not advance.
Hanover bought the site for $2.5M in May 2026. The seller, Spear Street, had paid $9.5M in 2015 (Bisnow).
Hanover built the 248-unit Point at Merrimack River in Andover in 2018.

Photos (all verified with identify as JPEG, at least 2448 px wide; credits in photo_credits.json).

File	Subject	Author	License
addison_gallery_phillips_academy_andover.jpg	Addison Gallery	Daderot	Public domain
phillips_academy_memorial_bell_tower.jpg	Phillips Academy bell tower	Daderot	Public domain
harold_parker_state_forest_berry_pond.jpg	Harold Parker SF, Berry Pond	John Phelan	CC BY-SA 4.0
harold_parker_state_forest_stearns_pond.jpg	Harold Parker SF, Stearns Pond	John Phelan	CC BY-SA 3.0
andover_mbta_commuter_rail_station.jpg	Andover MBTA station	John Phelan	CC BY-SA 3.0
downtown_andover_main_street.jpg	Downtown Main Street	John Phelan	CC BY 3.0
i495_bridge_merrimack_river_lawrence.jpg	I-495 bridge over the Merrimack, Lawrence	John Phelan	CC BY 3.0

Gaps and conflicts.

Minuteman Park photo: none found on Commons (Bisnow has a Google Maps-sourced photo, not freely licensed). The Merrimack photo is the I-495 bridge in Lawrence, not an Andover riverfront shot.
Schools:
- The subject-centered 4-mile call returned only 8 preschools for ELEMENTARY, and the same 10 nearest schools for MIDDLE and HIGH. It never reached Andover High or Wood Hill.
- For those schools I used a supplemental call centered on Andover High, stored in schools.json under ahs_supplement: Andover HS 10, Andover West Middle 8, West Elementary 8.
- From the subject-centered call: High Plain Elementary 9/10 and Wood Hill Middle 8/10, both about 2 mi away. Greater Lawrence Tech is 4/10.
- Neighboring Lawrence and Methuen schools score 2–5 or are unrated.
Interchanges: I-93/I-495 interchange points and the River Rd/Dascomb Rd exit did not geocode, so they are not in poi.json.
Unverified employers and tenants: MKS and the Alexandria/150 Minuteman Rd tenant list were not verified. Wegmans and Trader Joe's were not checked. MKS has only a 100–249 band, no count.
Employee counts:
- Pfizer: band of 1,000–4,999 (bond filing) versus about 1,500 in the taxpayer article.
- Mercury: band of 500–999.
- Vicor and Philips: no counts.
Distances:
- The Boston figure conflicts: Newmark says 25 mi, my ×1.3 estimate is about 31 mi. Use the sourced 25 mi.
- Logan's drive time is modeled only.
- The ~52-min train time comes from one sample train on the Transit app. A real-estate blog says 50–62 min.
Overlay boundary: The andoverma.gov overlay page body did not render, so the district boundary was not verified.
Not read: The andoverma.gov April 14 Select Board brief did not render either, and I did not read the Bisnow or Ledger articles in full. Statements from them come from the search text.

Instructions

GOAL: Supply the "Supply Pipeline", "Asset Sale Comps", and affordable-rent inputs for an institutional investment book for "Hanover Minuteman Park", a proposed 334-unit apartment community (250 market-rate + 84 units affordable at 80% AMI via Massachusetts Local Initiative Program / friendly 40B) at 300 Minuteman Road, Andover, MA 01810 (approx. lat 42.66, lon -71.20; near I-93/I-495). Investors use these pages to judge competing new supply and exit value.

TASK A — SUPPLY PIPELINE (within ~5–7 miles: Andover, North Andover, Tewksbury, Wilmington, Methuen, Lawrence, Lowell east side, North Reading, Dracut): find multifamily rental projects of ~50+ units that are under construction, approved/permitted, or proposed (2024–2026 news), plus anything delivered 2024–2026 still in lease-up. For each: name, address, developer, units, status (UC / approved / proposed / lease-up), expected delivery date, affordability (40B? % affordable), lat/lon (use google_places_search_text or geocode the address), and source URL. Search local outlets (andovermanews.com, andoverledger.com, eagletribune.com, lowellsun.com, bisnow.com/boston, bldup.com, bizjournals) and town planning/ZBA pages. Note MBTA Communities zoning overlays adopted in Andover/North Andover/Tewksbury/Wilmington and any known projects under them. Known context: Andover 2023 housing starts were 9 (andovermanews); Avalon North Andover (221u, 2022) and 505 Sutton (136u, 2022) in North Andover are recent deliveries. Also check the Datamart: use explore_data then query_data on entity property_mfr for properties in submarket_id IN ["0eb2cbd1c19f4cd0024a0390441e62b9","7bbccbf1e9cfa562703d937cbbe215ab"] with mf_property_attributes.year_built >= 2024 to catch lease-ups.

TASK B — ASSET SALE COMPS: find 4–6 multifamily sales (2022–2026) of 100+ unit properties built 2000 or later in the Merrimack Valley / I-93 / I-495 north suburbs (Andover, North Andover, Tewksbury, Wilmington, Burlington, Woburn, Lowell, Methuen, Haverhill, Chelmsford, Billerica, Peabody, Danvers). For each: property, address, year built, units, sale date, price, price/unit, cap rate (if reported), buyer/seller, source URL. Sources: Datamart (explore_data on property_mfr topics, then mf_sales_history and mf_property_attributes for candidate properties — e.g., query property_mfr with address_city IN [list] and mf_property_attributes.unit_count >= 100, mf_property_attributes.year_built >= 2000, then check mf_sales_history), plus web (Bisnow Boston Deal Sheet, Boston Business Journal, The Real Deal, MHN, press releases). Never invent a cap rate; leave null if not reported.

TASK C — AFFORDABLE RENT LIMITS: Andover MA falls in HUD's Lawrence, MA-NH HUD Metro FMR Area (verify). Find FY2025 (and FY2026 if published) HUD 80% AMI income limits by household size and the corresponding maximum affordable rents used for LIP/40B units in Massachusetts (EOHLC/DHCD guidance: rent ≤30% of 80% AMI income, household size = bedrooms + 1 typically; utility allowance deducted). Compute max gross rent for studio/1BR/2BR/3BR in a saved python script (subagents/pipeline/affordable_rents.py) and show the inputs. Source URLs required (huduser.gov income limits page, mass.gov LIP/40B guidelines).

RULES: Every figure needs a source URL or Datamart tool reference (state the entity id and field). Flag conflicts. Do not fabricate. Save work in subagents/pipeline/.

OUTPUT FILES (subagents/pipeline/): pipeline.json, sale_comps.json, affordable_rents.json (+ the .py script), notes.md with sources and caveats. RETURN: compact report listing pipeline projects (name, units, status, delivery, distance from subject), sale comps (property, date, price, $/unit, cap rate), affordable max rents by bedroom with the income limits used, and gaps/confidence.

Results

Subject: 300 Minuteman Rd geocodes to 42.69317, -71.20855 (Google Places); the brief's 42.66 was approximate. Status: Select Board first read 9/15/2026, second ~10/26. The plan is 3 buildings of 4–5 stories, 84 units (25%) at 80% AMI, via LIP ("friendly 40B"). Sources: andovermanews.com/andover-select-board-reviews-334-unit-minuteman-park-proposal; Bisnow deal sheet 134732 (site bought for $2.5M from Spear Street).

A. Supply pipeline (straight-line miles from subject; pipeline.json)

Zero Prescott St (East Mill), North Andover: 296 units, East Mill Housing LLC. Proposed 40B; ZBA hearing continued May 2026, draft decision dated 4/29/26. ~4.0 mi. The delivery date is not reported. The lat/lon is a proxy (downtown North Andover), not the parcel. Sources: patch.com/massachusetts/northandover/massive-housing-project-boards-agenda-north-andover; records.northandoverma.gov DocView id=487201.
946 Osgood St (China Blossom), North Andover: 108 units. Proposed LIP/friendly 40B, under town evaluation. ~4.7 mi. The developer is not identified. The unit count comes from a search snippet only, because the town page returned "Not Found".
Princeton Wilmington, 201–203 Jefferson Rd: 108 units, National Development / Princeton Properties. Broke ground Jul 2025; leasing Aug 2026, so it is in lease-up now. ~8.8 mi. The affordable share is unverified. Sources: natdev.com/project/wilmingtonapartments/; princetonproperties.com/coming-soon/.
MacLellan Oil site, 1167 Main St, Tewksbury: 24 units. Approved early 2024, site clearing in 2026. ~6.0 mi. It is below the 50-unit threshold but noted. Source: homenewshere.com Tewksbury Town Crier "Projects in progress".
Recent deliveries (Datamart property_mfr):
- Avalon North Andover: 221 units, 2022, id 02040dd87d588479c7e26dc25d280b2b, ~4.2 mi.
- 505 Sutton: 136 units, 2022, id e5161a7f949c7d6d4f3068750608034c, ~4.7 mi.
Datamart lease-up check: no property with year_built ≥ 2024 in submarkets 0eb2cbd1… and 7bbccbf1…. Newest are the 2022 deliveries above and Point at Merrimack Valley (2021).
Not competitive or out of radius: Dracut Avenue at Greenmont (56 senior affordable units, opened 6/11/2026); Lowell North Canal (160 affordable units, approved 6/2026); Tewksbury 1325 Main (14 units).
MBTA Communities:
- Andover is compliant (overlays at the Andover station, Ballardvale and River Road).
- North Andover's Town Meeting approved in May 2024 (535–253), with sub-districts at the North Andover Mall and Osgood Landing.
- Tewksbury's Town Meeting rejected in May 2024; its later status is unverified.
- Wilmington is not verified.
- No projects confirmed under the overlays.
- Sources: tbf.org 2025 Housing Report Card; mbtacommunities.bostonindicators.org.

B. Sale comps (Datamart property_mfr mf_sales_history / latest sale fields; sale_comps.json)

Property	Sale date	Price	$/unit	Cap rate	Buyer / seller
Residences at Crosspoint, Lowell (240u, 2020, id f62b35f22dd198f4644e2a4bb2730077)	11/27/2024	$85.075M	$354.5k	5.0% (CoStar via Wronka snippet)	Seller SMC Management, CBRE listing broker; buyer not retrieved
Villas at Old Concord, Billerica (324u, 2004)	9/27/2024	$114.5M	$353.4k	null	Buyer TruAmerica (yieldpro.com/2024/09/villas-at-old-concord/)
Point at Merrimack Valley, Methuen (156u, 2021)	3/28/2024	$58.1M	$372.4k	null	not retrieved
Woods at Merrimack, Methuen (140u, 2019)	11/3/2025	$37.537M	$268.1k	null	not retrieved
Legacy Park, Lawrence (104u, 2009)	6/25/2026	$30.2M	$290.4k	null	not retrieved
Point at Woburn (289u, 2022)	Jan 2026	$130.676M	$452.2k	null	Buyer Pantzer Properties; seller Toll Brothers / Carlyle affiliate (MarketScreener)

Conflicts: Traded.co shows $130M / $449.8k for Point at Woburn against Datamart's $130.676M. Woods at Merrimack at $268k/unit looks anomalous and needs verification.
Other Datamart candidates not used:
- Stonebridge at Burlington: 110u, 7/2025, $468.5k/unit.
- Lynnfield Commons: 200u, 6/2025, $407.5k/unit.
- The Val, Billerica: 211u, 9/2023, $438.6k/unit.
- Bradlee Danvers: 433u, 6/2022, $478.9k/unit.
Market cap rate: Wronka's Aug 2025 report shows a Boston market cap rate of about 5.1%.

C. Affordable rents (Lawrence, MA-NH HMFA; affordable_rents.py / .json)

FY2025 80% limits, 1–8 persons: 72,950 / 83,400 / 93,800 / 104,200 / 112,550 / 120,900 / 129,250 / 137,550. Area median $141,300. Sources: huduser.gov/portal/datasets/home-datasets/files/HOME_IncomeLmts_State_MA_2025.pdf; masshousing.com 2025-WFH-FMR-Lawrence.pdf.
FY2026 80% limits, effective 5/1/2026: 74,800 / 85,450 / 96,150 / 106,800 / 115,350 / 123,900 / 132,450 / 141,000. Area median $137,300 (-2.83%). Source: masshousing.com/wp-content/uploads/2026-WFH-FMR-LAWRENCE.pdf.
Method: gross rent = 30% × limit / 12, before any utility allowance. Household size is 1.5 persons per bedroom, with studio = 1 person.

Unit type	FY2026 max gross rent	FY2025 max gross rent
Studio	$1,870	$1,824
1BR	$2,003	$1,954
2BR	$2,404	$2,345
3BR	$2,777	$2,709

Bedrooms+1 sensitivity (FY2026): $1,870 / $2,136 / $2,404 / $2,670.
Flag: the FY2026 median fell but the HUD-published 80% limit rose, due to HUD's cap rule. I used the HUD-published 80% limit, not MassHousing's uncapped figure (109,850 for 4 persons).

Gaps:

The utility allowance is not deducted; it is project-specific and I did not find a value.
I did not open the official mass.gov EOHLC LIP guideline page. The 1.5 persons-per-bedroom convention comes from search snippets, so it is an assumption to confirm.
I did not open HUD's page confirming Andover is in the Lawrence HMFA. I relied on the brief and the MassHousing Lawrence sheet.
Market-rate 50+ unit projects in Methuen, Lawrence, Dracut, North Reading and Lowell-east were not surfaced by search. Treat that as a gap, not as "none".
Datamart gives only the latest sale, so buyer, seller and cap rate are mostly null.
Files are in subagents/pipeline/: pipeline.json, sale_comps.json, affordable_rents.json, affordable_rents.py, build_pipeline.py, notes.md. Raw Datamart pulls are in tool-outputs/new_mf.json and tool-outputs/comps_dm.json.

Instructions

GOAL: Gather visuals and entitlement facts for an investment book for "Hanover Minuteman Park", a proposed 334-unit apartment community by The Hanover Company at 300 Minuteman Road, Andover, MA 01810 (vacant 20.24-acre site in Minuteman Park, off River Road near I-93).

TASK 1 — HANOVER "SIMILAR PROJECT" RENDERINGS/PHOTOS: From hanoverco.com (portfolio pages, e.g., https://www.hanoverco.com/portfolio/ , property pages such as Hanover Weymouth (MA), Hanover Colony Place (Plymouth MA), Hanover Midtown, and any other suburban 4–5 story wood-frame communities), download 6–10 high-resolution exterior and amenity images (pool/courtyard, clubhouse, fitness, unit interior, exterior elevations). These will be labeled "Rendering/photo of similar Hanover project; subject project will vary." Use web_fetch to read pages and find image URLs (look for wp-content/uploads .jpg/.webp; strip size suffixes like -1024x683 to get originals), then curl them with a browser User-Agent. Convert webp to jpg with ImageMagick. Keep only images ≥1400px wide. Record the source page and project name for each.

TASK 2 — THE POINT AT MERRIMACK RIVER (30 Shattuck Rd, Andover; 248 units, built 2018 by Hanover, Chapter 40B): download 2–4 photos (property website or hanoverco.com) and gather facts: developer/owner history (did Hanover sell it? when, to whom, price?), affordability share, amenities, website. Sources required.

TASK 3 — RENT COMP PHOTOS: one good exterior photo each (property websites preferred) for: The Slate at Andover (50 Woodview Way, Andover), Berry Farms (4 Berry St, North Andover), Haven North Andover (1252 Osgood St, North Andover), Avalon North Andover (88 High St, North Andover), 505 Sutton (495 Sutton St, North Andover). Plus The Point (task 2). Record source URL.

TASK 4 — SITE / MINUTEMAN PARK IMAGES: try to find usable images of the 300 Minuteman Road site or Minuteman Park entrance/buildings (Bisnow article https://www.bisnow.com/news/boston/deal-sheet/houston-developer-acquires-andover-site-for-housing-development-boston-deal-sheet-134732 , Andover Ledger https://andoverledger.com/articles/andover-weighs-334-unit-apartment-complex-at-minuteman-park-mu4dwlit — its lead image is "Courtesy: Hanover Co." and may be a rendering or site plan; Alexandria/BGO/Optimum marketing pages for Minuteman Park lab buildings). Download and note credit lines. Also, if possible, generate an aerial image of the site: download Esri World Imagery tiles (https://server.arcgisonline.com/ArcGIS/rest/services/World_Imagery/MapServer/tile/{z}/{y}/{x}) around the site at zoom 16 and 17, stitch with Pillow into a ~2400x1600 image centered on the site (geocode the site first with google_places_search_text), and save as site_aerial_z16.jpg / site_aerial_z17.jpg with a metadata json giving the center lat/lon, zoom, and the pixel bounds→lat/lon transform (web mercator) so another script can overlay markers. Attribution: "Esri, Maxar, Earthstar Geographics".

TASK 5 — ENTITLEMENT STATUS: Build a dated timeline from public sources: 2021–2023 cGMP pharma special permit history for 300 Minuteman Rd; Hanover land purchase (May 2026, $2.5M from Spear Street Capital); Select Board LIP first readings (Apr 14, 2026 overview; Sep 15, 2026 presentation); Conservation Commission action (the Andover Ledger article says the Conservation Commission "unanimously denied a related..." — find out what exactly); scheduled second reading (~Oct 26, 2026); next steps (EOHLC project eligibility, ZBA comprehensive permit). Also capture issues raised (traffic, sewer capacity, archaeology at former Shattuck Farm, conversion of commercial land) and Hanover's commitments (sewer work, ~2 acres reserved for commercial, $1.5M/yr property tax estimate vs ~$150K today, 33–50 school children estimate). Fetch andoverledger.com, andovermanews.com, eagletribune.com, andoverma.gov meeting briefs/agendas. Every item needs a source URL and date.

RULES: Do not fabricate facts or credits. Verify every downloaded file with identify (real image, note pixel size). Save everything under subagents/images/ (subfolders: hanover_similar/, comps/, site/). Write images.json listing {filename, category, subject, source_url, credit, width, height}. Write entitlement.json (timeline items with date, event, source_url) and the_point.json. RETURN: compact report of all saved images (filename, subject, size, source), The Point facts, the entitlement timeline, and gaps.

Results

All files are under subagents/images/. Every image was checked with identify. Several items are gaps, listed at the end.

IMAGES All 12 Hanover images are 2400–2560 px wide. The Hanover photos are of completed projects, not renderings. Suggested label: "Rendering/photo of similar Hanover project; subject project will vary." Credit for all: The Hanover Company (hanoverco.com). I viewed them all and the subjects match the names below.

Hanover similar projects (hanover_similar/):

colony_place_pool_dusk.jpg, 2400x1600. Hanover Colony Place, Plymouth MA. Pool and courtyard at dusk. Source: hanoverco.com/property/hanover-colony-place/
colony_place_aerial.jpg, 2400x1350. Colony Place aerial.
colony_place_firepit_courtyard.jpg, 2400x1600. Colony Place courtyard with fire table.
colony_place_entrance.jpg, 2400x1600. Colony Place building entrance elevation.
colony_place_fitness.jpg, 2400x1600. Colony Place fitness center.
colony_place_clubhouse.jpg, 2400x1600. Colony Place clubhouse lounge.
weymouth_pool.jpg, 2400x1600. Hanover Weymouth, Weymouth MA. Pool. Source: hanoverco.com/property/hanover-weymouth/
weymouth_courtyard_dusk.jpg, 2400x1600. Weymouth courtyard with fountains.
weymouth_fitness.jpg, 2400x1600. Weymouth fitness center.
stoneham_exterior.jpg, 2560x1707. Hanover Stoneham, Stoneham MA. 5-story exterior. Source: hanoverco.com/property/hanover-stoneham/
stoneham_courtyard_dusk.jpg, 2560x1707. Stoneham courtyard at dusk.
stoneham_courtyard.jpg, 2560x1920. Stoneham courtyard.
Hanover Midtown images were downloaded (2400 px) but not kept. Midtown is a higher-density project, less comparable.
I did not find a separate unit-interior shot in the Hanover set. The clubhouse, fitness and lounge shots are the interiors.

Comps (comps/):

berry_farms_exterior.jpg, 1920x1280. Aerial of buildings and pool. Berry Farms property site berryfarmsapartments.com. Image URL: https://irp.cdn-website.com/aae78eeb/dms3rep/multi/31-web-or-mls-Berry+St-S0907-029.jpg
haven_north_andover_exterior.jpg, 1640x900. Building exterior at dusk. Source: havennorthandover.com/gallery
haven_north_andover_clubhouse_exterior.jpg, 1640x900. Clubhouse exterior at dusk. Source: havennorthandover.com/gallery
505_sutton_entrance.jpg, 1920x1280. Entrance monument with building. Source: 505sutton.com/gallery, image "505+Sutton-POI-10.jpg"
505_sutton_patio.jpg, 1920x1280. Patio with building exterior. Source: 505sutton.com/gallery, image "DSC_8382.jpg"
the_point_exterior_housingnavigator.jpg, 1920x1440. The Point exterior. https://search.housingnavigatorma.org/file/image/133303/lightbox, from listing HN-MA-009978. It is the listing's og:image. It is a property-supplied listing photo, but the credit is not explicit.
the_point_exterior_redfin.jpg, 1280x960. The Point exterior. Redfin lead photo: https://ssl.cdn-redfin.com/photo/rent/dc56bd6f-560f-41c1-a704-0196d0ca87ad/bigphoto/0_8.jpg. Licensing is not confirmed; it is a third-party listing feed.

Site (site/):

site_aerial_z16.jpg and site_aerial_z17.jpg, both 2400x1600. Esri World Imagery tiles stitched by subagents/images/make_aerial.py (rerunnable). Attribution: "Esri, Maxar, Earthstar Geographics". I viewed z17: the center is wooded, vacant land just west of I-93.
site_aerial_meta.json holds the center, zoom, world-pixel origin, and the top-left and bottom-right lat/lon for each image. The transform is Web Mercator: world_px = 256·2^z·((lon+180)/360, (1−ln(tan φ+sec φ)/π)/2), and image_px = world_px − origin. This is how another script can place overlay markers.
Geocode: 42.6931748, -71.2085521, from google_places_search_text ("300 Minuteman Rd", place id ChIJN6N0CxGo44kRm-1IqiHNW-I).
minuteman_park_entrance_sign.jpg, 1237x842. Lower resolution. Taken from the Bisnow article (og:image URL in images.json). Bisnow's caption reads "The Entrance To Andover's Minuteman Park Complex...". It looks like a Google Street View capture. There is no explicit photo credit.
300_minuteman_rd_building_photo.jpg, 984x450. Andover News lead image. It is an aerial of a blue-glass building beside a highway. It appears to be an existing Minuteman Park building, not the vacant site, and is low-res. It has no credit line.

images.json lists filename, category, subject, source_url, credit, width and height for all 23 images.

THE POINT AT MERRIMACK RIVER (the_point.json)

Address and size: 248 units in two buildings at 30 and 40 Shattuck Rd. Source: https://andoverma.gov/824/The-Point-at-Merrimack
Developer and build: built in 2018 by Hanover Co. as a Chapter 40B project. Eagle-Tribune and Andover Ledger both say so.
Datamart record (property_mfr id 9b7a968a7992ab189999245c08f4bc00):
- Fee owner is "ANDOVER APARTMENTS LLC".
- Latest sale date is 2017-04-21 with no price.
- 10.24 acres, 4 stories, 2 buildings.
- Rent type is MARKET_AND_AFFORDABLE.
- The datamart shows no true owner and no manager.
Hanover sale: I found NO source saying Hanover sold it. There is no buyer, date or price. Treat the ownership history as unverified.
Website: https://www.thepointatmerrimackriver.com/. Management is Pantzer on the property site; Housing Navigator says Panco Management.
Affordable units exist, but I found no source for a 25% share.
- SEB Housing lists 80% AMI rents of 1BR $1,791 and 2BR $2,118. Source: https://sebhousing.com/affordable-housing-opportunities/
- Housing Navigator lists fixed below-market 1/2/3BR units. Source: https://search.housingnavigatorma.org/navigator/listing/HN-MA-009978
Amenities (Zillow and Housing Navigator):
- 6,000 SF clubhouse
- 24-hour fitness center
- Resort-style pool
- Theater room with stadium seating
- Computer lounge
- Business center
- BBQ areas
- Outdoor fireplace
- Playground
- Garage parking at $250/month
Asking rents (Zillow, Oct 2026): 1BR 696 SF from $2,593; 2BR 1,140–1,227 SF $3,277–$3,552.

ENTITLEMENT TIMELINE (entitlement.json; every item has a source URL)

2015: Spear Street bought the site for $9.5M. Source: Bisnow, https://www.bisnow.com/news/boston/deal-sheet/houston-developer-acquires-andover-site-for-housing-development-boston-deal-sheet-134732
2021, before 9/14: the ZBA granted a parking variance. The exact date was not found.
2021-09-14: Planning Board opened the hearing on special permit SP21-06 for a 224,500 GSF cGMP facility. Source: https://www.andoverma.us/AgendaCenter/ViewFile/Minutes/_09142021-4351
- The infrastructure was built for a 320,000 SF office building.
- Traffic would be about 30% lower than office.
- Past construction created isolated wetlands, to be filled and replaced through a Notice of Intent to the Conservation Commission.
- Sewer and River Rd/Minuteman/Shattuck traffic mitigation was proposed.
2023: the cGMP project was approved, per the Eagle-Tribune. The Conservation Commission minutes mention a 2023 wetland delineation and an Order of Conditions that took over a year. The Conservation Commission attached a Conservation Restriction (CR) as mitigation. The exact Planning Board decision date was not found.
2026-03-17: Conservation Commission asked for more detail on the CR-timing request.
2026-04-14: Select Board overview. Hall gave the $1.5M vs ~$150K tax figures (per the Ledger). Bisnow says the town called the project the multifamily piece of a "broader, multi-phased redevelopment". The town brief page text did not load. Source: https://andoverma.gov/1725/17710/April-14-2026
2026-04-21: Conservation Commission UNANIMOUSLY (5-0) denied an Amendment to the Order of Conditions. Source: minutes, https://andoverma.gov/AgendaCenter/ViewFile/Minutes/_04212026-7393
- The amendment would have changed only the TIMING of recording the CR. It was procedural and did not deny the housing project itself.
- Attorney Mark Johnson said the sale must close before new permitting, and Hanover wants multifamily next to the MBTA overlay.
- The DEP Circuit Rider advised filing a new Notice of Intent instead. The 2023 delineation is still valid, and the applicant could choose between two valid Orders of Conditions.
- Commissioners noted that a 40B/LIP comprehensive permit would bypass the commission's review.
May 2026: Hanover bought for $2.5M from Spear Street Capital. Sources: Bisnow and the Boston Business Journal (5/27/26), https://www.bizjournals.com/boston/news/2026/05/27/texas-developer-buys-site-eyed-for-334-apartments.html. The deed date was not found.
2026-09-15: Select Board first-reading presentation of the Local Initiative Program (LIP) "friendly 40B". No vote was taken. Sources: Andover Ledger, https://andoverledger.com/articles/andover-weighs-334-unit-apartment-complex-at-minuteman-park-mu4dwlit; Andover News and Eagle-Tribune.
- Program: 334 units in 3 buildings (two 5-story, one 4-story).
- Parking: 515 spaces.
- Unit mix: 201 1BR, 110 2BR, 23 3BR.
- Affordable: 84 units (25%) at up to 80% AMI.
- Commercial: about 2 acres reserved.
- DATE CONFLICT: the Ledger says Tuesday 9/15; Andover News and the Eagle-Tribune say "Monday". Check the Select Board minutes.
2026-10-26 (expected): second reading.
Next steps: if the board supports it, the LIP application goes to the state (EOHLC) for project eligibility. Then the ZBA holds a comprehensive permit hearing; no date is set.
A new Notice of Intent to the Conservation Commission is also possible.

Issues raised:

Traffic: Vice Chair Melissa Danisch said "Traffic already stinks."
Sewer: the Department of Public Works wants an updated sewer capacity analysis and detail on repairs Hanover would fund.
Archaeology: Danisch asked for a review of the former Shattuck Farm. The Eagle-Tribune says it is a documented Native American site with multiple prior studies.
Land conversion: Member Kevin Coffey raised the loss of commercial and industrial land to housing. He also noted that 100 River Rd housing was recently approved.
Schools: Hanover estimates 33–50 school-age children. Town Counsel Doug Heim said enrollment is generally not a ZBA factor.

Hanover's commitments and claims:

Complete the sewer work previously required for the abandoned manufacturing project.
Reserve about 2 acres for commercial use.
$1.5M/yr property tax vs about $150K today.
Andover is already above 10% on the state's Subsidized Housing Inventory, so the project is voluntary (Hall, to Bisnow).
Amenities include a fitness center and coworking space.
A park-wide vision with housing, roads and grocery-anchored retail. This is not before the board.

GAPS

The Slate at Andover and Avalon North Andover: no exterior photo. Their websites block scripted download (HTTP 403). The Redfin lead photos I could fetch were interiors, so I did not keep them. These two need a manual download. Likely URLs: theslateatandoverapartments.com and avaloncommunities.com/massachusetts/north-andover-apartments/avalon-north-andover/.
The Point's own website also blocks download (403), so the two Point photos came from third-party listings.
No rendering or site plan of the subject project was found. The Ledger's "Courtesy: Hanover Co." lead image is just the Hanover logo (318x106), not a rendering.
Alexandria, BGO and Optimum marketing pages for Minuteman Park were not fetched.
The exact dates of the Planning Board, ZBA and Town Meeting actions on the 2021–2023 cGMP permit were not found.
The Hanover purchase closing date was not found.
The Point: no sale by Hanover was found, and the 25% affordable share is not source-confirmed.
Not read: the hanoverco.com text for Colony Place, Weymouth and Stoneham. Images only.
Colony Place image size suffixes were stripped; the files are full-size.

Content

171 lines

"""Aerial maps for the Hanover Minuteman Park investment book.
Esri World Imagery tiles (attribution: Esri, Maxar, Earthstar Geographics). Coordinates: Google Places / RealAI Datamart."""
import math, os, io, requests
from PIL import Image, ImageDraw, ImageFont, ImageFilter

OUT = os.path.join(os.path.dirname(__file__), "..", "assets", "maps")
CACHE = os.path.join(os.path.dirname(__file__), "..", "assets", "tilecache")
os.makedirs(OUT, exist_ok=True); os.makedirs(CACHE, exist_ok=True)
NAVY = (10, 34, 64); ORANGE = (240, 81, 36); WHITE = (255, 255, 255)
FB = "/usr/share/fonts/urw-base35/URWGothic-Demi.otf"
FR = "/usr/share/fonts/urw-base35/URWGothic-Book.otf"
FS = "/usr/share/fonts/urw-base35/C059-Roman.otf"
URL = "https://server.arcgisonline.com/ArcGIS/rest/services/World_Imagery/MapServer/tile/{z}/{y}/{x}"
UA = {"User-Agent": "Mozilla/5.0 (RealAI investment-book map builder)"}


def wpx(lat, lon, z):
    s = 256 * 2 ** z
    x = (lon + 180) / 360 * s
    lr = math.radians(lat)
    y = (1 - math.log(math.tan(lr) + 1 / math.cos(lr)) / math.pi) / 2 * s
    return x, y


def tile(z, x, y):
    p = os.path.join(CACHE, f"{z}_{x}_{y}.jpg")
    if not os.path.exists(p):
        r = requests.get(URL.format(z=z, x=x, y=y), headers=UA, timeout=30)
        r.raise_for_status()
        open(p, "wb").write(r.content)
    return Image.open(p).convert("RGB")


def basemap(bbox, z):
    (lat0, lon0, lat1, lon1) = bbox  # south, west, north, east
    x0, y0 = wpx(lat1, lon0, z); x1, y1 = wpx(lat0, lon1, z)
    tx0, ty0, tx1, ty1 = int(x0 // 256), int(y0 // 256), int(x1 // 256), int(y1 // 256)
    im = Image.new("RGB", ((tx1 - tx0 + 1) * 256, (ty1 - ty0 + 1) * 256))
    for tx in range(tx0, tx1 + 1):
        for ty in range(ty0, ty1 + 1):
            im.paste(tile(z, tx, ty), ((tx - tx0) * 256, (ty - ty0) * 256))
    ox, oy = tx0 * 256, ty0 * 256
    im = im.crop((int(x0 - ox), int(y0 - oy), int(x1 - ox), int(y1 - oy)))
    return im, (x0, y0)


def font(p, s):
    return ImageFont.truetype(p, s)


def label_box(d, xy, lines, fill, scale, anchor_pt=None, sub=None):
    f1 = font(FB, int(26 * scale)); f2 = font(FR, int(18 * scale))
    w = max(d.textlength(lines[0], font=f1), d.textlength(sub or "", font=f2)) + 28 * scale
    h = 38 * scale + (24 * scale if sub else 0)
    x, y = xy
    x0, y0 = x - w / 2, y - h / 2
    if anchor_pt:
        d.line([anchor_pt, (x, y)], fill=WHITE, width=max(2, int(3 * scale)))
        r = 7 * scale
        d.ellipse([anchor_pt[0] - r, anchor_pt[1] - r, anchor_pt[0] + r, anchor_pt[1] + r], fill=fill, outline=WHITE, width=max(2, int(3 * scale)))
    d.rounded_rectangle([x0, y0, x0 + w, y0 + h], radius=8 * scale, fill=fill, outline=WHITE, width=max(2, int(2 * scale)))
    d.text((x, y0 + 8 * scale), lines[0], font=f1, fill=WHITE, anchor="mt")
    if sub:
        d.text((x, y0 + 38 * scale), sub, font=f2, fill=WHITE, anchor="mt")


def subject_badge(d, pt, scale, name="MINUTEMAN PARK"):
    x, y = pt
    f1 = font(FS, int(34 * scale)); f2 = font(FS, int(17 * scale))
    w, h = 270 * scale, 78 * scale
    bx, by = x - w / 2, y - h - 34 * scale
    d.polygon([(x - 14 * scale, by + h), (x + 14 * scale, by + h), (x, y - 6 * scale)], fill=NAVY)
    d.rectangle([bx, by, bx + w, by + h], fill=NAVY, outline=WHITE, width=max(2, int(3 * scale)))
    d.text((x, by + 10 * scale), "HANOVER", font=f1, fill=WHITE, anchor="mt")
    d.line([(bx + 30 * scale, by + 50 * scale), (bx + w - 30 * scale, by + 50 * scale)], fill=WHITE, width=1)
    d.text((x, by + 55 * scale), name, font=f2, fill=WHITE, anchor="mt")
    r = 11 * scale
    d.ellipse([x - r, y - r, x + r, y + r], fill=WHITE, outline=NAVY, width=max(2, int(4 * scale)))


def num_pin(d, pt, n, scale, fill=ORANGE):
    x, y = pt; r = 24 * scale
    d.ellipse([x - r, y - r, x + r, y + r], fill=fill, outline=WHITE, width=max(2, int(4 * scale)))
    d.text((x, y), str(n), font=font(FB, int(28 * scale)), fill=WHITE, anchor="mm")


def make(name, bbox, z, subject, pois=(), pins=(), width=2600, darken=0.82, caption=None, rings=()):
    im, (ox, oy) = basemap(bbox, z)
    s = width / im.width
    im = im.resize((width, int(im.height * s)), Image.LANCZOS)
    im = Image.eval(im, lambda v: int(v * darken))
    d = ImageDraw.Draw(im)
    sc = width / 2600
    P = lambda lat, lon: ((wpx(lat, lon, z)[0] - ox) * s, (wpx(lat, lon, z)[1] - oy) * s)
    sp = P(*subject)
    for miles in rings:  # radius rings around subject
        lat, lon = subject
        dlat = miles / 69.0
        r = abs(P(lat + dlat, lon)[1] - sp[1])
        d.ellipse([sp[0] - r, sp[1] - r, sp[0] + r, sp[1] + r], outline=WHITE, width=max(2, int(3 * sc)))
        d.text((sp[0], sp[1] - r - 6 * sc), f"{miles} MILE", font=font(FB, int(22 * sc)), fill=WHITE, anchor="mb")
    for p in pois:  # dict(lat, lon, title, sub, dx, dy)
        a = P(p["lat"], p["lon"])
        label_box(d, (a[0] + p.get("dx", 0) * sc, a[1] + p.get("dy", -70) * sc), [p["title"]], p.get("fill", ORANGE), sc, anchor_pt=a, sub=p.get("sub"))
    for p in pins:
        num_pin(d, P(p["lat"], p["lon"]), p["n"], sc, p.get("fill", ORANGE))
    subject_badge(d, sp, sc)
    f = font(FR, int(16 * sc))
    txt = "Imagery: Esri, Maxar, Earthstar Geographics" + (f"  |  {caption}" if caption else "")
    tw = d.textlength(txt, font=f)
    d.rectangle([im.width - tw - 24 * sc, im.height - 30 * sc, im.width, im.height], fill=(0, 0, 0))
    d.text((im.width - 12 * sc, im.height - 15 * sc), txt, font=f, fill=WHITE, anchor="rm")
    p = os.path.join(OUT, name)
    im.save(p, quality=90)
    print(name, im.size)
    return p


SUBJ = (42.6931748, -71.2085521)

if __name__ == "__main__":
    # 1. Site aerial with points of interest (coords: Google Places via poi.json)
    pois = [
        dict(lat=42.6940678, lon=-71.2193602, title="PHILIPS", sub="MINUTEMAN PARK", dx=-230, dy=-40),
        dict(lat=42.688225, lon=-71.215805, title="MERCURY SYSTEMS", sub="50 MINUTEMAN RD", dx=-300, dy=70),
        dict(lat=42.6786052, lon=-71.1976241, title="SCHNEIDER ELECTRIC", sub="R&D CENTER", dx=60, dy=110),
        dict(lat=42.7048812, lon=-71.2014754, title="MARKET BASKET", sub="1.2 MI", dx=40, dy=-120),
        dict(lat=42.6978937, lon=-71.208099, title="MERRIMACK RIVER TRAIL", dx=-330, dy=-150),
        dict(lat=42.6857284, lon=-71.2118393, title="THE POINT AT MERRIMACK RIVER", sub="248 UNITS | HANOVER 2018", dx=-120, dy=170, fill=NAVY),
        dict(lat=42.6432881, lon=-71.1889269, title="RAYTHEON (RTX)", sub="10,000 EMPLOYEES", dx=-270, dy=-20),
        dict(lat=42.6573798, lon=-71.1450279, title="ANDOVER MBTA STATION", sub="HAVERHILL LINE", dx=-60, dy=-160),
        dict(lat=42.6550911, lon=-71.139565, title="DOWNTOWN ANDOVER", sub="MAIN STREET | WHOLE FOODS", dx=230, dy=-60),
        dict(lat=42.6492489, lon=-71.1319695, title="PHILLIPS ACADEMY", sub="ADDISON GALLERY", dx=190, dy=70),
        dict(lat=42.6571642, lon=-71.1554961, title="ANDOVER HIGH SCHOOL", sub="U.S. NEWS #38 IN MA", dx=-200, dy=110),
        dict(lat=42.6726839, lon=-71.1669523, title="ANDOVER COUNTRY CLUB", dx=60, dy=-110),
        dict(lat=42.6846681, lon=-71.1361867, title="NORTH ANDOVER MALL", sub="MARKET BASKET", dx=90, dy=-110),
        dict(lat=42.7436586, lon=-71.1594268, title="THE LOOP", sub="METHUEN RETAIL", dx=0, dy=90),
    ]
    make("site_aerial_poi.jpg", (42.633, -71.262, 42.752, -71.105), 15, SUBJ, pois=pois, width=2600)

    # 2. Rent comps (Datamart property coordinates)
    rc = [dict(n=1, lat=42.65070408582695, lon=-71.1843091249466),  # Slate
          dict(n=2, lat=42.63691753149041, lon=-71.074338555336),  # Berry Farms
          dict(n=3, lat=42.68572837114343, lon=-71.21183931827545, fill=NAVY),  # The Point (Hanover)
          dict(n=4, lat=42.72045224905023, lon=-71.11320912837982),  # Haven
          dict(n=5, lat=42.70448774099358, lon=-71.12665235996246),  # Avalon
          dict(n=6, lat=42.71139174699792, lon=-71.11989319324495)]  # 505 Sutton
    make("rent_comps_map.jpg", (42.615, -71.255, 42.745, -71.045), 14, SUBJ, pins=rc, width=2600)

    # 3. Supply pipeline (coords per subagents/pipeline/pipeline.json; East Mill coordinate is a proxy)
    pp = [dict(n=1, lat=42.7027, lon=-71.1304),  # East Mill
          dict(n=2, lat=42.7119164, lon=-71.1191918),  # 946 Osgood
          dict(n=3, lat=42.571006, lon=-71.1596422),  # Princeton Wilmington
          dict(n=4, lat=42.6081285, lon=-71.2288481),  # MacLellan Tewksbury
          dict(n=5, lat=42.70448774, lon=-71.12665236, fill=NAVY),  # Avalon (delivered)
          dict(n=6, lat=42.71139175, lon=-71.11989319, fill=NAVY)]  # 505 Sutton (delivered)
    make("pipeline_map.jpg", (42.555, -71.300, 42.745, -71.060), 14, SUBJ, pins=pp, width=2600, rings=(3, 5))

    # 4. Sale comps (Datamart coordinates)
    sc_ = [dict(n=1, lat=42.62004643678673, lon=-71.32143437862396),
           dict(n=2, lat=42.541924417018976, lon=-71.28705918788911),
           dict(n=3, lat=42.740423977375116, lon=-71.15863502025604),
           dict(n=4, lat=42.75350242853173, lon=-71.17054402828217),
           dict(n=5, lat=42.71201401948938, lon=-71.1402243375778),
           dict(n=6, lat=42.51728564500817, lon=-71.13894760608675)]
    make("sale_comps_map.jpg", (42.495, -71.380, 42.775, -71.090), 13, SUBJ, pins=sc_, width=2600)

    # 5. Close-up site aerial (z17), subject marker only
    make("site_closeup.jpg", (42.6865, -71.2215, 42.6995, -71.1957), 17, SUBJ, width=2600, darken=0.9,
         caption="Marker at geocoded address; parcel boundary not shown")

Command

40 lines

cd hanover/minuteman/scripts && python3 - <<'EOF'
s=open("maps.py").read()
s=s.replace('''def make(name, bbox, z, subject, pois=(), pins=(), width=2600, darken=0.82, caption=None, rings=(), ls=1.45):''','''import json
PARCEL = json.load(open(os.path.join(os.path.dirname(__file__), "..", "extracts", "parcel_300_minuteman.geojson")))["features"][0]["geometry"]["coordinates"][0]
PC = (sum(p[1] for p in PARCEL) / len(PARCEL), sum(p[0] for p in PARCEL) / len(PARCEL))  # parcel vertex-mean (lat, lon)


def make(name, bbox, z, subject, pois=(), pins=(), width=2600, darken=0.82, caption=None, rings=(), ls=1.45, parcel=False):''')
s=s.replace('''    for p in pins:
        num_pin(d, P(p["lat"], p["lon"]), p["n"], sc, p.get("fill", ORANGE))''','''    if parcel:
        pts = [P(la, lo) for lo, la in PARCEL]
        ov = Image.new("RGBA", im.size, (0, 0, 0, 0)); od = ImageDraw.Draw(ov)
        od.polygon(pts, fill=(240, 81, 36, 70))
        im.paste(Image.alpha_composite(im.convert("RGBA"), ov).convert("RGB")); d = ImageDraw.Draw(im)
        d.line(pts + [pts[0]], fill=ORANGE, width=max(3, int(5 * sc)), joint="curve")
    for p in pins:
        a = P(p["lat"], p["lon"])
        if "off" in p:
            b = (a[0] + p["off"][0] * sc, a[1] + p["off"][1] * sc)
            d.line([a, b], fill=WHITE, width=max(2, int(3 * sc)))
            r = 6 * sc; d.ellipse([a[0] - r, a[1] - r, a[0] + r, a[1] + r], fill=WHITE)
            a = b
        num_pin(d, a, p["n"], sc, p.get("fill", ORANGE))''')
s=s.replace('SUBJ = (42.6931748, -71.2085521)','SUBJ = None')
s=s.replace('if __name__ == "__main__":','if __name__ == "__main__":\n    SUBJ = PC')
# pipeline recenter and offsets
s=s.replace('''    pp = [dict(n=1, lat=42.7027, lon=-71.1304),  # East Mill
          dict(n=2, lat=42.7119164, lon=-71.1191918),  # 946 Osgood''','''    pp = [dict(n=1, lat=42.7027, lon=-71.1304, off=(-90, 60)),  # East Mill
          dict(n=2, lat=42.7119164, lon=-71.1191918, off=(40, -90)),  # 946 Osgood''')
s=s.replace('''          dict(n=5, lat=42.70448774, lon=-71.12665236, fill=NAVY),  # Avalon (delivered)
          dict(n=6, lat=42.71139175, lon=-71.11989319, fill=NAVY)]  # 505 Sutton (delivered)''','''          dict(n=5, lat=42.70448774, lon=-71.12665236, fill=NAVY, off=(90, 40)),  # Avalon (delivered)
          dict(n=6, lat=42.71139175, lon=-71.11989319, fill=NAVY, off=(110, -30))]  # 505 Sutton (delivered)''')
s=s.replace('box(42.640, -71.175, 12.5), 14, SUBJ, pins=pp, width=2600, rings=(3, 5))','box(42.650, -71.180, 12.8), 14, SUBJ, pins=pp, width=2600, rings=(3, 5))')
s=s.replace('''make("site_aerial_poi.jpg", box(42.688, -71.170, 7.8), 15, SUBJ, pois=pois, width=2600)''','''make("site_aerial_poi.jpg", box(42.688, -71.170, 7.8), 15, SUBJ, pois=pois, width=2600, parcel=True)''')
s=s.replace('''make("site_closeup.jpg", box(42.6925, -71.2085, 1.0), 17, SUBJ, width=2600, darken=0.9,
         caption="Marker at geocoded address; parcel boundary not shown")''','''make("site_closeup.jpg", box(PC[0] - 0.0005, PC[1], 0.95), 17, SUBJ, width=2600, darken=0.95, parcel=True,
         caption="Parcel 165-4 boundary: MassGIS Property Tax Parcels (FY2026)")''')
open("maps.py","w").write(s)
EOF
cd .. && python3 scripts/maps.py && cd assets/maps && for f in site_aerial_poi pipeline_map site_closeup; do convert $f.jpg -resize 1100x qa_$f.jpg; done && montage qa_pipeline_map.jpg qa_site_closeup.jpg -tile 2x1 -geometry 700x452+4+4 qa_all.jpg

Now the calc script for the derived figures, then the deck build.

Content

39 lines

"""Derived figures for the Hanover Minuteman Park investment book.
Inputs: RealAI Datamart (rent index as of 2026-09-26; SuperCensus 2026-08-15), FY2026 HUD 80% AMI limits
(Lawrence MA-NH HMFA, per extracts/affordable_rents.json), project program per Eagle-Tribune / Bisnow."""
import json, os
X = os.path.join(os.path.dirname(__file__), "..", "extracts")

# Program (Eagle-Tribune; Bisnow; Andover News)
units, aff_units, parking, acres, land = 334, 84, 515, 20.24, 2_500_000
mix = {"1BR": 201, "2BR": 110, "3BR": 23}
print(f"Market-rate units: {units - aff_units}; affordable share {aff_units/units:.1%}")
print(f"Parking ratio: {parking/units:.2f} per unit; density {units/acres:.1f} units/acre")
print(f"Land basis: ${land/units:,.0f}/unit; ${land/acres:,.0f}/acre")
print("Unit mix: " + ", ".join(f"{k} {v} ({v/units:.0%})" for k, v in mix.items()))

# Comp-set average in-place rent by bedroom (6 comps; simple average of reporting properties)
comp = {"1BR": [2408.32, 2515.42, 2561.99, 2512.29, 2475.56, 2799.66],
        "2BR": [2962.80, 2991.56, 3156.15, 3176.32, 2992.63, 3248.26],
        "3BR": [4171.46, 3736.07, 3886.40, 4051.36]}
avg = {k: sum(v) / len(v) for k, v in comp.items()}
# FY2026 80% AMI max gross rent, 1.5 persons/bedroom (affordable_rents.json)
aff = json.load(open(os.path.join(X, "affordable_rents.json")))["results"]["FY2026"]
for k in ["1BR", "2BR", "3BR"]:
    a = aff[k]["max_gross_rent_1.5"]; inc = aff[k]["income_limit_1.5"]
    print(f"{k}: comp avg in-place ${avg[k]:,.0f}; target income at 3.0x rent ${avg[k]*36:,.0f}; "
          f"80% AMI max gross rent ${a:,.0f} (income limit ${inc:,.0f}); affordable discount {a/avg[k]-1:+.1%}")
renter_med, zip_med = 95223, 180672
for k in avg:
    print(f"  {k} target income vs 01810 renter median ${renter_med:,}: {avg[k]*36/renter_med-1:+.0%}; vs all-HH median ${zip_med:,}: {avg[k]*36/zip_med-1:+.0%}")
# Weighted market rent reference across the proposed mix (illustrative, comp averages)
w = sum(avg[k] * mix[k] for k in mix) / units
print(f"Mix-weighted comp in-place rent reference: ${w:,.0f}/mo")

# Sale comps $/unit range (Datamart latest sale)
ppu = {"Crosspoint": 354479, "Villas at Old Concord": 353395, "Point at Merrimack Valley": 372436,
       "Woods at Merrimack": 268121, "Legacy Park": 290385, "Point at Woburn": 452166}
vals = sorted(ppu.values())
print(f"Sale comps $/unit: low ${vals[0]:,}, high ${vals[-1]:,}, median ${(vals[2]+vals[3])/2:,.0f}; 2020+ vintage avg "
      f"${(354479+372436+452166)/3:,.0f}")

Content

669 lines

"""Hanover Minuteman Park — investment book in the style of Hanover's Sugarloaf book (letter landscape).
Figures: RealAI Datamart, public sources noted on each page, scripts/book_calcs.py. Placeholders marked [TBD]."""
import os
from PIL import Image
from pptx import Presentation
from pptx.util import Inches, Pt
from pptx.dml.color import RGBColor
from pptx.enum.text import PP_ALIGN, MSO_ANCHOR
from pptx.enum.shapes import MSO_SHAPE
from pptx.chart.data import CategoryChartData
from pptx.enum.chart import XL_CHART_TYPE, XL_LEGEND_POSITION, XL_LABEL_POSITION
from pptx.oxml.ns import qn

ROOT = os.path.join(os.path.dirname(__file__), "..")
A = lambda *p: os.path.join(ROOT, "assets", *p)
os.makedirs(A("crops"), exist_ok=True)

NAVY = RGBColor(0x0A, 0x22, 0x40); SLATE = RGBColor(0x47, 0x5A, 0x70); STEEL = RGBColor(0x5B, 0x7C, 0x99)
GRAY = RGBColor(0xA2, 0xAA, 0xAD); LGRAY = RGBColor(0xE8, 0xEA, 0xEB); ORANGE = RGBColor(0xF0, 0x51, 0x24)
TXT = RGBColor(0x23, 0x23, 0x23); WHITE = RGBColor(0xFF, 0xFF, 0xFF); TBDFILL = RGBColor(0xFD, 0xEE, 0xE8)
SERIF = "Garamond"; SANS = "Century Gothic"
NAME = "Hanover Minuteman Park"

prs = Presentation()
prs.slide_width, prs.slide_height = Inches(11), Inches(8.5)
W, H = prs.slide_width, prs.slide_height
BL = prs.slide_layouts[6]
I = Inches


def crop(src, w, h, fy=0.5):
    im = Image.open(A("photos", src) if not os.path.isabs(src) and not src.startswith("maps") else A(src)).convert("RGB")
    tr = w / h; r = im.width / im.height
    if r > tr:
        nw = int(im.height * tr); x0 = (im.width - nw) // 2; im = im.crop((x0, 0, x0 + nw, im.height))
    else:
        nh = int(im.width / tr); y0 = int((im.height - nh) * fy); im = im.crop((0, y0, im.width, y0 + nh))
    if im.width > 2400: im = im.resize((2400, int(2400 / tr)), Image.LANCZOS)
    out = A("crops", f"{os.path.basename(src).rsplit('.',1)[0]}_{w:.2f}x{h:.2f}.jpg"); im.save(out, quality=88)
    return out


def pic(s, src, x, y, w, h, fy=0.5):
    return s.shapes.add_picture(crop(src, w, h, fy), I(x), I(y), I(w), I(h))


def spacing(run, pts):
    run._r.get_or_add_rPr().set("spc", str(int(pts * 100)))


def tb(s, x, y, w, h, text, size=11, color=TXT, font=SERIF, bold=False, align=PP_ALIGN.LEFT, italic=False,
       anchor=MSO_ANCHOR.TOP, sp=0, line=None, caps=False):
    box = s.shapes.add_textbox(I(x), I(y), I(w), I(h)); tf = box.text_frame; tf.word_wrap = True
    tf.margin_left = tf.margin_right = I(0.02); tf.margin_top = tf.margin_bottom = I(0.01); tf.vertical_anchor = anchor
    paras = text if isinstance(text, list) else [text]
    for i, ptxt in enumerate(paras):
        p = tf.paragraphs[0] if i == 0 else tf.add_paragraph(); p.alignment = align
        if line: p.line_spacing = line
        runs = ptxt if isinstance(ptxt, list) else [ptxt]
        for r_ in runs:
            if isinstance(r_, tuple):
                t, kw = r_[0], r_[1]
            else:
                t, kw = r_, {}
            r = p.add_run(); r.text = t.upper() if caps else t
            f = r.font; f.name = kw.get("font", font); f.size = Pt(kw.get("size", size)); f.bold = kw.get("bold", bold)
            f.italic = kw.get("italic", italic); f.color.rgb = kw.get("color", color)
            if sp: spacing(r, sp)
    return box


def rect(s, x, y, w, h, fill, line=None):
    r = s.shapes.add_shape(MSO_SHAPE.RECTANGLE, I(x), I(y), I(w), I(h)); r.fill.solid(); r.fill.fore_color.rgb = fill
    if line is None: r.line.fill.background()
    else: r.line.color.rgb = line; r.line.width = Pt(0.75)
    r.shadow.inherit = False
    return r


def hline(s, x, y, w, color=GRAY, wt=0.75):
    ln = s.shapes.add_connector(1, I(x), I(y), I(x + w), I(y)); ln.line.color.rgb = color; ln.line.width = Pt(wt); return ln


def page(title, num, sub=None):
    s = prs.slides.add_slide(BL)
    tb(s, 0.5, 0.32, 9.5, 0.6, title, 27, NAVY, SERIF, False, sp=1.5, caps=False)
    hline(s, 0, 0.95, 5.2 if len(title) < 28 else 7.2, GRAY)
    if sub: tb(s, 0.5, 1.0, 10, 0.3, sub, 10, STEEL, SANS, True, sp=0.5, caps=True)
    # footer like Sugarloaf: number + wordmark + rule
    even = num % 2 == 0
    if even:
        tb(s, 0.5, 8.05, 0.35, 0.25, str(num), 10, NAVY, SERIF, True)
        tb(s, 0.85, 8.05, 1.4, 0.25, "HANOVER", 10, GRAY, SERIF, sp=2)
        hline(s, 2.15, 8.17, 8.85, GRAY, 0.5)
    else:
        hline(s, 0, 8.17, 7.5, GRAY, 0.5)
        tb(s, 7.6, 8.05, 2.6, 0.25, NAME.upper(), 10, GRAY, SERIF, align=PP_ALIGN.RIGHT, sp=1)
        tb(s, 10.2, 8.05, 0.35, 0.25, str(num), 10, NAVY, SERIF, True, align=PP_ALIGN.RIGHT)
    return s


def head(s, x, y, w, text, color=STEEL, size=11):
    return tb(s, x, y, w, 0.3, text, size, color, SANS, True, caps=True, sp=0.3)


def body(s, x, y, w, h, paras, size=11, color=TXT):
    box = tb(s, x, y, w, h, paras, size, color, SERIF, line=1.08)
    for p in box.text_frame.paragraphs: p.alignment = PP_ALIGN.JUSTIFY; p.space_after = Pt(5)
    return box


def bullets(s, x, y, w, h, items, size=10.5, color=TXT, gap=4):
    box = s.shapes.add_textbox(I(x), I(y), I(w), I(h)); tf = box.text_frame; tf.word_wrap = True
    for i, it in enumerate(items):
        p = tf.paragraphs[0] if i == 0 else tf.add_paragraph(); p.space_after = Pt(gap)
        hd, txt = it if isinstance(it, tuple) else (None, it)
        r = p.add_run(); r.text = "•  "; r.font.size = Pt(size); r.font.color.rgb = ORANGE; r.font.name = SERIF
        if hd:
            r = p.add_run(); r.text = hd + " "; r.font.bold = True; r.font.size = Pt(size); r.font.color.rgb = NAVY; r.font.name = SERIF
        r = p.add_run(); r.text = txt; r.font.size = Pt(size); r.font.color.rgb = color; r.font.name = SERIF
    return box


def src(s, text, y=7.72, h=0.3):
    tb(s, 0.5, y, 10, h, text, 7, GRAY, SANS, italic=False)


def table(s, x, y, w, rows, cw, rh=0.28, size=8.5, hdr=NAVY, bold_rows=(), shade=(), tbd_cells=(), align_first=PP_ALIGN.LEFT,
          section=()):
    nr, nc = len(rows), len(rows[0])
    gs = s.shapes.add_table(nr, nc, I(x), I(y), I(w), I(rh * nr)); t = gs.table
    tp = gs._element.graphic.graphicData.tbl.tblPr; tp.set("bandRow", "0"); tp.set("firstRow", "0")
    for j, c in enumerate(cw): t.columns[j].width = I(c)
    for i, r in enumerate(rows):
        t.rows[i].height = I(rh)
        for j, v in enumerate(r):
            c = t.cell(i, j); c.margin_left = c.margin_right = I(0.05); c.margin_top = c.margin_bottom = I(0.01)
            c.vertical_anchor = MSO_ANCHOR.MIDDLE; tf = c.text_frame; tf.word_wrap = True
            p = tf.paragraphs[0]; p.alignment = align_first if j == 0 else PP_ALIGN.CENTER
            run = p.add_run(); run.text = str(v); run.font.size = Pt(size); run.font.name = SANS
            c.fill.solid()
            is_tbd = (i, j) in tbd_cells or str(v).startswith("[TBD")
            if i == 0:
                c.fill.fore_color.rgb = hdr; run.font.color.rgb = WHITE; run.font.bold = True
            elif i in section:
                c.fill.fore_color.rgb = SLATE; run.font.color.rgb = WHITE; run.font.bold = True
            elif is_tbd:
                c.fill.fore_color.rgb = TBDFILL; run.font.color.rgb = ORANGE; run.font.italic = True
            elif i in shade:
                c.fill.fore_color.rgb = LGRAY; run.font.color.rgb = NAVY; run.font.bold = True
            else:
                c.fill.fore_color.rgb = WHITE if i % 2 else RGBColor(0xF5, 0xF6, 0xF7); run.font.color.rgb = TXT
                run.font.bold = i in bold_rows
    for i in section: t.cell(i, 0).merge(t.cell(i, nc - 1))
    return t


def stat_panel(s, x, y, w, h, stats, title=None):
    rect(s, x, y, w, h, NAVY)
    yy = y + 0.2
    if title:
        tb(s, x + 0.2, yy, w - 0.4, 0.3, title, 10, GRAY, SANS, True, caps=True, sp=0.5); yy += 0.4
    step = (h - (yy - y) - 0.1) / len(stats)
    for v, l in stats:
        tb(s, x + 0.2, yy, w - 0.4, 0.45, v, 22, WHITE, SERIF, sp=0.5)
        tb(s, x + 0.2, yy + 0.43, w - 0.4, 0.3, l, 8, GRAY, SANS, True, caps=True, sp=0.3)
        yy += step


def tbd_tag(s, x, y, text="PLACEHOLDER — HANOVER TO PROVIDE"):
    r = rect(s, x, y, 3.2, 0.3, ORANGE)
    tb(s, x, y + 0.03, 3.2, 0.25, text, 8.5, WHITE, SANS, True, align=PP_ALIGN.CENTER, sp=0.5)


def caption(s, x, y, w, title, sub=None, dark=False):
    tb(s, x, y, w, 0.25, title, 9.5, NAVY if not dark else WHITE, SANS, True, caps=True, sp=0.4)
    if sub: tb(s, x, y + 0.25, w, 0.6, sub, 9, TXT if not dark else WHITE, SERIF)


# =============== 1. COVER ===============
s = prs.slides.add_slide(BL)
pic(s, "stoneham_exterior.jpg", 0, 1.95, 11, 6.55, 0.55)
rect(s, 0, 0, 11, 1.95, NAVY)
tb(s, 0, 0.28, 11, 0.7, NAME, 36, WHITE, SERIF, align=PP_ALIGN.CENTER, sp=1.5)
hline(s, 2.3, 1.0, 6.4, WHITE, 0.5)
tb(s, 0, 1.03, 11, 0.35, "334-Unit Multifamily Development Opportunity", 12, WHITE, SERIF, align=PP_ALIGN.CENTER, sp=2.5, caps=True)
hline(s, 2.3, 1.38, 6.4, WHITE, 0.5)
tb(s, 0, 1.45, 11, 0.4, "Andover  |  Massachusetts", 13, WHITE, SERIF, align=PP_ALIGN.CENTER, sp=3, caps=True)
tb(s, 0.45, 7.45, 4, 0.55, "HANOVER", 30, WHITE, SERIF, bold=True, sp=2)
tb(s, 0.47, 8.0, 6, 0.3, "Photo of similar Hanover project (Hanover Stoneham); subject project will vary", 9, WHITE, SERIF, italic=True)

# =============== 2. FULL-BLEED AREA PHOTO ===============
s = prs.slides.add_slide(BL)
pic(s, "downtown_andover_main_street.jpg", 0, 0, 11, 8.5, 0.6)
rect(s, 0, 7.75, 11, 0.75, NAVY)
tb(s, 0.5, 7.92, 6, 0.4, "Downtown Andover  —  Main Street", 13, WHITE, SERIF, sp=1.5, caps=True)
tb(s, 8.3, 7.86, 2.3, 0.45, "HANOVER", 22, WHITE, SERIF, bold=True, align=PP_ALIGN.RIGHT, sp=2)
tb(s, 0.5, 7.5, 6, 0.25, "Photo: John Phelan, CC BY 3.0, via Wikimedia Commons", 7, WHITE, SANS)

# =============== 3. TABLE OF CONTENTS ===============
s = page("Table of Contents", 3)
pic(s, "phillips_academy_memorial_bell_tower.jpg", 6.6, 0, 4.4, 8.0, 0.35)
toc = [("Executive Summary", 4), ("Site Aerial", 5), ("Financial Summary", 6), ("Cost & Operations Assumptions", 8),
       ("Why Minuteman Park", 10), ("Nearby Points of Interest", 11), ("Site Overview", 12), ("Building Summary & Entitlement", 13),
       ("Sample Renderings", 14), ("Demographics", 16), ("Supply Pipeline", 18), ("Rent Comps", 19), ("Leasing Momentum", 23),
       ("Asset Sale Comps", 24), ("Sponsor Track Record", 25), ("Key Contacts", 26)]
for i, (t, n) in enumerate(toc):
    y = 1.35 + i * 0.39
    tb(s, 0.7, y, 4.6, 0.32, t, 14, NAVY, SERIF)
    tb(s, 5.3, y, 0.8, 0.32, str(n), 14, SLATE, SERIF, align=PP_ALIGN.RIGHT)
    hline(s, 0.7, y + 0.34, 5.4, LGRAY, 0.5)
tb(s, 6.75, 7.65, 4, 0.3, "Phillips Academy Memorial Bell Tower. Photo: Daderot, public domain", 7, WHITE, SANS)

# =============== 4. EXECUTIVE SUMMARY ===============
s = page("Executive Summary", 4)
head(s, 0.5, 1.15, 3.5, "Overview")
body(s, 0.5, 1.42, 3.45, 6.2, [
    "Hanover Company (\u201cHanover\u201d) acquired an approximately 20.24-acre vacant site at 300 Minuteman Road in Andover, "
    "Massachusetts (the \u201cSite\u201d) in May 2026 for $2.5 million. Hanover intends to develop a 334-unit, Class A "
    "multifamily community (the \u201cProject\u201d) across three four- and five-story buildings with 515 parking spaces.",
    "The Project is proposed through the state\u2019s Local Initiative Program (\u201cfriendly 40B\u201d), with 84 units (25%) "
    "reserved for households earning up to 80% of area median income and 250 market-rate units. Roughly two acres are "
    "reserved for future commercial development.",
    "The Site sits inside Minuteman Park, a 1M+ SF office, R&D and lab campus directly off I-93, 25 miles north of Boston, "
    "next to the town\u2019s MBTA Multifamily Overlay District. It was previously approved for a 224,500 SF pharmaceutical "
    "manufacturing facility that did not advance."])
head(s, 4.2, 1.15, 3.4, "Location & Market")
body(s, 4.2, 1.42, 3.45, 6.2, [
    "Andover pairs one of Greater Boston\u2019s strongest employment bases (Raytheon alone employs ~10,000 in town) with a "
    "high-income renter pool. ZIP 01810\u2019s median household income is $180,672, about 45% above the Boston MSA, and "
    "households moving in earn $45K more than those moving out.",
    "Supply is limited. No 120+ unit community has delivered in Andover or North Andover since 2022, and Andover permitted just "
    "nine new homes in 2023. New product in the comp set earns 9–12% premiums on two- and three-bedroom units, and ZIP occupancy "
    "is 96.7%.",
    "Hanover built The Point at Merrimack River (248 units, 2018) about half a mile from the Site. It is 96.8% occupied."])
stat_panel(s, 7.95, 1.15, 2.55, 6.45, [("334", "Total units"), ("84 / 25%", "Affordable at ≤80% AMI"), ("20.24 AC", "Site area"),
                                         ("$2.5M", "Land basis ($7,485 / unit)"), ("515", "Parking spaces (1.54 / unit)"),
                                         ("$1.5M", "Est. annual property tax")], "Project at a Glance")
src(s, "Sources: Bisnow (May 2026); Andover News & Eagle-Tribune (Sep 2026); Andover Ledger; Newmark; RealAI Datamart (SuperCensus Aug 2026; Rent Index Sep 26, 2026). "
       "Property tax figure is Hanover's estimate as presented to the Select Board.")

# =============== 5. SITE AERIAL ===============
s = page("Site Aerial", 5)
s.shapes.add_picture(A("maps", "site_aerial_poi.jpg"), I(0.5), I(1.15), I(10.0), I(6.45))
src(s, "Imagery: Esri, Maxar, Earthstar Geographics. Site outline: MassGIS Property Tax Parcels (parcel 165-4). Locations: Google Places; RealAI Datamart. "
       "Raytheon employment per Andover News (2024); Andover High ranking per U.S. News.", y=7.65)

# =============== 6. FINANCIAL SUMMARY (PLACEHOLDER) ===============
s = page("Financial Summary", 6)
tbd_tag(s, 7.3, 0.45)
head(s, 0.5, 1.15, 4, "Project Summary")
rows = [["Item", "Value"], ["Product Type", "Suburban garden / mid-rise"], ["Construction Type", "[TBD]"],
        ["Levels", "4 and 5 stories (3 buildings)"], ["Units", "334 (250 market / 84 affordable)"], ["Site Area", "20.24 acres (16.5 units / acre)"],
        ["NRSF", "[TBD]"], ["Average Unit Size", "[TBD]"], ["Going-In Rent (PSF / Nominal)", "[TBD]"],
        ["Parking", "515 spaces / 1.54 per unit"], ["Land Closing", "May 2026 (closed)"], ["Construction Start", "[TBD]"],
        ["Construction Duration", "[TBD]"], ["First Units Delivered", "[TBD]"], ["Stabilization", "[TBD]"]]
table(s, 0.5, 1.45, 4.6, rows, [2.3, 2.3], rh=0.36, size=9)
head(s, 5.5, 1.15, 5, "Returns")
for i, (lbl) in enumerate(["Going-In Return on Cost", "Stabilized Return on Cost", "Unlevered Project IRR", "Levered Project IRR",
                            "Equity Multiple", "Total Project Cost"]):
    x = 5.5 + (i % 2) * 2.55; y = 1.45 + (i // 2) * 1.35
    rect(s, x, y, 2.4, 1.2, LGRAY)
    tb(s, x, y + 0.18, 2.4, 0.5, "[TBD]", 24, ORANGE, SERIF, align=PP_ALIGN.CENTER, italic=True)
    tb(s, x, y + 0.75, 2.4, 0.3, lbl, 8.5, NAVY, SANS, True, align=PP_ALIGN.CENTER, caps=True)
head(s, 5.5, 5.6, 5, "Market Reference Points (Not Underwriting)")
bullets(s, 5.5, 5.88, 5.0, 1.8, [
    ("Comp in-place rent:", "$2,822/mo average, weighted to the proposed unit mix (six 2016–2022 Andover/North Andover comps)."),
    ("80% AMI rent caps (FY2026):", "1BR $2,003, 2BR $2,404, 3BR $2,777 gross, before utility allowance."),
    ("Sale comps:", "$268K–$452K/unit across six 2024–2026 trades. 2020+ vintages average $393K/unit.")], size=9.5)
src(s, "[TBD] = to be provided from Hanover's underwriting model. Program per Eagle-Tribune / Bisnow; reference points from RealAI Datamart and HUD FY2026 income limits (Lawrence, MA-NH HMFA).")

# =============== 7. UNIT MIX + SOURCES & USES (PLACEHOLDER) ===============
s = page("Financial Summary", 7, "Unit Mix  |  Sources & Uses")
tbd_tag(s, 7.3, 0.45)
rows = [["Unit Mix", "Units", "% Total", "Market", "Affordable", "Avg SF"],
        ["One Bedrooms", "201", "60%", "[TBD]", "[TBD]", "[TBD]"], ["Two Bedrooms", "110", "33%", "[TBD]", "[TBD]", "[TBD]"],
        ["Three Bedrooms", "23", "7%", "[TBD]", "[TBD]", "[TBD]"], ["Total", "334", "100%", "250", "84", "[TBD]"]]
table(s, 0.5, 1.5, 6.2, rows, [1.7, 0.8, 0.8, 0.95, 0.95, 1.0], rh=0.34, size=9, shade=(4,))
uses = [["Uses", "Total", "Per Unit", "Per NRSF"], ["Land Costs", "$2,500,000", "$7,485", "[TBD]"], ["Hard Costs", "[TBD]", "[TBD]", "[TBD]"],
        ["Soft Costs", "[TBD]", "[TBD]", "[TBD]"], ["Developer Fee", "[TBD]", "[TBD]", "[TBD]"], ["Contingency", "[TBD]", "[TBD]", "[TBD]"],
        ["Financing Costs", "[TBD]", "[TBD]", "[TBD]"], ["Total Uses", "[TBD]", "[TBD]", "[TBD]"]]
table(s, 0.5, 3.55, 4.9, uses, [1.6, 1.15, 1.05, 1.1], rh=0.34, size=9, shade=(7,))
srcs = [["Sources", "Total", "Per Unit", "% Total"], ["Limited Partner", "[TBD]", "[TBD]", "[TBD]"], ["Hanover (GP)", "[TBD]", "[TBD]", "[TBD]"],
        ["Construction Loan", "[TBD]", "[TBD]", "[TBD]"], ["Total Sources", "[TBD]", "[TBD]", "[TBD]"]]
table(s, 5.65, 3.55, 4.85, srcs, [1.6, 1.15, 1.05, 1.05], rh=0.34, size=9, shade=(4,))
src(s, "Unit mix and affordable count per Eagle-Tribune (Sep 2026); the split of affordable units by bedroom is not public. Land cost per Bisnow / public records ($2.5M, May 2026).")

# =============== 8. COST ASSUMPTIONS (PLACEHOLDER) ===============
s = page("Cost Assumptions", 8)
tbd_tag(s, 7.3, 0.45)
blocks = [("Project Name & Location", [["Project Name", NAME], ["Project Location", "Andover, MA"], ["Land Closing", "May 2026"],
                                        ["Construction Start", "[TBD]"], ["Construction Duration", "[TBD]"]]),
          ("Operating & Sale Assumptions", [["Residential Cap Rate", "[TBD]"], ["Retail / Commercial Cap Rate", "[TBD]"], ["Income Growth", "[TBD]"],
                                             ["Expense Growth", "[TBD]"], ["Closing Costs on Sale", "[TBD]"]]),
          ("Equity Assumptions", [["LP / GP Contributions", "[TBD]"], ["Total Equity", "[TBD]"], ["Preferred Return", "[TBD]"],
                                   ["Promote Structure", "[TBD]"], ["Equity Timing", "[TBD]"]]),
          ("Debt & Additional Inputs", [["Construction Loan LTC", "[TBD]"], ["Interest Rate / Index", "[TBD]"], ["Loan Term", "[TBD]"],
                                         ["Hard Cost Contingency", "[TBD]"], ["Developer Fee", "[TBD]"]])]
for k, (t, rws) in enumerate(blocks):
    x = 0.5 + (k % 2) * 5.1; y = 1.25 + (k // 2) * 3.0
    head(s, x, y, 4.8, t)
    table(s, x, y + 0.32, 4.9, [["Input", "Assumption"]] + rws, [2.5, 2.4], rh=0.36, size=9)
src(s, "All [TBD] inputs to be populated from Hanover's underwriting model; none of these figures have been sourced.")

# =============== 9. OPERATIONS ASSUMPTIONS (PLACEHOLDER) ===============
s = page("Operations Assumptions", 9)
tbd_tag(s, 7.3, 0.45)
head(s, 0.5, 1.15, 6, "Market Unit Mix")
rows = [["Unit Type", "Units", "Avg SF", "Comp In-Place Rent¹", "80% AMI Max Gross Rent²", "Underwritten Rent", "Underwritten $/SF"],
        ["1BR", "201", "[TBD]", "$2,546", "$2,003", "[TBD]", "[TBD]"], ["2BR", "110", "[TBD]", "$3,088", "$2,404", "[TBD]", "[TBD]"],
        ["3BR", "23", "[TBD]", "$3,961", "$2,777", "[TBD]", "[TBD]"], ["Blended", "334", "[TBD]", "$2,822", "—", "[TBD]", "[TBD]"]]
table(s, 0.5, 1.45, 10.0, rows, [1.2, 0.8, 0.9, 1.75, 1.95, 1.7, 1.7], rh=0.36, size=9, shade=(4,))
head(s, 0.5, 3.45, 6, "Annual Operating Expenses")
ox = [["Expense Category", "Input Type", "Total", "Per Unit", "Per SF"]] + [[c, "[TBD]", "[TBD]", "[TBD]", "[TBD]"] for c in
      ["Controllable (payroll, R&M, turnover, marketing, G&A)", "Management Fee", "Real Estate Taxes", "Insurance", "Utilities", "Replacement Reserves"]] + \
     [["Total Operating Expenses", "", "[TBD]", "[TBD]", "[TBD]"]]
table(s, 0.5, 3.75, 6.6, ox, [2.9, 0.9, 0.95, 0.95, 0.9], rh=0.34, size=8.5, shade=(7,))
head(s, 7.4, 3.45, 3, "Other Income")
table(s, 7.4, 3.75, 3.1, [["Item", "Assumption"], ["Parking", "[TBD]"], ["Amenity Fee", "[TBD]"], ["Storage", "[TBD]"], ["Pet / Other", "[TBD]"],
                          ["Vacancy & Credit Loss", "[TBD]"]], [1.7, 1.4], rh=0.34, size=8.5)
src(s, "¹ Simple average in-place rent of six 2016–2022 comps (RealAI Rent Index, Sep 26, 2026); the blended figure is weighted to the proposed mix.  ² FY2026 HUD 80% AMI limits, "
       "Lawrence MA-NH HMFA, 30% of income, 1.5 persons/bedroom, gross before utility allowance; confirm against EOHLC LIP guidelines.", y=7.55, h=0.45)

# =============== 10. WHY MINUTEMAN PARK ===============
s = page("Why Minuteman Park?", 10)
pic(s, "andover_mbta_commuter_rail_station.jpg", 7.15, 1.15, 3.35, 2.4)
tb(s, 7.15, 3.57, 3.35, 0.25, "Andover MBTA station (Haverhill Line). Photo: John Phelan, CC BY-SA 3.0", 7, GRAY, SANS)
body(s, 0.5, 1.15, 6.4, 3.6, [
    [("Economic base. ", {"bold": True, "color": NAVY}), "Andover is a diversified employment hub where two interstates meet. Raytheon is the largest "
     "employer, with about 10,000 employees in 2024 (nearly a third of the town\u2019s private-sector jobs), followed by Pfizer and Smith & Nephew. "
     "Minuteman Park is 1M+ SF of office, R&D and lab space; eight of its assets traded for $341 million in 2022. Philips and Mercury Systems sit within a mile of the Site."],
    [("Access. ", {"bold": True, "color": NAVY}), "The Site is directly off I-93, five miles south of the New Hampshire border and 25 miles north of Boston, "
     "with I-495 just to the south. Two MBTA Commuter Rail stations serve Andover, with scheduled trips to North Station of about 52 minutes."],
    [("Quality of life. ", {"bold": True, "color": NAVY}), "Andover High ranks #38 in Massachusetts (U.S. News, 97% graduation rate). Phillips Academy "
     "and the Addison Gallery anchor the town\u2019s cultural life. Market Basket is 1.2 miles away and the Merrimack River trail system less than half a mile."]], size=11)
stats = [("~10,000", "Raytheon employees in Andover"), ("25 mi", "To Boston via I-93"), ("#38", "Andover High, U.S. News MA ranking"),
         ("$341M", "2022 Minuteman Park asset trade"), ("$180.7K", "ZIP 01810 median HH income"), ("10x", "Projected tax revenue vs. today")]
for i, (v, l) in enumerate(stats):
    x = 0.5 + (i % 3) * 3.35; y = 4.95 + (i // 3) * 1.25
    rect(s, x, y, 3.15, 1.1, NAVY if i % 2 == 0 else SLATE)
    tb(s, x + 0.15, y + 0.12, 2.9, 0.5, v, 24, WHITE, SERIF)
    tb(s, x + 0.15, y + 0.68, 2.9, 0.35, l, 8.5, WHITE, SANS, True, caps=True)
src(s, "Sources: Andover News (Andover by the Numbers: employers; top taxpayers); Newmark (2022); U.S. News Best High Schools; MBTA timetable via Transit app; Andover Ledger (Sep 2026; "
       "$1.5M vs ~$150K tax); RealAI SuperCensus. Distances approximate.", y=7.5, h=0.45)

# =============== 11. POINTS OF INTEREST ===============
s = page("Nearby Points of Interest", 11)
pois = [("downtown_andover_main_street.jpg", "Downtown Andover", "Main Street dining and shops, Whole Foods and the commuter rail station, ~5.7 mi.", "John Phelan, CC BY 3.0"),
        ("addison_gallery_phillips_academy_andover.jpg", "Phillips Academy & Addison Gallery", "Phillips Academy and its free Addison Gallery of American Art, ~6.4 mi.", "Daderot, public domain"),
        ("harold_parker_state_forest_berry_pond.jpg", "Harold Parker State Forest", "3,000+ acres of trails and ponds, reached via I-93/I-495.", "John Phelan, CC BY-SA 4.0"),
        ("i495_bridge_merrimack_river_lawrence.jpg", "Merrimack River & I-495", "The river borders Minuteman Park. The Merrimack River Trail is ~0.4 mi from the Site.", "John Phelan, CC BY 3.0")]
for i, (f, t, d, cr) in enumerate(pois):
    x = 0.5 + (i % 2) * 5.1; y = 1.15 + (i // 2) * 3.25
    pic(s, f, x, y, 4.9, 2.35)
    caption(s, x, y + 2.4, 4.9, t, d)
    tb(s, x, y + 2.95, 4.9, 0.2, "Photo: " + cr + ", via Wikimedia Commons", 6.5, GRAY, SANS)
src(s, "Distances are approximate straight-line × 1.3 drive factor from the Site. Harold Parker State Forest acreage per Massachusetts DCR (confirm). Photo locations per Wikimedia Commons metadata.")

# =============== 12. SITE OVERVIEW ===============
s = page("Site Overview", 12)
s.shapes.add_picture(A("maps", "site_closeup.jpg"), I(0.5), I(1.15), I(6.6), I(4.26))
head(s, 7.35, 1.15, 3.2, "Site Facts")
rows = [["Item", "Detail"], ["Address", "300 Minuteman Rd, Andover MA"], ["Parcel", "Map 165, Lot 4"], ["Site Area", "20.24 acres"],
        ["Zoning", "ID2 (Industrial)"], ["Current Use", "Vacant, developable industrial land"], ["Assessed Value (FY26)", "$6,344,300"],
        ["Acquisition", "May 2026, $2.5M (Spear Street Capital)"], ["Prior Approval", "224,500 SF cGMP facility (2023)"],
        ["Adjacency", "MBTA Multifamily Overlay District"]]
table(s, 7.35, 1.45, 3.15, rows, [1.25, 1.9], rh=0.395, size=8)
head(s, 0.5, 5.6, 6, "Site Highlights")
bullets(s, 0.5, 5.88, 10.0, 1.6, [
    ("Infrastructure in place:", "Site work was originally built for a 320,000 SF office building. Hanover expects to complete the sewer work required of the prior project."),
    ("Highway frontage:", "The parcel sits along I-93 at the River Road interchange, next to Philips, Mercury Systems and the Alexandria lab campus."),
    ("Catalyst for Minuteman Park:", "Hanover's preliminary park-wide vision adds housing, new roads and grocery-anchored retail. Only this Project is before the town.")], size=10)
src(s, "Sources: MassGIS Property Tax Parcels (FY2026 assessor data; owner of record not yet updated after the May 2026 sale); Bisnow; Eagle-Tribune; Andover Planning Board minutes (Sep 14, 2021); Andover News (Sep 2026).")

# =============== 13. BUILDING SUMMARY & ENTITLEMENT ===============
s = page("Building Summary & Entitlement", 13)
head(s, 0.5, 1.15, 4.5, "Project Summary")
bullets(s, 0.5, 1.45, 4.6, 3.6, [
    ("Buildings.", "Three elevator-served buildings (two five-story, one four-story) totaling 334 units: 201 one-bedroom, 110 two-bedroom and 23 three-bedroom."),
    ("Affordability.", "84 units (25%) restricted to households at or below 80% AMI under a Local Initiative Program comprehensive permit. 250 units at market rents."),
    ("Parking.", "515 spaces (1.54 per unit). Garage vs. surface split [TBD]."),
    ("Amenities.", "Fitness center and coworking space per public presentations. Full amenity program, construction type and unit sizes [TBD]."),
    ("Commercial.", "~2 acres reserved for future commercial / retail development.")], size=10)
head(s, 5.4, 1.15, 5, "Entitlement Timeline")
tl = [("2021–2023", "Prior owner permits a 224,500 SF cGMP pharmaceutical facility (approved 2023; did not advance)."),
      ("Apr 14, 2026", "Select Board overview: multifamily as phase one of a multi-phase Minuteman Park redevelopment."),
      ("Apr 21, 2026", "Conservation Commission denies a procedural amendment to the prior Order of Conditions (timing of a conservation restriction). It does not rule on the housing project."),
      ("May 2026", "Hanover acquires the Site for $2.5M."),
      ("Sep 15, 2026", "LIP first reading before the Select Board. No vote; board to submit questions."),
      ("Oct 26, 2026", "Second reading targeted (expected)."),
      ("Next", "State (EOHLC) project eligibility, then a Zoning Board of Appeals comprehensive permit hearing.")]
y = 1.48
for d, e in tl:
    rect(s, 5.4, y + 0.05, 0.12, 0.12, ORANGE)
    tb(s, 5.62, y - 0.02, 1.25, 0.3, d, 9, NAVY, SANS, True)
    tb(s, 6.9, y - 0.02, 3.6, 0.6, e, 9.5, TXT, SERIF)
    y += 0.62
head(s, 0.5, 5.3, 4.6, "Issues Raised by the Town")
bullets(s, 0.5, 5.58, 4.6, 1.9, ["Traffic on River Road and nearby interchanges", "Updated sewer-capacity analysis requested by DPW",
                                 "Archaeological review of the former Shattuck Farm", "Conversion of commercial and industrial land to housing"], size=9.5, gap=2)
rect(s, 5.4, 5.95, 5.1, 1.45, LGRAY)
tb(s, 5.55, 6.02, 4.8, 1.35, [[("Tax impact. ", {"bold": True, "color": NAVY}),
     "Hanover estimates the Project would generate ~$1.5M a year in property taxes, about 10x the ~$150K the vacant parcel pays today. "
     "It also estimates 33–50 school-age children; town counsel noted enrollment is generally not a ZBA consideration."]], 9.5, TXT, SERIF)
src(s, "Sources: Andover News & Andover Ledger (Sep 16, 2026); Eagle-Tribune; Bisnow; Andover Conservation Commission minutes (Apr 21, 2026); Andover Planning Board minutes (Sep 14, 2021). "
       "The Sep 15 date is per the Ledger; other outlets say \u201cMonday.\u201d")

# =============== 14. SAMPLE RENDERINGS — INTERIOR ===============
s = page("Sample Renderings: Interior Amenities", 14)
pic(s, "colony_place_clubhouse.jpg", 0.5, 1.15, 6.3, 6.2)
pic(s, "colony_place_fitness.jpg", 6.95, 1.15, 3.55, 3.05)
pic(s, "weymouth_fitness.jpg", 6.95, 4.3, 3.55, 3.05)
tb(s, 0.5, 7.42, 10, 0.25, "Note: Photos of similar Hanover projects (Hanover Colony Place, Plymouth MA; Hanover Weymouth, Weymouth MA); subject project will vary.", 8.5, SLATE, SERIF, italic=True)

# =============== 15. SAMPLE RENDERINGS — EXTERIOR ===============
s = page("Sample Renderings: Exterior Amenities", 15)
pic(s, "colony_place_pool_dusk.jpg", 0.5, 1.15, 10.0, 3.9, 0.55)
pic(s, "weymouth_pool.jpg", 0.5, 5.15, 3.25, 2.2)
pic(s, "colony_place_firepit_courtyard.jpg", 3.875, 5.15, 3.25, 2.2)
pic(s, "stoneham_courtyard_dusk.jpg", 7.25, 5.15, 3.25, 2.2)
tb(s, 0.5, 7.42, 10, 0.25, "Note: Photos of similar Hanover projects (Hanover Colony Place, Hanover Weymouth, Hanover Stoneham); subject project will vary.", 8.5, SLATE, SERIF, italic=True)

# =============== 16. DEMOGRAPHICS ===============
s = page("Demographics", 16)
rows = [["", "ZIP 01810", "Essex County", "Boston MSA"],
        ["Population", "35,648", "823,938", "5,025,517"], ["Households", "13,298", "314,816", "1,970,240"],
        ["5-Year Population Growth", "–0.5%", "+5.1%", "+4.0%"], ["Median Household Income", "$180,672", "$109,581", "$124,803"],
        ["Average Household Income", "$211,050", "$174,660", "$196,182"], ["Renter Median Household Income", "$95,223", "$58,138", "$74,832"],
        ["Bachelor's Degree or Higher", "96.3%", "57.9%", "68.1%"], ["Renter-Occupied Housing", "20.1%", "37.6%", "39.0%"],
        ["Median Home Value", "$1,013,410", "$714,736", "$728,727"], ["5-Year Home Value Change", "+38.8%", "+33.1%", "+31.3%"],
        ["Median Net Worth", "$1M – $2M", "—", "$500K – $750K"]]
table(s, 0.5, 1.2, 6.0, rows, [2.4, 1.2, 1.2, 1.2], rh=0.37, size=9)
head(s, 6.85, 1.15, 3.7, "Household Income — ZIP 01810")
cd = CategoryChartData(); cd.categories = ["< $50K", "$50K–$100K", "$100K–$200K", "> $200K"]; cd.add_series("Share of households", (0.13, 0.14, 0.30, 0.44))
ch = s.shapes.add_chart(XL_CHART_TYPE.COLUMN_CLUSTERED, I(6.85), I(1.45), I(3.65), I(2.9), cd).chart
ch.has_legend = False; pl = ch.plots[0]; pl.gap_width = 60; pl.has_data_labels = True
dl = pl.data_labels; dl.number_format = '0%'; dl.number_format_is_linked = False; dl.font.size = Pt(9); dl.font.name = SANS; dl.position = XL_LABEL_POSITION.OUTSIDE_END
pl.series[0].format.fill.solid(); pl.series[0].format.fill.fore_color.rgb = NAVY
ch.value_axis.visible = False; ch.value_axis.has_major_gridlines = False; ch.value_axis.maximum_scale = 0.55
ch.category_axis.tick_labels.font.size = Pt(8); ch.category_axis.tick_labels.font.name = SANS
rect(s, 6.85, 4.55, 3.65, 2.85, NAVY)
tb(s, 7.05, 4.68, 3.3, 0.3, "Migration Cohort (Aug 2024 – Jul 2026)", 9, GRAY, SANS, True, caps=True)
tb(s, 7.05, 5.05, 3.3, 0.5, "$297,090", 24, WHITE, SERIF); tb(s, 7.05, 5.52, 3.3, 0.3, "Median HHI of households moving in", 8, GRAY, SANS, True, caps=True)
tb(s, 7.05, 5.9, 3.3, 0.5, "$251,631", 24, WHITE, SERIF); tb(s, 7.05, 6.37, 3.3, 0.3, "Median HHI of households moving out", 8, GRAY, SANS, True, caps=True)
tb(s, 7.05, 6.75, 3.3, 0.55, "+$45K in-mover advantage, vs. –$19K for the Boston MSA. In-movers are younger (41 vs. 52).", 9, WHITE, SERIF)
tb(s, 0.5, 5.75, 6.0, 1.6, [[("Takeaway. ", {"bold": True, "color": NAVY}),
    "Andover's resident base is among the wealthiest in Greater Boston, and the households arriving out-earn those leaving. "
    "Renters are only 20% of households, so new rental supply serves younger, high-earning newcomers who are still building wealth. "
    "Population is roughly flat, so the case rests on the quality of new residents rather than how many are arriving."]], 10, TXT, SERIF)
src(s, "Sources: RealAI SuperCensus (Aug 15, 2026), US Census ACS 2024 (population, households, tenure), municipal deed/AVM data (home values, Aug 2026). Income distribution: ACS 2024 5-year "
       "via Census Reporter (shares sum to 101% from rounding). Bachelor's share is SuperCensus adults. Migration = tracked households (1,163 in / 1,284 out).", y=7.5, h=0.45)

# =============== 17. TARGET INCOME / AFFORDABILITY ===============
s = page("Demographics", 17, "Target Income  |  Affordable Program")
rows = [["Unit Type", "Comp In-Place Rent", "Target Income (3.0x Rent)", "vs. 01810 Renter Median", "80% AMI Income Limit", "80% AMI Max Gross Rent", "Affordable Discount"],
        ["1BR", "$2,546", "$91,639", "–4%", "$80,125", "$2,003", "–21%"],
        ["2BR", "$3,088", "$111,166", "+17%", "$96,150", "$2,404", "–22%"],
        ["3BR", "$3,961", "$142,608", "+50%", "$111,075", "$2,777", "–30%"]]
table(s, 0.5, 1.45, 10.0, rows, [1.0, 1.4, 1.6, 1.6, 1.5, 1.5, 1.4], rh=0.42, size=9)
cd = CategoryChartData(); cd.categories = ["1BR target", "2BR target", "3BR target", "01810 renter median", "01810 all-HH median", "In-mover median"]
cd.add_series("Annual household income", (91639, 111166, 142608, 95223, 180672, 297090))
gf = s.shapes.add_chart(XL_CHART_TYPE.BAR_CLUSTERED, I(0.5), I(3.4), I(6.3), I(3.95), cd); ch = gf.chart
ch.has_legend = False; pl = ch.plots[0]; pl.gap_width = 45; pl.has_data_labels = True
pl.data_labels.number_format = '"$"#,##0'; pl.data_labels.number_format_is_linked = False; pl.data_labels.font.size = Pt(8.5); pl.data_labels.font.name = SANS
ser = pl.series[0]; ser.format.fill.solid(); ser.format.fill.fore_color.rgb = NAVY
for idx in (3, 4, 5):
    pt = ser.points[idx]; pt.format.fill.solid(); pt.format.fill.fore_color.rgb = ORANGE
ch.value_axis.visible = False; ch.value_axis.has_major_gridlines = False; ch.value_axis.maximum_scale = 340000
ch.category_axis.tick_labels.font.size = Pt(8.5); ch.category_axis.tick_labels.font.name = SANS; ch.category_axis.reverse_order = True
head(s, 7.1, 3.4, 3.4, "What It Means")
bullets(s, 7.1, 3.7, 3.4, 3.7, [
    ("1BR rents fit the existing renter pool.", "The 1BR target income ($91.6K) is just below the 01810 renter median ($95.2K)."),
    ("2BR and 3BR units depend on high earners.", "They need $111K–$143K incomes, well below the ZIP's all-household median ($180.7K) and the in-mover median ($297K)."),
    ("The affordable units price 21–30% below market.", "That gives them a deep waiting-list pool. Rents are capped by HUD's 80% AMI limits, which rose for FY2026.")], size=9.5)
src(s, "Target income = 36x monthly rent (rent at 33% of income), the convention in Hanover's Sugarloaf book. Comp rents: RealAI Rent Index (Sep 26, 2026), six-comp simple average. 80% AMI: HUD FY2026, "
       "Lawrence MA-NH HMFA, 1.5 persons/bedroom, 30% of income, gross before utility allowance. Calculations: book_calcs.py.", y=7.5, h=0.45)

# =============== 18. SUPPLY PIPELINE ===============
s = page("Supply Pipeline", 18)
s.shapes.add_picture(A("maps", "pipeline_map.jpg"), I(0.5), I(1.15), I(5.6), I(3.61))
rows = [["#", "Project", "Units", "Status", "Est. Delivery", "Dist."],
        ["1", "Zero Prescott St (East Mill), N. Andover", "296", "Proposed 40B — ZBA review", "TBD", "~4 mi"],
        ["2", "946 Osgood St, N. Andover", "108", "Proposed LIP / 40B", "TBD", "~5 mi"],
        ["3", "Princeton Wilmington, Wilmington", "108", "Lease-up (from Aug 2026)", "2026", "~9 mi"],
        ["4", "MacLellan Oil site, Tewksbury", "24", "Approved; site clearing", "TBD", "~6 mi"],
        ["5", "Avalon North Andover", "221", "Delivered", "2022", "~4 mi"],
        ["6", "505 Sutton, N. Andover", "136", "Delivered", "2022", "~5 mi"]]
table(s, 6.3, 1.15, 4.2, rows, [0.25, 1.55, 0.45, 1.0, 0.55, 0.4], rh=0.5, size=7.5)
head(s, 0.5, 4.95, 10, "Pipeline Takeaways")
bullets(s, 0.5, 5.25, 10.0, 2.2, [
    ("No competing units under construction within 5 miles.", "The two nearest proposals (East Mill and 946 Osgood) are still in comprehensive-permit or town review, so delivery dates are unknown."),
    ("Andover itself is barely building.", "The town permitted 9 new homes in 2023 and 21 in 2024. North Andover's 2022 deliveries (357 units) were 95.0% and 86.8% occupied as of September 2026."),
    ("Zoning is opening up near transit.", "Andover's MBTA Multifamily Overlay (2024) and North Andover's sub-districts could add supply over time. No overlay projects have been confirmed yet.")], size=10)
src(s, "Sources: North Andover ZBA records & Patch (East Mill); Town of North Andover (946 Osgood, search listing); National Development / Princeton Properties; Tewksbury Town Crier; RealAI Datamart; "
       "Andover News (housing starts); Boston Foundation 2025 Housing Report Card. Distances are straight-line. Market-rate projects in Methuen, Lawrence and Lowell were not fully surveyed.", y=7.5, h=0.45)

# =============== 19. RENT COMPS SUMMARY ===============
s = page("Rent Comps", 19)
s.shapes.add_picture(A("maps", "rent_comps_map.jpg"), I(0.5), I(1.15), I(4.4), I(2.84))
rows = [["#", "Property", "Units", "Built", "Avg SF", "1BR", "2BR", "3BR", "$/SF", "Occ."],
        ["S", NAME + " (proposed)", "334", "TBD", "[TBD]", "[TBD]", "[TBD]", "[TBD]", "[TBD]", "—"],
        ["1", "The Slate at Andover", "225", "2016", "987", "$2,408", "$2,963", "$4,171", "$2.81", "95.6%"],
        ["2", "Berry Farms", "196", "2016", "960", "$2,515", "$2,992", "$3,736", "$2.98", "93.4%"],
        ["3", "The Point at Merrimack River*", "248", "2018", "986", "$2,562", "$3,156", "$3,886", "$2.94", "96.8%"],
        ["4", "Haven North Andover", "192", "2019", "1,070", "$2,512", "$3,176", "—", "$2.71", "95.3%"],
        ["5", "Avalon North Andover", "221", "2022", "980", "$2,476", "$2,993", "$4,051", "$2.95", "95.0%"],
        ["6", "505 Sutton", "136", "2022", "899", "$2,800", "$3,248", "—", "$3.26", "86.8%"],
        ["", "Comp Average", "1,218", "", "", "$2,546", "$3,088", "$3,961", "", "94.3%"],
        ["", "Andover Submarket", "", "", "", "$2,437", "$2,764", "$3,618", "$2.66", "93.2%"]]
table(s, 5.1, 1.15, 5.4, rows, [0.25, 1.65, 0.45, 0.42, 0.45, 0.48, 0.48, 0.48, 0.4, 0.34], rh=0.31, size=7, shade=(8, 9))
head(s, 0.5, 4.2, 10, "Rent Comp Takeaways")
bullets(s, 0.5, 4.5, 10.0, 2.8, [
    ("The premium is in larger units.", "Against submarket in-place rents, the comps run +11.7% on 2BRs and +9.5% on 3BRs, but only +4.5% on 1BRs. 1BRs make up 60% of the proposed mix."),
    ("Rents cluster tightly.", "Average in-place rents fall between $2,719 and $2,855 across all six comps. The unit-weighted average is $2,805, with 94.3% occupancy."),
    ("Where rents top out.", "505 Sutton (2022) gets the set's highest rent at $3.26/SF but is only 86.8% occupied."),
    ("Hanover's own comp.", "The Point at Merrimack River, Hanover's 2018 Andover 40B half a mile from the Site, is 96.8% occupied at $2.94/SF.")], size=10)
src(s, "* Developed by Hanover (2018); current ownership to be confirmed. In-place rent = average rent on leased units, RealAI Rent Index as of Sep 26, 2026. Comps 3–6 sit in the Datamart's adjacent "
       "\u201cLawrence\u201d submarket; under strict criteria only comps 1–2 qualify. 3BR average n = 4. Rent and $/SF for the Subject to come from Hanover's underwriting.", y=7.45, h=0.5)


# =============== 20–22. RENT COMP PROFILES ===============
def profile(s, x, title, addr, img, data, comments, credit):
    tb(s, x, 1.15, 4.9, 0.32, title, 15, NAVY, SERIF, sp=0.5)
    tb(s, x, 1.45, 4.9, 0.25, addr, 9, SLATE, SANS)
    if img:
        pic(s, img, x, 1.75, 4.9, 2.55)
        tb(s, x, 4.32, 4.9, 0.2, credit, 6.5, GRAY, SANS)
    else:
        rect(s, x, 1.75, 4.9, 2.55, LGRAY)
        tb(s, x, 2.85, 4.9, 0.4, "[Property photo — to be added]", 11, ORANGE, SERIF, align=PP_ALIGN.CENTER, italic=True)
    table(s, x, 4.55, 4.9, data, [1.15, 0.85, 0.95, 0.95, 1.0], rh=0.27, size=8)
    tb(s, x, 6.25, 4.9, 1.2, comments, 9, TXT, SERIF)


def comp_rows(units, yb, sf, rt, occ, b1, b2, b3, psf):
    return [["Building Data", "", "Bedroom", "In-Place", "$/SF"],
            ["Units", units, "1BR", b1[0], b1[1]], ["Year Built", yb, "2BR", b2[0], b2[1]], ["Avg SF", sf, "3BR", b3[0], b3[1]],
            ["Rent Type", rt, "Total", "", psf], ["Occupancy", occ, "", "", ""]]


s = page("Rent Comps", 20)
profile(s, 0.5, "The Point at Merrimack River", "30–40 Shattuck Rd, Andover, MA 01810  |  Developed by Hanover (2018)", "the_point_exterior_housingnavigator.jpg",
        comp_rows("248", "2018", "986", "Mkt & Aff.", "96.8%", ("$2,562", "$3.12"), ("$3,156", "$2.64"), ("$3,886", "$2.66"), "$2.94"),
        "Chapter 40B community 0.5 mi from the Site. Amenities include a 6,000 SF clubhouse, resort-style pool, 24-hour fitness, theater room and "
        "garage parking ($250/mo). October 2026 listings show 1BR asking rents from $2,593 and 2BR from $3,277 to $3,552.",
        "Photo: property listing via Housing Navigator MA")
profile(s, 5.6, "Haven North Andover", "1252 Osgood St, North Andover, MA 01845", "haven_north_andover_exterior.jpg",
        comp_rows("192", "2019", "1,070", "Market", "95.3%", ("$2,512", "$2.90"), ("$3,176", "$2.51"), ("—", "—"), "$2.71"),
        "Market-rate community on the Route 125 corridor. It has the set's largest average unit (1,070 SF) and the highest 2BR in-place rent "
        "outside 505 Sutton. No 3BR units are tracked.", "Photo: havennorthandover.com")
src(s, "Source: RealAI Rent Index (Sep 26, 2026); The Point amenities and asking rents per Housing Navigator / Zillow listings (Oct 2026).")

s = page("Rent Comps", 21)
profile(s, 0.5, "505 Sutton", "495 Sutton St, North Andover, MA 01845", "505_sutton_entrance.jpg",
        comp_rows("136", "2022", "899", "Market", "86.8%", ("$2,800", "$3.09"), ("$3,248", "$2.94"), ("—", "—"), "$3.26"),
        "MINCO's 2022 market-rate delivery beside North Andover's new senior center. It has the highest rent per SF in the set, but occupancy of 86.8% shows "
        "where pricing tops out for smaller units.", "Photo: 505sutton.com")
profile(s, 5.6, "Berry Farms", "4 Berry St, North Andover, MA 01845", "berry_farms_exterior.jpg",
        comp_rows("196", "2016", "960", "Market", "93.4%", ("$2,515", "$3.18"), ("$2,992", "$2.80"), ("$3,736", "$3.05"), "$2.98"),
        "Market-rate garden community in the Andover submarket. Its 1BR in-place rent of $3.18/SF is the highest of the strict two-property set.",
        "Photo: berryfarmsapartments.com")
src(s, "Source: RealAI Rent Index (Sep 26, 2026); Town of North Andover (505 Sutton context).")

s = page("Rent Comps", 22)
profile(s, 0.5, "Avalon North Andover", "88 High St, North Andover, MA 01845", None,
        comp_rows("221", "2022", "980", "Market", "95.0%", ("$2,476", "$3.17"), ("$2,993", "$2.58"), ("$4,051", "$2.72"), "$2.95"),
        "AvalonBay's 2023 Mill District delivery next to downtown North Andover. It is 95.0% occupied, with studios at $2,123 in-place.", "")
profile(s, 5.6, "The Slate at Andover", "50 Woodview Way, Andover, MA 01810", None,
        comp_rows("225", "2016", "987", "Mkt & Aff.", "95.6%", ("$2,408", "$2.94"), ("$2,963", "$2.58"), ("$4,171", "$2.76"), "$2.81"),
        "Mixed-income community in ZIP 01810. It has the comp set's highest 3BR in-place rent ($4,171), supporting family-sized units in Andover.", "")
src(s, "Source: RealAI Rent Index (Sep 26, 2026). Photos for these two properties could not be downloaded (site access blocked) and should be added manually.")

# =============== 23. LEASING MOMENTUM ===============
s = page("Leasing Momentum", 23, "ZIP 01810 Multifamily")
cd = CategoryChartData()
cd.categories = ["Sep-25", "Oct-25", "Nov-25", "Dec-25", "Jan-26", "Feb-26", "Mar-26", "Apr-26", "May-26", "Jun-26", "Jul-26", "Aug-26"]
cd.add_series("Median asking rent", (2732, 2750, 2620, 2590, 2565, 2628, 2640.5, 2640, 2800, 2824, 2700, 2621))
cd.add_series("Median in-place rent", (2550, 2545, 2544, 2540, 2540.5, 2541, 2545, 2530, 2541, 2545, 2550, 2559))
ch = s.shapes.add_chart(XL_CHART_TYPE.LINE_MARKERS, I(0.5), I(1.5), I(6.4), I(4.6), cd).chart
ch.has_legend = True; ch.legend.position = XL_LEGEND_POSITION.BOTTOM; ch.legend.include_in_layout = False; ch.legend.font.size = Pt(9); ch.legend.font.name = SANS
va = ch.value_axis; va.minimum_scale, va.maximum_scale, va.major_unit = 2400, 2900, 100
va.tick_labels.font.size = Pt(8); va.tick_labels.number_format = '"$"#,##0'; va.tick_labels.number_format_is_linked = False
va.major_gridlines.format.line.color.rgb = LGRAY; ch.category_axis.tick_labels.font.size = Pt(8)
for ser, col in zip(ch.plots[0].series, [NAVY, ORANGE]):
    ser.format.line.color.rgb = col; ser.format.line.width = Pt(2.25); ser.smooth = False
    ser.marker.format.fill.solid(); ser.marker.format.fill.fore_color.rgb = col; ser.marker.format.line.color.rgb = col
kp = [("$2,820", "Median asking rent", "+0.8% YoY; 10.2% above $2,559 in-place"), ("96.7%", "Occupancy", "97.1% a year ago"),
      ("+3.6%", "New-lease tradeout", "Mar–Aug 2026, 131 leases; latest 30 days +5.85%"), ("59 days", "Median days on market", "Leases signed past 30 days (36)")]
for i, (v, l, d) in enumerate(kp):
    y = 1.5 + i * 1.18
    rect(s, 7.2, y, 3.3, 1.05, NAVY if i % 2 == 0 else SLATE)
    tb(s, 7.35, y + 0.08, 3.0, 0.45, v, 22, WHITE, SERIF)
    tb(s, 7.35, y + 0.5, 3.0, 0.25, l, 8.5, WHITE, SANS, True, caps=True)
    tb(s, 7.35, y + 0.72, 3.0, 0.3, d, 8, GRAY, SANS)
tb(s, 0.5, 6.3, 10, 1.0, [[("Read-through. ", {"bold": True, "color": NAVY}),
    "Occupancy is tight and new-lease pricing has turned up, from –0.2% over Sep 2025–Feb 2026 to +3.6% over Mar–Aug 2026. Asking rents are seasonal, "
    "bottoming in January ($2,565) and peaking in June ($2,824). A two-month median days on market says renters are paying the 10% asking premium "
    "gradually rather than competing for units."]], 10, TXT, SERIF)
src(s, "Source: RealAI Rent Index (snapshot Sep 26, 2026; monthly series through Aug 2026), ~1,213 tracked units in ZIP 01810. Tradeout is lease-weighted; monthly samples are small (10–28 leases).")

# =============== 24. SALE COMPS ===============
s = page("Asset Sale Comps", 24)
s.shapes.add_picture(A("maps", "sale_comps_map.jpg"), I(0.5), I(1.15), I(4.2), I(2.71))
rows = [["#", "Property", "Built", "Units", "Sale Date", "Price", "Per Unit", "Cap Rate", "Buyer"],
        ["S", NAME, "TBD", "334", "—", "[TBD]", "[TBD]", "[TBD]", "(Project cost)"],
        ["1", "Residences at Crosspoint, Lowell", "2020", "240", "Nov-24", "$85.1M", "$354,479", "5.0%", "n/a"],
        ["2", "Villas at Old Concord, Billerica", "2004", "324", "Sep-24", "$114.5M", "$353,395", "n/a", "TruAmerica"],
        ["3", "Point at Merrimack Valley, Methuen", "2021", "156", "Mar-24", "$58.1M", "$372,436", "n/a", "n/a"],
        ["4", "Woods at Merrimack, Methuen", "2019", "140", "Nov-25", "$37.5M", "$268,121", "n/a", "n/a"],
        ["5", "Legacy Park, Lawrence", "2009", "104", "Jun-26", "$30.2M", "$290,385", "n/a", "n/a"],
        ["6", "The Point at Woburn, Woburn", "2022", "289", "Jan-26", "$130.7M", "$452,166", "n/a", "Pantzer"]]
table(s, 4.9, 1.15, 5.6, rows, [0.22, 1.7, 0.4, 0.4, 0.5, 0.58, 0.65, 0.5, 0.65], rh=0.34, size=7)
head(s, 0.5, 4.1, 10, "Sale Comp Takeaways")
bullets(s, 0.5, 4.4, 10.0, 2.9, [
    ("Newer product trades at $354K–$452K per unit.", "The three 2020+ vintages average $393K/unit, led by The Point at Woburn ($452K, Jan 2026)."),
    ("Merrimack Valley trades are cheaper.", "Methuen and Lawrence sales of $268K–$372K/unit reflect older or less affluent locations than Andover."),
    ("Few cap rates are disclosed.", "Only Crosspoint reports one (5.0%, CoStar). Wronka puts the Boston market cap rate near 5.1% (Aug 2025). Exit cap and project cost per unit [TBD] from Hanover's model.")], size=10)
src(s, "Sources: RealAI Datamart (deed-recorded latest sale); YieldPro (Villas at Old Concord); MarketScreener (Point at Woburn buyer); Wronka Boston Multifamily Capital Market Report (Aug 2025). "
       "Woods at Merrimack $/unit appears low and should be verified. n/a = not reported.", y=7.45, h=0.5)

# =============== 25. SPONSOR TRACK RECORD ===============
s = page("Sponsor Track Record", 25)
pic(s, "the_point_exterior_redfin.jpg", 0.5, 1.15, 3.6, 2.45)
tb(s, 0.5, 3.62, 3.6, 0.2, "The Point at Merrimack River. Photo: listing via Redfin (confirm rights)", 6.5, GRAY, SANS)
head(s, 4.35, 1.15, 6, "Hanover in Andover — The Point at Merrimack River")
bullets(s, 4.35, 1.45, 6.15, 2.3, [
    "248-unit Chapter 40B community at 30–40 Shattuck Rd, developed by Hanover in 2018, about half a mile from the Site.",
    "96.8% occupied at $2.94/SF average in-place rent (Sep 2026), the highest occupancy in the six-property comp set.",
    "Gives Hanover direct operating history with Andover's permitting boards and renter base through the same 40B structure now proposed.",
    "Current ownership and management to be confirmed: the fee owner of record is Andover Apartments LLC, and Pantzer manages on site."], size=10)
head(s, 0.5, 4.0, 10, "Hanover Atlanta — Performance vs. Submarket (Sep 2026)")
rows = [["Metric", "Hanover Edgewood", "Submarket¹", "Premium", "Hanover Midtown", "Midtown Submarket", "Premium"],
        ["Units / Year Built", "422 / 2022", "—", "—", "421 / 2023", "—", "—"],
        ["Average In-Place Rent", "$1,990", "$1,744", "+14.1%", "$2,740", "$2,496", "+9.8%"],
        ["2BR In-Place Rent", "$2,339", "$1,858", "+25.9%", "$3,396", "$3,022", "+12.4%"],
        ["Average $/SF", "$2.20", "$1.97", "+11.7%", "$2.86", "$2.81", "+1.8%"],
        ["Median Days on Market²", "35", "63", "28 days faster", "68", "82", "14 days faster"],
        ["Occupancy", "89.8%", "94.0%", "–4.1 pts", "80.5%", "85.3%", "–4.8 pts"]]
table(s, 0.5, 4.3, 10.0, rows, [2.0, 1.4, 1.3, 1.25, 1.4, 1.4, 1.25], rh=0.36, size=8.5, shade=(5,))
src(s, "¹ Datamart-assigned submarket (Chandler-McAfee/West Belvedere Park). Against its own ZIP (30307), Edgewood rents 2.6% below average.  ² Leases signed in the past 30 days (Edgewood 23; Midtown 46). "
       "Both properties trail submarket occupancy. Source: RealAI Rent Index (Sep 26, 2026); Eagle-Tribune; Andover Ledger; Town of Andover.", y=7.45, h=0.5)

# =============== 26. KEY CONTACTS ===============
s = page("Key Contacts", 26)
cts = [("Brandt Bowden", "Chief Executive Officer", "713 580 1203", "bbowden@hanoverco.com"),
       ("Theresa Blades", "Chief Investment Officer", "713 580 1150", "tblades@hanoverco.com"),
       ("Emily Coty", "Capital Markets Managing Director", "713 580 1027", "ecoty@hanoverco.com"),
       ("Drew Hunnicutt", "Capital Markets Director", "713 580 1376", "dhunnicutt@hanoverco.com"),
       ("David Hall", "Regional Development Partner", "[Phone — TBD]", "[Email — TBD]"),
       ("Steve Dazzo", "[Title — TBD]", "[Phone — TBD]", "[Email — TBD]"),
       ("Ethan Kao", "Capital Markets Analyst", "713 580 1358", "ekao@hanoverco.com")]
for i, (n, t, p, e) in enumerate(cts):
    x = 0.5 + (i % 4) * 2.55; y = 1.4 + (i // 4) * 2.2
    tb(s, x, y, 2.4, 0.35, n, 15, NAVY, SERIF)
    tb(s, x, y + 0.38, 2.4, 0.5, t, 9, SLATE, SANS, True, caps=True)
    tb(s, x, y + 0.9, 2.4, 0.25, "Direct: " + p, 9, ORANGE if "TBD" in p else TXT, SERIF)
    tb(s, x, y + 1.15, 2.4, 0.25, "Email: " + e, 9, ORANGE if "TBD" in e else TXT, SERIF)
hline(s, 0.5, 6.05, 10, GRAY, 0.5)
tb(s, 0.5, 6.15, 10, 0.3, "LEGAL NOTICE", 9, NAVY, SERIF, True, align=PP_ALIGN.CENTER, sp=2)
tb(s, 0.5, 6.45, 10, 0.9, "The information contained in this brochure is for informational purposes only and all market analysis and projections are merely examples. "
   "The examples are not intended to represent or guarantee that the results will be achieved. You understand that The Hanover Company is not warranting the accuracy, "
   "completeness, merchantability or fitness of any information contained in this brochure. You recognize any business endeavor has inherent risk for loss of capital "
   "and that there are no guarantees on your investment.", 8.5, TXT, SERIF, align=PP_ALIGN.CENTER)
tb(s, 0.5, 7.4, 10, 0.25, "©2026 Hanover Company", 8.5, GRAY, SERIF, align=PP_ALIGN.CENTER)
src(s, "Contacts carried over from the Hanover Sugarloaf book (Apr 2025); confirm the Minuteman Park deal team. David Hall and Steve Dazzo are named in Andover press coverage.", y=7.7)

out = os.path.join(ROOT, "Hanover_Minuteman_Park_Investment_Book.pptx")
prs.save(out); print("saved", out, len(prs.slides), "slides")

Hanover Minuteman Park – Investment Book.pptxAI Created

PowerPoint·22.6 MB·Oct 5, 2026

Hanover Minuteman Park – Investment Book (PDF export).pdfAI Created

PDF·12.0 MB·Oct 5, 2026

The 26-page Hanover Minuteman Park investment book is built in the Sugarloaf format, as an editable PowerPoint plus a PDF export. It follows the Sugarloaf structure: navy cover band, serif headings, and the same footer, page order and photo conventions. The financial, cost and operating pages are placeholders, with every input marked [TBD] in orange .

The deck presents the program you confirmed: 250 market-rate units plus 84 affordable units at 80% of area median income under the Local Initiative Program.

Visuals I created or sourced

Aerial maps: a site map with 14 points of interest, plus rent-comp, supply-pipeline (with 3- and 5-mile rings) and sale-comp maps .
Site close-up: the actual parcel boundary from state assessor records (Map 165, Lot 4, 20.24 acres, industrial zoning) .
Photos:
- Similar Hanover communities (Stoneham, Colony Place, Weymouth) on the cover and the two sample-renderings pages, labeled "similar project," as in Sugarloaf.
- Freely licensed Andover photos for the points of interest, with credits on the page.
- Comp photos where the property websites allowed downloads.

What the area research adds

Unit mix: 201 one-bedroom, 110 two-bedroom and 23 three-bedroom units, according to the Eagle-Tribune .
Entitlement timeline: from the 2023 approval of a pharmaceutical plant on the site through the Select Board's Oct. 26 second reading. The April Conservation Commission denial was procedural, not a ruling on the housing .
Target income (rent × 36, Sugarloaf's convention) :
- Comp one-bedroom rents need about $91.6K a year, just under the ZIP's $95.2K renter median.
- Two- and three-bedroom units need $111K–$143K.
Affordable units: FY2026 rent caps are 21–30% below comp rents .
Supply: nothing within 5 miles is under construction. The nearest competitors are two North Andover proposals of 296 and 108 units, both still in town review .
Sale comps: $268K–$452K per unit across six trades from 2024–2026 .

Gaps to close before it goes out

Your logo file is no longer accessible, so the book uses the "HANOVER" wordmark from the Sugarloaf book. Please re-upload the logo if you'd rather use it.
Two comps have no photo. Avalon North Andover and The Slate blocked downloads, so they show a gray placeholder box.
Some photos need rights cleared. The Point at Merrimack River photos come from third-party listings (Housing Navigator, Redfin).
Confirm the track record. The Point's owner of record is "Andover Apartments LLC" and it is managed by Pantzer. I found no record of a sale by Hanover, so the slide says "developed by Hanover" only.
Confirm the contacts. They are carried over from Sugarloaf. David Hall's and Steve Dazzo's phone numbers, emails and Dazzo's title are [TBD].
The PDF shows substitute fonts. The book is set in Garamond and Century Gothic, which display correctly in PowerPoint.

A few questions:

Can you share any underwriting outputs (rents, costs, return on cost, IRR) to fill pages 6–9? Partial numbers are fine.
Do you have a site plan or renderings for 300 Minuteman? Sugarloaf had a preliminary site plan page, which this book currently lacks.
Should the Atlanta comparison stay on the track-record page, or should the page be Andover-only now that The Point is the lead story?

Files are in the library:

Hanover Minuteman Park – Investment Book.pptx
Hanover Minuteman Park – Investment Book (PDF export).pdf