INTERNAL DRAFT, not yet client-ready. For Jon's review. Data-driven, plain presentation only; no reference to any prior conversation with Benjamin should appear anywhere in this document.
VERSION 2 (Option 1), for side-by-side comparison only. The same-store trend below is trimmed to June through August 2026, the only readings unaffected by the analytics.unit_nights data-floor issue (Relay #13946). The Dec 2025 to May 2026 "before Pacer" argument and the regional trend history before June are removed entirely, not approximated.
Hello Holt · Performance & Opportunity Review
Where things stand, and what we're fixing
A plain look at same-store performance since Pacer took over pricing, the specific factors currently holding bookings back, and the concrete work already underway to address each one.
Same-store performance, month by month
Hello Holt currently manages 154 active units. Every point below compares the same units with at least 24 months of booking history as of that reading (57 units for June and July, 60 for August), the same cohort discipline used throughout this document. Trend begins in June 2026, when Pacer took over day-to-day pricing on this book.
Occupancy YoY ADR YoY RevPAR YoY
Source: Pacer (KeyData, in-service-both-spans cohort). Each point is a trailing-12-month reading ending that month, using only units with a service-start date on or before the start of that reading's comparison year that are still active today. RevPAR = adjusted occupancy × fee-inclusive ADR.
RevPAR YoY, trailing 12 months
-9.0%
Ranged -9.0% to -9.7% across the three readings since Pacer took over pricing
Pacer (KeyData, same-store cohort, 60 units)
Occupancy YoY, trailing 12 months
-12.3%
The metric to watch: down further in each of the three readings, from -9.7% in June
Pacer (KeyData, same-store cohort)
ADR YoY, trailing 12 months
+1.0%
Improved in each reading, from -3.1% in June to +1.0% in August
Pacer (KeyData, same-store cohort)
Even now, near-term bookings take time to turn over: 68% of June's revenue, 38% of July's, and 26% of August's had already been booked before Pacer took over pricing on June 3.
Two markets, two different stories
Bucharest/Brasov ("Romania") and San Miguel de Allende/Oaxaca ("Mexico") make up the same-store book. Splitting the trend line above by market shows they are not behaving the same way, and are not the same kind of property.
Romania same-store RevPAR YoY, range
-7.6% to -8.4%
23-25 units across the three readings since June; mostly small city apartments
Pacer (KeyData, in-service-both-spans)
Mexico same-store RevPAR YoY, range
-11.0% to -12.5%
32-33 units across the three readings since June; a mix of studios and large villas
Pacer (KeyData, in-service-both-spans)
Mexico's mature same-store units are running meaningfully worse than Romania's (double digits vs. high single digits). This is not explained by an external market event. See the Puerto Vallarta section below, which is a separate, newer set of units entirely. It's an open question, and we don't have a cause for it yet.
Romania
SS units
SS YoY
June 2026
23
-7.6%
July 2026
23
-7.7%
August 2026
25
-8.4%
Source: Pacer (KeyData, in-service-both-spans cohort), verified against prod 2026-08-26.
Mexico
SS units
SS YoY
June 2026
32
-12.3%
July 2026
32
-12.5%
August 2026
33
-11.0%
Source: Pacer (KeyData, in-service-both-spans cohort), verified against prod 2026-08-26.
Why Mexico swings harder: across the full current portfolio in these two markets (not just the same-store subset above), Romania's 90 units are almost entirely small city apartments (54% are 1-2BR, nothing larger than 4BR). Mexico's 45 units split between small studios and a real cluster of large 4-8BR villas, a fundamentally different, higher-variance product. A handful of villas moving in or out of a cohort swings the blended number far more than the same change would in Romania's more uniform apartment stock.
A different question: how much summer revenue was already on the books at the same point each year. Confirmed reservation revenue for June-August stays, booked on or before June 3 of that year, same same-store units as above.
Romania was ahead of its own prior-year pace in every single month. The deficit is concentrated entirely in Mexico, and specifically June: Mexico's June position was $31,980 behind the year before, more than the rest of the summer's movement combined. July ran modestly behind; August ran ahead. This is a real, dated data point, not a complete explanation for Mexico's weaker same-store RevPAR shown above, but it is the first specific, quantified lead on where that gap originates.
Puerto Vallarta: a documented market shock, separate from the above
Puerto Vallarta is not part of the same-store analysis above. None of its 7 units have the 24 months of history required. It's a newer, smaller part of the book, and it has a clear, independently-verifiable story of its own.
Timeline: a cartel-related shootout in the Puerto Vallarta area in mid-February 2026, followed by three murders (May 10, 15, and 21) that drew national press coverage as a possible serial killer around May 26-27.
Hello Holt's Puerto Vallarta units, occupancy
85% → 63.6% → 55.7%
Jan/Feb 2026 vs. March vs. April, a two-month decline beginning the month after the February incident
Pacer Portal reservation data, Hello Holt's 7 PV units
Market-wide occupancy, past year
-9.7%
Entire Puerto Vallarta market, not just our units, independently confirmed via AirDNA
AirDNA, Puerto Vallarta market
Zona Hotelera booking lead time
-11.4%
Guests booking much closer to arrival, a classic hesitancy signal, in HS13011's own benchmark submarket
AirDNA, Zona Hotelera submarket
Hello Holt's Puerto Vallarta units recovered through the summer (63.8% June, 72.8% July, 70.5% August), tracking the same shape as the broader market. This is a real, documented external event with independent third-party confirmation. It is not a reason our own units gave, but a market condition every operator in the area faced.
Guest experience
Review scores across the portfolio, same period.
Portfolio guest rating
4.85
1,976 reviews across 85 rated units, unchanged from seven weeks ago
Pacer (Airbnb review export, 2026-08-26)
Guest Favorite units
20
Up from 0 seven weeks ago, as Airbnb expanded the badge across the book
Pacer (Airbnb review export)
Units with a real score decline
4 of 85
80 units flat, 1 improved over the same period
Pacer (Airbnb review export)
Cleanliness (4.89), Communication (4.88), and Location (4.85) are all strong. Check-in (4.77) and Value (4.76) run lowest, and Check-in is the single most common weak point across the book. Two established listings sit below portfolio average and are worth direct attention: HS13011 (Puerto Vallarta, 4.62 across 26 reviews) and CANT12 (Spain, 4.71 across 21 reviews).
Minimum-rate floors: a concrete, fixable lever
Avon's minrate scan has covered 48 of 149 units so far. It found specific units where the price floor is set above what confirmed demand has actually been paying, quietly blocking bookings that would otherwise clear.
Recoverable, next 90 days
Up to $9,184
9 units flagged with a floor set too high for their market and bedroom segment
Pacer (Avon Minrate Scan, 48 of 149 units)
Already-cleared bookings below floor
$118,938
43 units took bookings below their own stated floor in the trailing 12 months, rent that cleared, but a floor that wasn't enforced
Pacer (Avon Minrate Scan)
Pacer upside at 25% commission
+$958 to +$2,296
From fixing the 9 flagged floors alone, conservative to aggressive
Pacer (Avon Minrate Scan)
Unit
Market
Current floor
Recommended floor
90-day recovery
BREZ21A-VC (Bucharest, 1BR)
Bucharest
$52
$41-49
+$160 to +$656
PV2E (San Miguel, 2BR)
Mexico Urban
$123
$96-112
+$393 to +$1,131
HS13011 (Puerto Vallarta, 2BR)
Zona Hotelera
$145
$126-143
+$197 to +$1,319
VC206-PQ (Bucharest, 1BR)
Bucharest
$115
$41-49
+$2,129 to +$2,625
MONT14-M3 / M4 (San Miguel, studios)
Mexico Urban
$47
$33-41
+$59 to +$135 each
Source: Pacer (Avon Minrate Scan, Hello Holt, 48 of 149 units scanned as of 2026-08-26). Conservative = highest floor where 60% of trailing-12-month comparable bookings would still clear. Aggressive = highest floor where 80% still clear. This is a live, ongoing scan; 101 units have not yet been scored.
Listing content: photo captions not yet applied
A second concrete, no-cost lever available to this portfolio right now.
Pacer's AI photo-captioning pass, which rewrites every listing photo caption for OTA search relevance and click-through, is live in production on three other portfolios but has not yet been run on Hello Holt. This is a straightforward, zero-cost addition. Recommend prioritizing Hello Holt for the next captioning cohort alongside the minrate fixes above. We don't yet have a Hello Holt-specific lift number to quote; that will come once it's run here.
Forward booking pace, next 65 days
Bookings made so far for stays through October 30, compared to what was on the books at the same point last year.
Forward RevPAR, same-store
+43.4%
$65.39 booked so far vs. $45.60 at this point last year
$232.93 booked so far vs. $187.38 at this point last year
Pacer (KeyData forward pace)
Forward hold nights: 581 nights (+232% YoY, up from 175)
Occupancy above is simply booked nights divided by available nights; hold nights are a separate figure and don't change that ratio. What this number does say: 581 nights of inventory are being pulled off the market for owner use or maintenance over this window, up from 175 last year, nights guests can't book into no matter how strong demand is. Worth confirming this reflects a real increase in owner activity, not a change in how holds are being recorded, before using it externally.
Greenshoots: verified reservations
Three individual reservations, one per market, each checked against its matched year-ago comparable.
Three specific, real reservations, one from each major market in the portfolio. Real, verified proof that when demand is present and pricing is right, this team captures it.
What's already in motion
Pricing & availability
Far-out pricing report for Bucharest, the largest single market in the portfolio (86 units); Romania overall (Bucharest and Brasov combined) is 87
Property-by-property review of San Miguel de Allende and Oaxaca same-store units, since their underperformance isn't explained by a market-wide event
Minimum-rate floor corrections for the 9 flagged units above