RevParPro · Hotel Data Assembly · Pilot
One hotel, every data source assembled in one place, with the depth and freshness of each. Built as the pilot for a standard RevParPro onboarding intake. The pattern that matters: almost none of the useful material came from anything we built. It came from reports the brand already produces.
Eleven distinct sources feed this hotel. Four are ours, five come from IHG or CoStar and arrive by email, and two are read from the property systems. Depth is what determines whether a source can answer a seasonal question or only a right-now question.
| Source | What it answers | Depth held | Status |
|---|---|---|---|
| STR daily benchmark | Us vs comp set, every day, by day of week | 2024-01-01 to 2026-06-30, 912 days | STRONG |
| STR monthly benchmark | Monthly occupancy, rate, revenue index | through 2026-06 | STRONG |
| STR weekly benchmark | Rolling 28 day comp set, current | 1 week pulled 2026-08-13 | NEW |
| IHG DSS Property Detail | Our segment mix on the books vs last year | Aug 6 file, monthly forward | NOT INGESTED |
| IHG Production Analysis | Segment production, 3 year same-week compare | Aug 6 file | NOT INGESTED |
| IHG Booking Analysis | Reservation level, and segment by lead time | Aug 6 file, 6,749 room nights | NOT INGESTED |
| On the books pace | How full we are N days out | 2026-03-06 to 2026-08-13, 86 days | THIN |
| Competitor rate shop | Public rates, ours and theirs, daily | 2026-04-06 to 2026-08-13 | LIVE |
| Night audit | Occupancy, out of order, rate, revenue | daily, current | LIVE |
| Profit and loss, live QBO | Does the hotel cover its costs | 2022-01-01 to 2026-08-31, 13,645 txns | LIVE |
| IHG daily snapshot PDF | Brand forecast and strategy notes | daily email, link only | NEVER READ |
Depth figures are row counts and date ranges read from the RevParPro tables str_daily_data, str_monthly_data, booking_pace, rate_snapshots, night_audit_current and portfolio_pl_canonical on 2026-08-13, and from the file dates on the IHG workbooks attached to the Aug 6 "HOUUS Off Call Reports" email (Spark 337753).
The finding that matters for onboarding
The three richest sources on this list arrived by email on Aug 6, went unread for a week, and are still not ingested. IHG's revenue manager attached them and asked for feedback. Every hard question today was answered by those files or by the weekly benchmark report, not by anything RevParPro built.
The gap between what we can answer and what we cannot falls almost entirely along one line: how far back the source reaches.
| Question | Answerable today | Why |
|---|---|---|
| What does a normal October Tuesday finish at? | Yes | 2.5 years of daily benchmark data covers every month |
| Are we beating the comp set, and where? | Yes | Daily benchmark carries both sides, by day of week |
| Which segments did we lose? | Yes | IHG segment reports carry 3 year same-week compares |
| How far ahead does each segment book? | Yes | Booking Analysis holds lead time by segment |
| Did the market lose it or did we? | Partly | Benchmark shows total, not segment. Demand360 would close it |
| What does a normal October Tuesday look like 5 days out? | No | Pace data starts 2026-03-06 and has never seen October |
| Does a $6 discount change booking behavior? | No | Needs rate by reservation over time. One file is not a series |
| Does the hotel cover its costs this year? | Yes | Live QBO through the last closed month, May 2026 |
Date ranges as listed in section 01. The pace gap is structural: 86 snapshot days sit inside a 160 day calendar span, and pickup requires consecutive days, so the usable sample is smaller than 86.
This is the part that needs no new data and no waiting. Two and a half years of daily benchmark data, averaged by month, us against the comp set.
| Month | Days | Our occupancy | Comp set | Our rate | Comp rate | Revenue index |
|---|---|---|---|---|---|---|
| January | 93 | 54.3% | 52.2% | $91.70 | $85.97 | 110 |
| February | 85 | 64.0% | 65.2% | $96.67 | $91.56 | 104 |
| March | 93 | 76.0% | 69.9% | $105.94 | $97.17 | 119 |
| April | 90 | 67.0% | 63.2% | $99.11 | $94.51 | 112 |
| May | 93 | 73.1% | 60.4% | $100.50 | $97.07 | 126 |
| June | 90 | 72.6% | 60.8% | $106.88 | $102.44 | 126 |
| July | 62 | 73.4% | 61.5% | $113.61 | $99.65 | 139 |
| August | 62 | 66.2% | 61.4% | $103.14 | $95.81 | 117 |
| September | 60 | 60.6% | 56.4% | $97.76 | $96.10 | 109 |
| October | 62 | 64.6% | 64.9% | $98.80 | $96.28 | 102 |
| November | 60 | 61.0% | 59.6% | $96.01 | $91.31 | 107 |
| December | 62 | 54.7% | 52.8% | $92.80 | $84.83 | 111 |
RevParPro table str_daily_data, HOUUS comp set, 912 daily rows spanning 2024-01-01 to 2026-06-30. Revenue index is our revenue per room divided by the comp set's, times 100.
Three things fall out of this that nobody has been pricing against. July is our strongest month against the market at index 139, and it is also our highest rate month at $113.61. October is the one month we lose, index 102, and the only month where the comp set runs fuller than we do, 64.9% against our 64.6%. And the current month, August, normally runs 66.2% occupancy. The last four weeks have run 61.0%.
The segment picture comes entirely from the IHG files that were never opened.
| Segment | Booked within 3 days | Booked 4+ days out | Total room nights | Share booked 4+ out |
|---|---|---|---|---|
| Unfenced, our headline rate | 2,428 | 416 | 2,844 | 15% |
| Corporate negotiated | 487 | 498 | 985 | 51% |
| Fenced, the discount product | 164 | 727 | 891 | 82% |
| Extended stay | 179 | 460 | 639 | 72% |
| Opaque | 350 | 221 | 571 | 39% |
| Membership marketing | 288 | 181 | 469 | 39% |
| All segments | 4,102 | 2,647 | 6,749 | 39% |
IHG Booking Analysis workbook, sheet SP_crosstab, file dated 2026-08-06, attached to Spark 337753.
Our headline rate is a close-in product: 85% of it books inside three days. The two segments that collapsed, corporate negotiated and extended stay, book half to three quarters of their volume four or more days out, and neither buys on the headline rate. They buy on contracts. Extended stay production for the same week fell from 93 room nights in 2024 to 37 in 2025 to 4 in 2026 (IHG Production Analysis, Segmentation sheet, week of Jul 26 to Aug 1). August on the books is down 150 room nights year over year, all of it extended stay at minus 87 and corporate negotiated at minus 63 (IHG DSS Property Detail, as of 2026-08-06).
Benchmark archives run years and the brand segment reports carry two to three years inside them. A hotel can arrive with real seasonal depth on day one instead of us waiting twelve months for our own history to accumulate. That is the direct fix for the one gap we cannot otherwise close.
Comp set benchmark, monthly archive plus weekly going forward. Comp set by segment, which is the only thing that separates "the market lost it" from "we lost it." Our own segment mix with history. And the reservation level extract carrying booking date, arrival date, rate code, channel and room type, which is the only source that supports real pricing analysis rather than inference.
What discounts exist, when they trigger, how deep they run, and what the channel floors are. On this hotel the discount trigger and depth had to be inferred from five days of rate shopping, and it is still unclear whether it is one automated rule or a rule plus manual forward pricing.
Including which portal actually serves which report. The benchmark reports here are reachable only through the brand portal's single sign on, not by going to the vendor directly.
The three most useful sources on this hotel arrive weekly by email and none of them lands anywhere. A source with no ingestion path is not a source.
Whether the pattern generalizes. This hotel has 86 days of pace history. The other nine properties have between 1 and 9 days. If the intake standard assumes a HOUUS-level stack, most of the portfolio fails it on day one.
Whether reservation level and pace history can be requested retroactively at all, or only collected forward. That answer differs by brand and it determines whether onboarding buys us history or only a head start.
And whether to stay silent where depth is missing. A brief that says nothing about October pace is more trustworthy than one that guesses, but silence has a cost too.