Insights / Operating / Hospitality
Field note

RevPAR married to FP&A — why hotel groups run them in silos.

Revenue management and FP&A run in adjacent silos at most independent hotel groups, and the gap costs real money. What marrying them looks like — the joined weekly report, rate-vs-occupancy decomposition, GM scorecard reshape, 11-week calendar.

In every independent hotel group I have worked with across 2023 through 2026 — five groups, six to eighteen properties each, urban boutique, resort, and select-service across Canada, the US, and the UK — the revenue-management team and the FP&A team have run on different numbers. Same underlying data, different aggregations, different audiences. The revenue manager runs daily against pacing data, talking ADR and pickup and RGI. The FP&A lead runs monthly against the operating ledger, talking GOP and flow-through and the rolling forecast. The two teams produce reports that should reconcile and almost never do, and the gap is not a competence problem — it is a cadence and definitions problem that compounds quietly until the owner meeting where three different RevPAR numbers land on the same page. Marrying the two outputs into a joined weekly report is unglamorous, takes about eleven weeks of focused work to land cleanly, and changes the decisions both teams make. The framework below is what I have used five times.

01 Why the silos exist in the first place

Revenue management runs daily against pacing data, with rate and occupancy as the primary levers, and the conversation oriented toward decisions that need to be made before the next booking window closes. FP&A runs monthly against the operating ledger, with EBITDA and contribution as the primary levers, and the conversation oriented toward decisions that the owner or board will see in the next pack. Both cadences are correct for their respective audiences — the revenue manager is making a rate decision before lunch; the CFO is preparing a forecast that will land in twelve days. The silo is created not by either team running poorly but by neither team seeing the other team's output before the decision lands.

Underneath the cadence problem sits a definitions problem. RM uses PMS and RMS data — by stay date, by rate code, segmented the way the revenue strategy is constructed, with rooms revenue net of taxes and sometimes inclusive of packages. FP&A uses GL data — by business or posting date, with accruals and reclasses and the trade-deduction line that the OTA channels generate two months after the booking. The two source systems are both correct for their respective uses and produce different numbers when asked the same question. The RM weekly RevPAR does not match the FP&A weekly rooms revenue line, and over time the operating team learns to defend their version and the finance team learns to defend theirs and nobody trusts a single dashboard. The GM gets caught in the middle of the owner meeting answering "why am I seeing three different RevPAR numbers" and the integration project either gets prioritised or the silos calcify for another year.

The third layer is segmentation. Revenue manages by rate code, channel, and demand segment — BAR, OTA, corporate, group, package, wholesale. Finance manages by department and account in the chart of accounts, sometimes with a top-line segment that does not reconcile to the RM logic. Channel profitability — Net RevPAR by channel after distribution costs and commissions — sits between the two systems and is not visible in either. The brand-managed groups paper over this with the franchise reporting layer; the independent groups do not have that intermediary and the answer has to come from inside the operation.

The silo is not a competence problem. It is a cadence and a definitions problem. The fix is the joined report, the shared meeting, and one KPI dictionary both teams sign.
— From a 2025 revenue-and-finance integration, US boutique group

02 The joined weekly report — what it actually contains

The joined report sits on top of both teams' source data and produces a single document that both teams read each Tuesday for the week ended the prior Sunday. It is not a replacement for the daily RM flash and not a replacement for the monthly FP&A pack — it is the weekly bridge that makes both of those reports useful to the other audience. The structure is the same across the five groups where we have built it.

Top section: the portfolio snapshot. Occupancy, ADR, RevPAR by property and at group level, weekly. Variance vs Budget, vs prior year, vs the rolling forecast that FP&A last locked. Top three winners and bottom three by RGI movement. STR indexes (MPI, ARI, RGI) where the STR feed is weekly. The point of the top section is the same five-minute read the steering meeting opens with — what changed this week, where is it good, where is it broken.

Middle section: the rate-vs-occupancy decomposition. RevPAR equals occupancy times ADR; for any period versus its comparator, the variance decomposes into a rate effect, an occupancy effect, and a small interaction term. We compute rate effect as the ADR delta multiplied by prior-period occupancy, and occupancy effect as the occupancy delta multiplied by prior-period ADR, with the interaction term grouped against the larger of the two. The decomposition runs against Budget, against prior year, and against the rolling forecast — three columns per property. Underneath sits the segment and channel mix: how much of the RevPAR variance is coming from corporate vs leisure vs group, brand.com vs OTA vs wholesale, weekday vs weekend. The decomposition is what turns a "we missed RevPAR by 4%" sentence into "we held rate but lost compression nights to the compset," which is a different decision than "we discounted into a soft week."

Bottom section: the FP&A bridge. Contribution margin by property, weekly, with the linkage from RevPAR drivers to the contribution line visible. Rooms departmental contribution after distribution costs and rooms variable expense. Flow-through on incremental revenue versus Budget — what percentage of the revenue beat converted to GOP. Labor percent of revenue for the week, with the link back to the labor cadence post for the full methodology. The bottom section is what makes the joined report useful to the CFO — it shows the rate decisions and their margin implications on the same page, so the conversation about whether to chase the compression weekend with a promo is a margin conversation, not a RevPAR conversation.

$1.4M
Median annualised contribution improvement from joining the two reports — Putra & Co cohort, 5 independent hotel groups, 2023–2026.
~50%
Median reduction in quarter-ahead GOP forecast variance after the integration, same cohort (from ~12% MAPE to ~6%).
11 weeks
Median implementation timeline from scoping to go-live, with steering and data-council cadence in place.

The joined report is the artefact. The harder work — and the work that the artefact forces — is the KPI dictionary and the mapping specification that sit underneath it. One page per KPI: definition, formula, source system, timing basis (stay date vs business date), inclusions and exclusions. One page per mapping: rate code to segment, PMS revenue category to GL account, property-specific overrides. The KPI dictionary is what stops the next owner meeting from going sideways when the three numbers do not match.

03 How rate decisions change when both teams see the same numbers

The single most important change post-integration is the rate-vs-occupancy split becomes visible in the rate conversation itself, not in the post-mortem six weeks later. The revenue manager who has been buying share with discount sees the ADR Index trailing the RevPAR Index by four points on the dashboard and the conversation with the CFO opens with "we held occupancy, gave up rate, and gave up Net RevPAR after channel costs." That conversation does not happen in the silo. In the silo, the revenue manager sees a RevPAR Index above 100 and reports a win; FP&A sees the channel-cost line up and reports a margin drag; nobody connects the two until the quarter closes.

In practice, three rate decisions change inside the first quarter post-integration. First, the compression-night discount disappears. Properties stop running the "fill the last 15 rooms at BAR minus 20%" tactic on Friday-Saturday compression because the joined report has shown them, week after week, that the rooms revenue gain is fully offset by the OTA-commission and the package-allocation drag, and the rate-driven flow-through underneath is negative. Second, the package mix gets rebuilt. Package revenue, which had been booked as a single line in the PMS and reconciled to rooms vs F&B on a quarterly true-up, gets split at the booking level with the segment and channel mapping consistent — and what looks like a 65% gross margin package after the F&B allocation often turns out to be a 35% margin package once distribution and loyalty costs are loaded. Third, the corporate-rate negotiation shifts from volume targets to volume-plus-Net-RevPAR targets, because the procurement-driven rate cut that the corporate side wants to give to land the contract has a quantified contribution-margin cost that the finance side can defend in the negotiation.

Duetto and IDeaS have both been pushing this shift in their commentary since 2023 — the revenue conversation needs to be 3-4 minutes on STR and then 30 minutes on the forward-looking demand and the profit consequence of the rate choices being made. The joined report is the operationalisation of that commentary. Without the report, the conversation stays where the silo puts it; with it, the GM and the revenue manager and the CFO walk out of the Tuesday meeting with the same rate playbook for the next twelve days.

The compression-night discount disappears in the first quarter post-integration. The CFO is in the room when the rate is set; the OTA cost lands on the same page as the rooms revenue.
— From a working session with a UK boutique group, late 2025

04 Reshaping the GM scorecard so the incentives match

If the joined report is the data infrastructure, the GM scorecard is the incentive infrastructure that has to match it. The legacy independent-group GM plan I see most often pays a target bonus of 30-40% of base, weighted roughly 40% on RevPAR Index, 30% on GOP dollars, 20% on guest satisfaction, 10% on a discretionary basket. That structure rewards top-line RevPAR — including the occupancy-bought RevPAR that the integration was supposed to discourage — and treats GOP as a backstop rather than the primary number. Post-integration, the weights move.

The reshape I have run five times, with HSMAI's 2024-2026 commentary and HVS's management-fee work as the public anchors, looks like this. Forty percent on GOPPAR vs Budget and prior year — profitability dominates. Fifteen percent on flow-through, computed as the incremental GOP divided by the incremental revenue when revenue beats Budget, with a flex computation underneath when revenue misses. Ten percent on a rate-quality metric — ADR Index versus the compset, with a penalty if ADR Index trails RevPAR Index by more than three points, which is the formal way of saying "you bought share with occupancy, we are not paying for that." Ten percent on RevPAR Index itself. Fifteen percent on guest satisfaction. Ten percent on people and compliance. Profitability and rate quality together are 65% of the scorecard; volume and occupancy show up only through their effect on the profit metrics.

  1. 01
    GOPPAR gatekeeper: If GOPPAR comes in below 95% of Budget, the RevPAR Index and rate-quality components do not pay regardless of the index numbers. This is the structural fix for the "great RevPAR, terrible GOP" year that the old plan paid for. Documented extraordinary events (a citywide cancellation, an unbudgeted refurb) get a written carve-out approved by the steering committee before the year closes.
  2. 02
    Cadence split: Quarterly on-account payouts on the commercial components (RevPAR Index, ADR Index, rate-quality) — this aligns with the HSMAI sales-incentive cadence and gives the GM forward-looking signal. Annual true-up on GOPPAR, flow-through, guest, and people, because those metrics need the full year to settle without seasonality distortion.
  3. 03
    Multi-property GMs: For GMs running clusters of two or three properties, 70-80% of the scorecard weights at portfolio level (aggregate GOPPAR, portfolio flow-through, portfolio RGI). The remaining 20-30% splits equally across each property's controllable GOP variance and guest score, so a single weak property cannot hide inside the cluster average.

The plan reshape lands in roughly one full bonus cycle. The transition year is messy — GMs who have been paid on RevPAR for a decade need the rate-quality logic walked through line by line, and HR needs to handle the conversations where the new plan would have paid less in the legacy comp year. The structural argument that wins those conversations is the same one the joined report makes: a 100 RGI bought with discount is not a 100 RGI worth paying for. The HSMAI 2026 commentary, the HVS work on management fees, and the STR analysis showing profit grows 1.5-2x top-line when ADR drives RevPAR are the supporting citations that turn the plan reshape from a CFO preference into an industry direction.

05 What FP&A learns from reading the RM dashboard

The reverse direction matters as much as the forward direction. The CFO and FP&A team that start reading the RMS dashboard alongside the operating ledger learn things the operating ledger cannot tell them, and the rolling forecast quality improves in measurable ways.

First, pace data — the booking velocity versus the same point last year — becomes the leading indicator for the next two quarters of room revenue. The operating ledger lags pace by 6-12 weeks depending on booking window; reading pace weekly compresses the lag and lets the forecast adjust before the variance shows up in the close. Second, unconstrained demand — the RMS estimate of demand that would have booked if rate and inventory had been unconstrained — becomes the upper-bound scenario for the forecast, and the gap between constrained and unconstrained becomes the quantified case for the pricing experiment or the inventory release. Third, channel mix shifts in pace data — the BAR-to-OTA migration that is happening in real time, the corporate-to-leisure rotation post-event — show up four to six weeks before the channel-cost line moves in the ledger, which lets the FP&A team adjust the distribution-cost forecast before the variance lands.

Across the five-group cohort, the measurable forecast-quality improvement post-integration runs in a consistent band. Quarter-ahead RevPAR forecast variance — median across the cohort — drops from roughly 8% MAPE pre-integration to roughly 4% post. Quarter-ahead GOP variance drops from roughly 12% MAPE to roughly 6%. The 13-week cash forecast accuracy on the week-4 ending balance moves from ±9% pre to ±5% post. None of those numbers are heroic individually; together they materially change the owner conversation, because the forecast the CFO is defending in the quarterly owner meeting is half as wrong as it was the year before.

8% → 4%
Median quarter-ahead RevPAR forecast variance, pre vs post integration (Putra & Co cohort, 5 groups).
12% → 6%
Median quarter-ahead GOP forecast variance, pre vs post integration, same cohort.
±9% → ±5%
Median 13-week cash forecast accuracy on the week-4 ending balance, same cohort.

The other direction the FP&A team learns is what does not improve. Group wash — the cancellation and attrition that hits group blocks 30-90 days out — does not become more predictable from reading the RMS dashboard, because the dashboard does not show it. Group wash needs a separate sales-pipeline read and a group-block forecasting layer that sits next to the RMS feed. Citywide event volatility — the conference that gets cancelled, the festival that gets moved — also does not improve from the RMS feed alone; it needs the local commercial team's judgment overlay. Knowing what the integration does not solve is part of running the meeting honestly.

06 The 11-week implementation calendar

The implementation calendar is unglamorous and predictable. Across the five groups, weeks 1-3 are about definitions and stakeholders; weeks 4-7 are about plumbing; weeks 8-11 are about pilot, rollout, and the cutover from the legacy parallel reports. The governance is what keeps it on schedule — a steering committee meeting fortnightly with the CFO or COO, the head of revenue, the FP&A lead, the data/BI owner, and a rotating GM rep; a data council meeting weekly with the FP&A and RM analysts, the BI developer, and a property controller. Without both forums, the project drifts into either a finance-led project that the commercial team resists or a commercial-led project that finance does not trust.

  1. 01
    Weeks 1-3 — scope, discovery, KPI dictionary: Week 1 locks the scope (which properties, which systems, what success looks like) and stands up the steering committee. Week 2 inventories the source systems — PMS exports, RMS feeds, GL chart of accounts, existing RM and FP&A reports. Week 3 runs the KPI workshop with RM, FP&A, and one GM in the room, locks the KPI dictionary v1 (rooms revenue, net rooms revenue, OCC, ADR, RevPAR, Net RevPAR, GOP, GOPPAR, labor metrics), and gets the steering committee sign-off on the report wireframes.
  2. 02
    Weeks 4-7 — data model, ETL, prototype, forecast layer: Week 4 designs the joined data model and the rate-code-to-segment and PMS-to-GL mappings. Week 5 builds the ETL — pipelines from PMS, RMS, and accounting into the BI layer. Week 6 builds the prototype dashboard and runs the first PMS-vs-GL reconciliation for two or three recent weeks, agreeing which differences to adjust for and which to disclose. Week 7 adds the FP&A forecast and budget layers, the pickup and pace from RMS, and the variance analysis (actuals vs forecast vs budget).
  3. 03
    Weeks 8-11 — pilot, rollout, governance hardening, go-live: Week 8 pilots the joined report in one or two properties, replacing (not adding to) the existing decks in those weekly meetings. Week 9 rolls out to all in-scope properties with access controls (property-level users see own + group) and automated data-quality checks (refresh status, PMS-vs-GL reconciliation variance, missing days). Week 10 finalises the runbook, the KPI dictionary, and the change-request process — who can request a new metric, who approves, what the data-council cadence is. Week 11 gets the steering committee go-live sign-off and retires the legacy parallel reports.

The failure modes are also predictable. Scope creep — the temptation to add F&B and TRevPAR and channel profitability net of commissions in the first build, instead of locking the rooms-and-GOP core and shipping. Definition ambiguity — letting the KPI dictionary stay vague to avoid the RM vs FP&A argument, which guarantees the argument resurfaces in week 9 with worse consequences. No formal cutover — letting the legacy RM and FP&A weekly reports run alongside the joined report indefinitely, which means the joined report is treated as one more dashboard rather than the source of truth. The discipline that prevents all three is the steering committee's willingness to defer Phase 2 enhancements, lock definitions before the build, and formally retire the legacy reports at week 11.

07 What changes at the property and at the group level

Property-level changes show up first. The GM and the revenue manager and the hotel controller walk into the Tuesday meeting with the same three pages and the same vocabulary, and the meeting compresses from sixty minutes of reconciling numbers to twenty minutes of deciding what to do about them. The rate-quality conversation that used to sit in the corporate revenue call now happens inside the property, because the data is in front of the GM in real time. The package mix conversation, which used to happen quarterly when the F&B allocation got trued up, happens weekly because the segment and channel mapping is consistent.

Group-level changes follow. The CFO and the head of revenue stop running parallel forecasts and start running one. The owner pack — the monthly or quarterly document the institutional or family-office owners read — gets cleaner, because the RevPAR variance bridge and the GOP variance bridge reconcile to the same drivers. The annual budget process compresses, because the segmentation and the chart-of-accounts mapping are already locked. Capital decisions get sharper, because the room-by-room return on a refurb gets a credible revenue forecast underneath it (from the RMS feed) and a credible cost-and-margin forecast (from the operating ledger), and the two sides do not have to be reconciled in a separate exercise.

The harder change is cultural. The RM team that was used to running its own meeting with its own slides has to get comfortable with the CFO in the room asking margin questions; the FP&A team that was used to its monthly cadence has to get comfortable with the weekly rhythm and the operational tempo of the commercial side. The cultural change runs about two quarters past the technical go-live. Groups that have done the technical integration and not the cultural integration end up with a joined report that nobody trusts; groups that do both get the contribution improvement and the forecast accuracy improvement and, more importantly, get a hotel-finance organisation that runs as one team instead of two.

For groups thinking about whether to start this work, the answer is shaped less by size and more by the diagnostic in section one. If your RM weekly RevPAR does not reconcile to your FP&A rooms revenue line, if your CFO has not read the RMS dashboard in the last month, if your GM scorecard still pays primarily on RevPAR Index without a rate-quality modifier — the silo is costing you contribution dollars every quarter. Eleven weeks of focused work, run inside the cadence above, gets the integration to a defensible go-live. The 18-month exit-prep calendar applies if you are heading toward a sale; the labor cadence and the boutique-CFO playbook apply for the operating discipline. The integration is the connective tissue that makes all three work together.

Frequently asked questions

Why do revenue management and FP&A run in silos at most independent hotel groups?
The split is a cadence and definitions problem, not a competence problem. RM runs daily against PMS and RMS data with rate and occupancy as the levers; FP&A runs monthly against the operating ledger with GOP and contribution as the levers. The two source systems use different timing bases (stay date vs business/posting date) and different segmentation (rate code/channel vs department/account), so the weekly RevPAR from RM and the weekly rooms revenue line from FP&A do not reconcile out of the box. Until a shared KPI dictionary and a joined weekly report exist, both teams defend their own numbers and the GM gets caught between three different RevPAR figures in the owner meeting.
What does the joined weekly report contain?
Three sections. Top: portfolio snapshot — OCC, ADR, RevPAR by property weekly, variance vs Budget/LY/rolling forecast, STR indexes (MPI, ARI, RGI) where the feed is weekly. Middle: rate-vs-occupancy decomposition — rate effect (ADR delta × prior-period OCC), occupancy effect (OCC delta × prior-period ADR), interaction term, run against Budget/LY/forecast with segment and channel mix underneath. Bottom: FP&A bridge — contribution margin by property weekly, rooms departmental contribution after distribution costs, flow-through on incremental revenue, labor percent of revenue. The CFO sees rate decisions and their margin implications on the same page; the revenue manager sees the margin consequence of the rate choices.
How do rate decisions actually change after the integration?
Three changes inside the first quarter. The compression-night discount disappears — properties stop discounting to fill the last 15 rooms on Friday/Saturday compression because the joined report shows the OTA-commission and package-allocation drag fully offsets the rooms revenue gain. The package mix gets rebuilt at the booking level with consistent segment and channel mapping, surfacing packages that look 65% gross margin but are 35% margin after distribution and loyalty. The corporate-rate negotiation shifts from volume-only targets to volume-plus-Net-RevPAR targets, with a quantified contribution-margin defence for the rate floor.
How should the GM scorecard be reshaped post-integration?
Move from a RevPAR-dominated plan (40% RevPAR Index, 30% GOP, 20% guest, 10% discretionary) to a profitability-and-rate-quality plan. Recommended weights: 40% GOPPAR vs Budget/LY, 15% flow-through on incremental revenue, 10% rate-quality (ADR Index with a penalty if ADR Index trails RevPAR Index by more than 3 points), 10% RevPAR Index, 15% guest satisfaction, 10% people and compliance. Add a GOPPAR gatekeeper — if GOPPAR comes in below 95% of Budget, the commercial components do not pay regardless of the index numbers. Quarterly on-account payouts on commercial metrics, annual true-up on GOPPAR and flow-through.
What forecast-quality improvement does the integration actually produce?
Across the Putra & Co cohort of 5 independent hotel groups (6-18 properties each, 2023-2026), median quarter-ahead RevPAR forecast variance moved from roughly 8% MAPE pre-integration to roughly 4% post; median quarter-ahead GOP variance from roughly 12% to roughly 6%; and 13-week cash forecast accuracy on the week-4 ending balance from ±9% pre to ±5% post. The mechanism is straightforward — FP&A reads pace and unconstrained demand from the RMS dashboard weekly, which compresses the 6-12 week lag between booking activity and the ledger.
How long does the implementation actually take?
Eleven weeks from scope to go-live, with the right governance in place. Weeks 1-3: scope, system discovery, KPI dictionary lock with RM + FP&A + GM in the room. Weeks 4-7: data model and mapping design, ETL build, prototype dashboard, PMS-vs-GL reconciliation, FP&A forecast and budget layers added. Weeks 8-11: pilot in 1-2 properties (replacing not adding to existing decks), rollout to all in-scope properties with access controls and data-quality checks, runbook and KPI dictionary finalisation, steering committee go-live sign-off, legacy parallel reports retired. A steering committee (fortnightly) and a data council (weekly) run alongside the build.
What does the integration not solve?
Group wash — the cancellation and attrition that hits group blocks 30-90 days out — does not improve from the RMS dashboard because the dashboard does not show it. Group wash needs a separate sales-pipeline read and a group-block forecasting layer next to the RMS feed. Citywide event volatility (cancelled conferences, moved festivals) also does not improve from the RMS feed alone; it needs the local commercial team's judgment overlay. The cultural integration — RM getting comfortable with the CFO in the room asking margin questions, FP&A getting comfortable with the weekly operational tempo — runs about two quarters past the technical go-live; groups that do the technical work and not the cultural work end up with a joined report nobody trusts.
Notes

RM and RMS commentary: STR / CoStar — Total revenue management in hotel profitability; Duetto — Five hospitality leaders on the shift to profit-thinking; HotelTechReport — Best BI Software for Hotels 2026; Lighthouse — How independent hotels can conquer revenue challenges.

GM incentive plan structure: HSMAI Hotel Management Company Sales Incentive Plans special report; Hospitality Net — Why 2026 Will Force a Redesign of Pay, Incentives & Performance Expectations; Hotel Financial Coach — Management Incentive Plans; HVS — A New Approach to Hotel Management Fees.

FP&A and forecasting integration: Cherry Bekaert — FP&A Forecasting Strategies; InsightSoftware — Why Integration is the Key to FP&A Today; Phoenix Strategy Group — Hospitality Financial Dashboards; Atlar — 13-week cash flow forecast.

Forecast-quality and contribution-margin improvement figures: Putra & Co internal cohort data, 5 independent hotel groups (6-18 properties each), 2023-2026 engagements. Pre/post integration comparison on quarter-ahead variance and 13-week cash forecast accuracy.

Full source list at content-pipeline/research/revpar-married-to-fp-and-a-hotel-groups/sources.md in the Putra & Co content pipeline.

About the author
Sid Ahuja
Partner · Operating

Sid Ahuja

Senior Partner

Capital markets and M&A background. Multi-unit specialist — hotel groups, dental and medical DSOs, real-estate operating cos, professional services firms, construction platforms. Leads sell-side processes and roll-up sequencing where unit economics are the deal. RevPAR, same-store and unit-economics rebuilds.