# Why hotel operators at Otelier's summit want daily numbers before AI agents

By MarketScale Newsroom · Published 2026-11-17 · Hospitality on MarketScale
Canonical: https://www.marketscale.com/industries/hospitality/why-hotel-operators-at-oteliers-summit-want-daily-numbers-before-ai-agents

> Real hotel GOPPAR still trails 2019 by 20%. Otelier's customers are moving finance checks to daily and making trusted data the price of entry for AI.

## Key points

- A day that balances can still be wrong, and an AI agent reading that day will act on it with full confidence, so trusted data and a named reviewer come before automation.
- For 2027 budgeting, Sigala points to a lower starting point for rate assumptions—roughly 2.9%—while noting that undistributed expenses are growing slightly faster than revenue.
- Hotels that budget once a year can start reforecasting weekly; operators already expect slower group pace in Q1 2027, which makes the first quarter the earliest test.

Twenty-five people who run hotels for a living sat down at Hammock Beach Golf Resort & Spa, and one number hung over the table: real GOPPAR is still 20% below 2019. Otelier CEO Chris Sigala, who hosted the group at his company's Otelier Optimize 2026 customer event, published what he heard in a Nov. 12 column for Hotel-Online. The room included revenue, finance and technology leaders from CoralTree Hospitality, Crescent Hotels & Resorts, Highgate, Mandarin Oriental Hotel Group, MCR Hotels and InnVest.

Nobody expected 2027 to be easy. Operators pointed to softer group pace in the first quarter and difficult leisure comparisons against last year. They also described a K-shaped market, in which luxury properties hold rate while extended stay cannot. Costs, technology spending among them, are climbing faster than revenue. The response had two parts: review the numbers much more often, and hand AI a task only after the data behind it can be trusted.

Dana Cariss, senior VP of revenue optimization at CoralTree, described the stakes in blunt terms. In her view, operators face an uncomfortable reality: they must run their businesses in a new way, since neither the approach in place 10 years ago nor the one hotels rely on now will serve as the model of the future. The column's headline picks up on that line.

## Rate gains that thin out before reaching profit

The wider industry figures point the same way. At the Lodging Conference in Phoenix, STR senior director of analytics Isaac Collazo said the industry has regained momentum after a rocky 2025 but not its margins, Hotel Dive's Noor Adatia reported. Top-line performance is stronger, yet undistributed expenses are growing slightly faster.

His own summary came in two parts. Collazo said demand and RevPAR are both heading the right way, and that the business now feels somewhat normal and sustainable, something he had not been able to say for a while. On the other side, he said, hotels still have limited pricing power, and rising expenses keep squeezing margins.

Block Field

The gap between those two rate figures matters for anyone writing a 2027 budget. The tournament was a 2026 event, so the roughly 2.9% figure is the more honest starting point for next year's rate assumption, and it leaves little room once expense growth is subtracted. That's why Sigala wrote that flow-through, the portion of revenue that reaches profit, will define success in 2027, more than top-line growth.

JLL's Adam McGaughy, writing in Hotel Management, supplied a useful benchmark from STR data: across the top 25 U.S. markets, the average spread between ADR growth and inflation sits at about 20 basis points. In a typical big market, then, rate is barely outrunning prices. That leaves almost no real pricing gain to absorb the expense growth Collazo described, and it helps explain why an inflation-adjusted profit measure is the one still stuck below 2019.

Block Field

Sigala attached a test to every technology purchase: name the operating outcome it will improve and how it will be measured. The same test covers knowing the true cost of every booking. For a procurement or IT lead with a stack of renewals, that turns a feature comparison into a question with a number attached.

## Moving month-end cleanup to every morning

Crescent CFO Rob Smith put the finance argument most bluntly: in his view, every decision in a hotel is a finance decision. Most hotel companies aren't set up that way. Decisions made on property turn into numbers at month-end, often after a tired night auditor has served as the last line of defense.

Highgate's Shawn Goodwin, VP of revenue and distribution, gave the example that explains the problem. The day balances and the income journal looks clean, but an entry has landed in a general ledger account where it doesn't normally go. Balancing checks that totals match. It doesn't check that each dollar went to the right place, and that kind of error can sit unnoticed until someone reconciles weeks later.

Sigala said the sharpest finance teams are shifting that work earlier in the process. They write down who is responsible for every data handoff, and they convert month-end cleanup into a daily early warning. Documenting ownership costs little, and it is the step that opens the way to the next one.

The same logic is reshaping how hotels want their reporting tools to behave. For years, Sigala wrote, hotel business intelligence sold visibility: more reports, more charts, more logins. Todd Wood, VP of technology at Mandarin Oriental Hotel Group, described moving from dashboards to cockpits or operational consoles. Sigala's point was that a chart leaves interpretation to someone who doesn't have time for it, while an alert shows overtime about to hit or an expense running off pattern, with enough context to act.

CoralTree Hospitality is already moving that way. Its AI agents go through each P&L across the portfolio and flag gaps more quickly than an analyst could. For management companies overseeing many properties, this is the workable model: software reads every statement and passes only the exceptions to people. Sigala expects the next generation of hotel tools to be judged on what they recommend and whether anyone follows through.

## Scoping AI against the biggest line on the P&L

Labor remains the biggest line on a hotel P&L. Attendees pushed back on the notion that AI just means a smaller headcount. After COVID, properties were already running lean. Wood said the opportunity he sees is AI helping optimize labor rather than replacing staff.

This is where the case for daily data becomes a hard requirement. Sigala's guardrail was short: verify everything, because AI raises the stakes on data quality. An experienced controller will catch a figure that looks wrong. An agent may use that same figure with no hesitation at all. A day can balance and still be wrong, and an AI agent working from that day will act on it with full confidence. As Sigala told the room, context is everything. He laid out this sequence to prevent that:

1. Skip the headcount target as a starting point.
2. Pick a few defined tasks the team dislikes.
3. Confirm the data behind those tasks is trusted.
4. Name the person who reviews the output.
5. Expand only as the results hold up.

Finding those tasks doesn't require a vendor's list. Hotel AI consultant Are Morch argues on his blog that the holiday rush shows them clearly: double entry, re-keying between systems that don't integrate, and manual reconciliations that only get painful at volume. Those are close cousins of the posting errors Goodwin described, which suggests the first AI candidates and the first data-ownership fixes often sit in the same workflows.

Morch's broader point is about timing. In his telling, the quiet months let a hotel paper over its gaps with extra effort, while a full house and thin staffing leave no slack to hide them. With the holiday season arriving just as 2027 plans firm up, that suggests a practical order for a general manager: log where staff lose time during the busy weeks, then choose the first AI tasks from that log.

Outside research supports being careful about expecting uniform results. An NBER working paper by James Brand, Mert Demirer, Connor Finucane and Avner Kreps, revised in December 2025 and not peer-reviewed, used CPU utilization data from nearly 100,000 firms and found large, persistent differences in how productively companies use the same cloud computing. That study covers cloud infrastructure, not hotel AI. Still, it indicates that buying the same tool as a competitor doesn't guarantee the same return, and that how a firm runs the tool drives the outcome. That fits Sigala's insistence on a named reviewer and a measured result.

The paper's details sharpen the point. Firms improved their cloud productivity by 33% in the first year after adoption and reached a stable level only after four years. Above-median firms were more than twice as likely as weaker ones to shut down idle machines, a monitoring habit that resembles the daily early warning the finance teams described. Gains also stayed mostly inside the division that earned them, with minimal transfer across divisions.

That last finding would matter for a management company if the pattern carries over to hotel AI. A successful pilot at one property would not automatically teach the rest of the portfolio, and the review loop in Sigala's sequence would need to run at each site.

## Where the urgency is lower

The strongest case against urgency is demand. Bernie Baumohl, chief global economist at The Economic Outlook Group, told the Lodging Conference the industry is in fairly good shape, with strong demand and severely constrained supply. McGaughy also cited STR projections of 0.4% supply growth in 2026 and 0.6% in 2027, and pointed to downtown Chicago leading the top 25 U.S. central business districts with 15.2% RevPAR growth in the second quarter.

Chicago's group calendar also runs against the national worry. As of June 7, 2026, McCormick Place bookings for 2027 were pacing at 104% of 2019 levels, and Chicago's year-to-date ADR growth was running about 210 basis points ahead of inflation, against roughly 20 for the top-25 average. The softer first-quarter group pace the Otelier group described is clearly not uniform across markets.

For hotels in markets like that, rate may cover cost growth for a while, and daily finance discipline becomes a nice-to-have. The pressure falls hardest on segments that can't hold rate, which the Otelier group placed on the extended-stay side of the K.

Baumohl's upbeat read still flagged three risks for 2027: rising energy costs, electricity rates among them; a potential pullback in consumer spending; and how unpredictable President Trump's decisions may be during the final two years of his term. That first risk sits on the cost side, not the demand side, so even a hotel in a strong market has reason to spot an expense drifting off pattern early.

The read also comes from one vendor's own customers. Otelier says it serves more than 10,000 hotels, and the Hammock Beach table was 25 of its top customers, a small group, though STR's expense comments point the same direction.

## Forecasts that go stale by spring

Budgeting got the same treatment as the books. MCR Hotels' Kyle McCaig, VP of technology operations, compared the accuracy of many outside forecasts to exit polls. The operators Sigala admired use external forecasts as context, build from their own trends and stop treating the budget as an annual ritual, because in a fast market it's stale by spring. His advice for hotels where daily reforecasting feels like too much: start weekly, so staffing and spending decisions move sooner.

InnVest CIO McLean Xavier set the limit on all of it, saying hospitality is still a people business and these tools should help staff be more personable. Sigala added a practical warning: a platform built to simplify work fails if the team sees it as one more login. The operators asked for fewer screens and numbers they don't have to second-guess. Sigala called that a high bar for every tech provider, Otelier's included.

Block Field

The first real test comes early. Operators already flagged slowing group pace for the first quarter of 2027. A hotel reforecasting weekly from its own data should see that softness in time to adjust schedules. One running on an annual budget may not see it until the quarter's numbers close.

## Sources

- [The Uncomfortable Truth About Running Hotels in 2027](https://www.hotel-online.com/news/the-uncomfortable-truth-about-running-hotels-in-2027)
- [4 key takeaways from 2026's Lodging Conference](https://www.hoteldive.com/news/4-hospitality-takeaways-2026-lodging-conference/833278/) (Hotel Dive)
- [Chicago's hotel comeback is already here](https://www.hotelmanagement.net/business-strategy/chicagos-hotels-are-breaking-records-narrative-hasnt-caught) (Hotel Management)
- [AI in Hotel Operations: Let Your Busiest Season Show You ...](https://aremorch.com/ai-hotel-operations-peak-season/)
- [Firm Productivity and Learning with Digital Technologies](https://www.nber.org/system/files/working_papers/w32938/w32938.pdf) (National Bureau of Economic Research)

Tags: Otelier, hotel operations, hotel finance, GOPPAR, AI agents, hospitality technology, CoralTree Hospitality, Mandarin Oriental, labor management, forecasting and budgeting, business intelligence, hotel management companies

---
Source: MarketScale, https://www.marketscale.com/industries/hospitality/why-hotel-operators-at-oteliers-summit-want-daily-numbers-before-ai-agents. Published for AI indexing and citation; cite the canonical URL. Site guide for agents: https://www.marketscale.com/llms.txt
