AI use case

AI copilot for hotel revenue management

An employee facing AI system that forecasts demand for a hotel or portfolio by date, room type and segment, recommends or automatically adjusts room prices and availability controls within limits a revenue manager sets, and scores group and event enquiries for true profitability, so a small revenue team can run pricing that used to need daily manual adjustment in the property management system.

By Len Debets · Last verified 28 September 2026 · 2 public deployments

33%
Reported revenue uplift
Hôtel Swexan, vendor claim.
USD 1.2 million to USD 8 million
Indicative value per year
A hotel group with 20 properties and average annual room revenue of USD 3 million per property. Worked example, see how it is calculated.

What problem does it solve?

A hotel's price should change with demand: a room that rents cheaply on a quiet Tuesday can be worth much more on a night a conference fills the market. Before a revenue management system, that means one person watching pickup, competitor rates and events across every room type and segment, and updating prices by hand in the property management system. RIMC Hotels & Resorts Group describes this before it adopted Duetto: pricing managed manually, with adjustments made through the property management system, a process it calls time consuming and inefficient, especially for pricing and capacity control.

The problem compounds with scale and complexity. Hôtel Swexan runs a 134 room luxury property with more than 20 room types, eight premium suites and five food and beverage outlets, with one person, Director of Revenue Jessica Schiele, responsible for the whole pricing strategy. Before an AI assisted revenue management system, every hour she spent pushing rates by hand was an hour not spent on analysis or strategy, and macro level tools such as closing the whole house or a blanket minimum stay were the only practical levers, because granular, segment level rules took too long to maintain manually.

How does it work?

  1. Pull the data continuously. The system reads live occupancy, rate and booking pace from the property management or central reservation system, plus competitor rates and, where used, group and food and beverage data.
  2. Forecast demand. It predicts occupancy and demand by date, room type and segment, updating the forecast as new bookings and cancellations arrive.
  3. Recommend or execute a price. Within a floor, a ceiling and a maximum daily rate change the revenue manager configures, it either recommends a rate for approval or pushes it directly to the property management system.
  4. Score group and event enquiries. For a group or function space request, it weighs room revenue, food and beverage minimums, room rental and the displacement of other business to recommend a profitable quote, not just an available rate.
  5. Flag what needs a person. Forecast misses, a recommendation outside the configured range, a new room type or segment and any group quote below the profitability threshold go to the revenue manager to review or approve.
Audience
Employee facing
Autonomy
Supervised agent
Adoption
Early adopters
Channels
Internal tools

What is it worth?

Benchmarks are computed from the public deployments below: one data point per organization per KPI, with who made each claim.

Value benchmarks for AI copilot for hotel revenue management
KPIMedianReported rangeData pointsClaimed by
Revenue upliftToo few to pool
28.4% to 33%
21 organization, 1 vendor

Value drivers: Revenue growth, Employee productivity, Speed and cycle time.

Indicative value

A hotel group with 20 properties and average annual room revenue of USD 3 million per property

USD 1.2 million to USD 8 million

Additional room revenue from AI assisted pricing per year

How this is calculated

Formula: properties * roomRevenuePerProperty * revparUplift. The low scenario uses every low input, the high scenario every high input.

InputLowHighBasis
Properties in the portfolio properties, properties2020The reference group.
Annual room revenue per property before the change roomRevenuePerProperty, USD per property per year2,000,0004,000,000Editorial assumption, replace with your own room revenue.
RevPAR uplift attributable to AI assisted revenue management revparUplift, fraction of room revenue0.030.1Conservative against the evidence on this page. RIMC Hotels & Resorts Group reports a 28.44% RevPAR increase at its Polish property, with no period or baseline stated for that figure, and, separately, an increase in RevPAR at every hotel in its portfolio compared to the previous year since adopting Duetto in 2022. Duetto's case study reports 33% total hotel RevPAR growth at Hôtel Swexan, 2025 versus 2024, with no date given for when Hôtel Swexan itself went live on Duetto. RIMC's own account describes its pricing before Duetto as fully manual, adjusted through the property management system, with no revenue management system of any kind in place before 2022: a group already running a mature revenue management system and adding only an AI forecasting layer on top of it should expect a smaller further gain than a group's first system delivered. Neither hotel's own case study uses the words AI or machine learning about its pricing; both run Duetto's revenue management platform, and Duetto's own product page for that platform, GameChanger, describes pairing data with "AI-driven rate recommendations" (checked with usecases:source, 2026-09-28).

What it leaves out: Gross room revenue uplift only. It leaves out the cost of the revenue management system and its integration, any change in distribution cost or food and beverage revenue, and the fact that part of a RevPAR gain in any single case can come from market conditions rather than the system itself.

Who already uses it?

2 public deployments, strongest evidence first. Grades: A regulator or audit, B the organization itself, C vendor case study, D anonymous or estimate.

Hôtel Swexan

United States · Travel and hospitality · 2025

ProductionGrade C

Hôtel Swexan, a 134 room luxury property in Dallas with more than 20 room types, eight premium suites and five food and beverage outlets, runs its entire revenue management function with one person, Director of Revenue Jessica Schiele. Before Duetto, the hotel relied on a remote third party consultant with no visibility into daily operations, and rate changes were slow with no room for granular, segment level pricing. Duetto now forecasts occupancy, recommends demand driven rates by room type and segment, executes automated pricing strategies around the clock based on occupancy thresholds and segment logic Schiele defines, and scores group and event enquiries for profitability (room rate, food and beverage minimums, room rental and displacement) so any team member can quote a profitable group rate. Comparing 2025 with 2024, Duetto's case study reports double digit RevPAR growth in every segment.

  • Revenue uplift: 33%, 2025 versus 2024
    "Total hotel RevPAR grew 33% year on year — a result that reflects more than favourable market conditions."
    Claimed by: vendor
  • Revenue uplift: 45%, 2025 versus 2024
    "Suite RevPAR grew 45% year on year — occupancy up 29%, ADR up 13%."
    Claimed by: vendor

RIMC Hotels & Resorts Group

Germany · Travel and hospitality · 2022

ProductionGrade C

RIMC Hotels & Resorts Group, a hotel association headquartered in Hamburg with business, city and holiday hotels across several countries, replaced manual pricing, made through the property management system and time consuming for pricing and capacity control, with Duetto's cloud revenue management system in 2022. Duetto forecasts demand and recommends real time, demand driven room prices across the portfolio, in place of the group's previous manual adjustments. Head of Revenue Henning Möhn describes an increase in RevPAR at every Duetto hotel in the portfolio and a 28.44 percent RevPAR increase at the group's Polish property, plus daily time saved on pricing and monitoring per hotel.

  • Revenue uplift: 28.4%
    "We've seen a 28.44% increase in RevPAR for our Polish property and an overall RevPAR increase across our portfolio."
    Claimed by: organization

How do you implement it?

A model agnostic playbook: what to prepare, the order to build in, and what goes wrong.

Data you need

  • At least one to two years of historical occupancy, rate and booking pace by room type and segment
  • A competitor set and a rate shopping feed
  • Group, event and food and beverage data if group profitability scoring is in scope
  • Agreed rate floors, ceilings and maximum daily change per room type and segment

Systems to integrate

  • Property management system or central reservation system
  • Channel manager and distribution system
  • Competitor rate shopping tool
  • Point of sale or catering system for group and food and beverage profitability
  • Business intelligence or reporting tool for the revenue manager's dashboard

Complexity: Medium

The forecasting and pricing logic itself is largely the vendor's; the work is integrating the property management or central reservation system, a channel manager and a competitor rate feed cleanly enough that the forecast is trustworthy, and agreeing the floor, ceiling and approval rules before turning on any automatic pricing.

  1. 1

    Connect the data before you connect the automation

    Integrate the property management system, channel manager and rate shopping feed first, and run the forecast in shadow mode against actual outcomes before any price is pushed automatically.

  2. 2

    Set the guardrails first

    Agree the rate floor, ceiling and maximum single day change per room type and segment with the revenue manager before switching on automatic pricing, not after.

  3. 3

    Start with one segment or room type

    Prove forecast accuracy and RevPAR impact on transient, individual bookings before extending automation to groups, suites or a new property.

  4. 4

    Add group and event profitability scoring once transient pricing is stable

    Bring in food and beverage minimums, room rental and displacement so any team member, not just the revenue manager, can quote a group at a profitable rate.

  5. 5

    Review overrides every week

    Look at every case where the revenue manager overrode the recommendation, and why; a pattern of overrides in the same direction means the forecast or the guardrails need adjusting.

Guardrails

  • Automatic pricing bounded by a floor, a ceiling and a maximum single day rate change, set by the revenue manager
  • Any recommendation outside the configured range, or for a new room type or segment, requires human approval before it goes live
  • Group and event quotes below the agreed profitability threshold route to a person, not straight to the quote

KPIs to instrument

  • Forecast accuracy against actual occupancy and pickup, by room type and segment
  • RevPAR, ADR and occupancy against the competitor set, before and after
  • Override rate: how often the revenue manager changes or rejects a recommendation, and why
  • Time the revenue team spends on manual pricing and monitoring tasks

Human in the loop

The revenue manager sets and periodically revisits the pricing thresholds and rules, approves anything the system flags as outside them, and owns the final call in unusual situations such as a citywide event, a competitor's distress pricing or a local disruption the forecast has not seen before.

Common failure modes

A forecast blind to what has not happened before
A model trained on historical patterns misses a new event, a cancellation wave or a local disruption. Keep an anomaly alert and a fast human override path for exactly these moments.
Automated pricing that moves against a competitor's algorithm
Automated systems in the same market can drift in either direction: repeatedly undercutting each other compresses rates for everyone, and, when several hotels lean on the same vendor's pricing recommendations, the arrangement can itself become an antitrust question. Two federal appeals courts have reached different outcomes on different shared pricing software complaints: the Ninth Circuit upheld the dismissal of a Sherman Act suit over shared Las Vegas Strip hotel pricing software in 2025, and the Third Circuit revived a similar suit over Atlantic City casino hotel pricing software in 2026, based on that complaint's own allegations (see the incidents on this page). Enforce both a rate floor and a ceiling, and review competitor reactions, not just your own pickup.
Chasing short term pickup at the expense of the base
The system discounts aggressively on a soft looking date and displaces higher value corporate or group business that would have booked later. Weigh displacement, not only occupancy, in the recommendation.

What are the risks and rules?

EU AI Act

Minimal risk

A demand forecasting and pricing tool used by hotel staff is not listed in Annex III and is not a system that decides on a natural person's access to an essential service; it prices a hotel room, not a person. It is not customer facing, so the Article 50 transparency duty for conversational AI does not apply. The AI literacy obligation on staff who use AI systems (Article 4) still applies.

Rules that apply

Guidance

  • Article 4, AI literacy (European Union, Europe). Providers and deployers must take measures to ensure staff and other people who operate and use an AI system on their behalf have a sufficient level of AI literacy.

Controls to put in place

  • Rate floor, ceiling and maximum daily change configured per room type and segment, reviewed periodically
  • Every automatic price change logged with the forecast and inputs that produced it, for audit
  • Regular comparison of AI recommended rates against revenue manager overrides to catch systematic bias

When it went wrong elsewhere

  • Gibson v. Cendyn Group: Ninth Circuit affirms dismissal of a Las Vegas Strip hotel pricing algorithm suit. A Sherman Act Section 1 class action alleged that competing hotels on the Las Vegas Strip fixed room prices by all licensing Cendyn Group's algorithmic pricing software. The Ninth Circuit's summary states it is "affirming the district court's dismissal," holding that several competitors independently choosing the same pricing software, followed by higher prices, does not by itself show an anticompetitive agreement. The US Department of Justice's Antitrust Division appeared as amicus curiae; the opinion does not state which side the government argued for. Decided August 15, 2025 (No. 24-3576, 9th Cir.).
  • Cornish-Adebiyi v. Caesars Entertainment: Third Circuit revives an Atlantic City hotel pricing algorithm suit. A parallel Sherman Act Section 1 suit against Atlantic City casino hotels that used the same Cendyn Group pricing software was dismissed by the district court in 2024. On July 29, 2026 the Third Circuit reversed that dismissal and remanded, writing that it will "reverse the District Court's dismissal of the Complaint," and holding that the complaint's well pleaded allegations of a hub and spoke agreement run through Cendyn's shared pricing software are sufficient to proceed (No. 24-3006, 3rd Cir.). The opinion cites Gibson v. Cendyn Group, the Ninth Circuit's ruling in the Las Vegas Strip case, without disagreeing with it: the two courts reached different outcomes on different complaints with different allegations, not opposite readings of the same rule.

Frequently asked questions

How much RevPAR gain can an AI revenue management copilot deliver?
It depends heavily on the starting point. RIMC Hotels & Resorts Group reports a 28.44% RevPAR increase at its Polish property, with no period or baseline stated for that figure, and, separately, a RevPAR increase across its whole portfolio compared to the previous year since adopting Duetto in 2022; Duetto's case study reports 33% total hotel RevPAR growth at Hôtel Swexan, 2025 versus 2024. Treat both as single case study results rather than a guaranteed uplift: RIMC moved from pricing it describes as fully manual, adjusted through the property management system, with no revenue management system in place before 2022, and the Swexan case study does not state when its own Duetto deployment began relative to that comparison. Neither hotel's own case study calls its pricing AI or machine learning; both run Duetto's revenue management platform, which Duetto's own product page describes as pairing data with "AI-driven rate recommendations".
Does the AI set prices on its own?
It depends on the deployment. Hôtel Swexan's case study describes automated pricing strategies that execute around the clock "without manual intervention," based on occupancy thresholds and segment logic the Director of Revenue, Jessica Schiele, defines in advance, with no need to change rates by hand. The oversight in that setup is Schiele setting and revising those rules and running a daily check the case study describes as a few minutes rather than a few hours, not a person approving each price. Setting an explicit rate floor, a ceiling and a maximum daily change before switching on automatic pricing, as the implementation guidance on this page recommends, is good practice, but neither cited deployment states those specific limits publicly.
What data does a hotel need before it can use one?
At least a year or two of historical occupancy, rate and booking pace by room type and segment, a defined competitor set with rate shopping data, and, for group profitability scoring, food and beverage and function space data.
Is this the same as a hotel guest chatbot?
No. A revenue management copilot is an internal tool for the revenue team that sets prices; it does not talk to guests. A guest facing booking or service assistant is a separate use case.

How to cite this page

Blits.ai AI Use Case Library, "AI copilot for hotel revenue management", last verified 28 September 2026, https://www.blits.ai/ai-use-cases/hotel-revenue-management-copilot. Licensed under CC BY 4.0. Method: how we verify use cases.

Changelog
  • 28 September 2026: First published

Related use cases

Insurance

AI copilot for insurance pricing and actuarial analysis

AI that speeds up the work of pricing and actuarial teams, from automated, transparent risk and demand model building to natural language analysis of rate filings, experience data and reserving diagnostics, while actuaries select the models, sign off the rates and own the professional judgment.

Deployments
5 public, best grade B
Reported productivity gain
5x
Generali France, organization claim
Capital marketsWealth and asset management

AI assistant for deal sourcing and M&A due diligence

An AI assistant that screens the market for acquisition or investment targets, builds company profiles, and speeds up due diligence by reading data room documents, extracting key terms and risks and drafting the investment or diligence memo, for the deal team to verify and decide.

Deployments
4 public, best grade B
Reported productivity gain
up to 80%
Datasite, vendor claim
BankingCross industry

AI cash flow forecasting for corporate treasury

Machine learning and conversational analytics, offered by some banks inside their cash management platforms, that categorise a company's cash flows, forecast positions across accounts and currencies, and answer treasurers' questions in plain language, so the treasury team decides on funding and idle balances with better information and less spreadsheet work.

Deployments
5 public, best grade B
Reported productivity gain
about 90%
JPMorgan Chase, organization claim
Telecommunications

AI for mobile network planning and capacity optimization

Machine learning that forecasts where and when a mobile network will run out of capacity, recommends where to add cells, spectrum or hardware, and continuously tunes radio parameters so existing capacity carries more traffic, with planners approving investments and major changes.

Deployments
5 public, best grade B
Reported cycle time reduction
at least 95%
Deutsche Telekom, organization claim