The Plumbing Problem: What Holds Back AI in the Hospitality Industry

Share on

The fastest way to fail with AI is not choosing the wrong tool. It is feeding the right tool the wrong data.

Every hotel conference this year has an AI panel. Chatbots, dynamic pricing and personalised offers fill the agenda. Very few sessions ask what the AI is actually fed.

For owners and marketing managers, the pattern is familiar:

• A new tool arrives with a strong demo.

• Six months later, the monthly review still opens with "which number is right?"

• Marketing, revenue and distribution each report a good month. Profit barely moves.

That gap sits below the tools. It is the plumbing: how data moves between your systems and your teams. It decides whether AI in hospitality improves hotel commercial performance or simply adds another dashboard.

AI in the Hospitality Industry Has an Adoption Gap

Adoption is no longer the question. Impact is.

• McKinsey's 2025 State of AI survey found that only 39% of respondents attribute any EBIT impact to AI, and most of those put it below 5% of EBIT. McKinsey & Company

• Gartner reports that 63% of organisations either lack, or are unsure they have, the right data management practices for AI. Gartner

• Gartner also predicts that through 2026, organisations will abandon 60% of AI projects that are not supported by AI-ready data. Gartner

Read together, the message is plain for any owner weighing an AI budget. The models are ready. The data underneath them often is not.

AI in the hospitality industry faces this constraint with an added burden. Hotel commercial data sits across more systems, is owned by more teams and updates at different speeds. That makes the plumbing question sharper for hotels than for most sectors adopting AI in hospitality today.

What "Plumbing" Means in a Hotel

Too many systems, too few connections

A typical commercial stack includes:

• a PMS holding reservations and guest records

• a booking engine and channel manager

• a revenue management system

• GA4, plus Google and Meta ad accounts

• a CRM that few people open weekly

Each system works on its own. Few of them share data cleanly.

Skift Research reports that 63% of hotel tech budgets still go to maintaining legacy systems, many of which are not AI-compatible or cannot support complex data integration. Every rupee, baht or dollar spent keeping old systems alive is money not spent connecting them, and hotel commercial performance pays for it. Skift

Four teams, four scoreboards

• Marketing optimises ROAS.

• Revenue optimises RevPAR.

• Distribution optimises channel mix.

• Loyalty optimises enrolment.

All four can hit target in the same month while the hotel loses margin.

A campaign drives low-cost clicks into dates that were already selling at full rate. A promotion fills rooms that would have filled anyway. Each team's report looks healthy.

This is where hotel commercial performance leaks: in the gap between scoreboards, where no single team is accountable for the whole. We covered the board-level view of this in Siloed Hotel Data: Why Commercial Intelligence Is Now a Board Question.

Numbers nobody fully trusts

Ask three systems how many bookings last month's campaigns produced. Expect three answers.

• Ad platforms count conversions by their own rules.

• GA4 counts sessions that ended on a confirmation page.

• The PMS counts stays that arrived and paid.

Until those three figures are reconciled, nobody can say with confidence which campaign paid for itself.

When the numbers disagree, meetings turn into reconciliation sessions. Decisions wait. Any hotel marketing strategy built on those numbers inherits the disagreement.

Why AI in Hotels Fails on Bad Plumbing

AI amplifies whatever it is fed

AI does not repair inconsistent data. It scales it.

• If "direct booking" means one thing in GA4 and another in the PMS, the model learns both definitions.

• If cancellations never flow back to the ad account, the model keeps rewarding campaigns that produce cancellations.

• The output looks confident. It is still wrong.

Much of the failure of AI in hotels starts with definitions. Our hotel commercial glossary: seven terms your AI tools cannot read shows where the common ones break.

Attribution breaks between click and stay

Marketing sees the click. Revenue sees the stay. The path between them is rarely visible to either.

• Most hotel marketing strategy reviews stop at cost per acquisition.

• The net value of the stay, after commission, cancellation and discount, seldom reaches the marketing team.

• Without that value, AI in hotels can only optimise for booking volume.

We look at the exact points where tracking fails in Where Does Hotel Marketing Attribution Stop Working?

The guest has already moved on

Deloitte's 2026 Travel Industry Outlook notes that generative AI use is rising in travel shopping, even though commerce and content are not yet fully connected. Deloitte Insights

Guests are adopting AI in hospitality, through search and booking, faster than most hotel back-ends are connecting. AI in the hospitality industry will keep moving on the demand side whether hotels are ready or not.

Fixing the Plumbing: A Hotel Commercial Strategy Checklist

Profit is now the board's first question. Deloitte's 2025 European Hotel Industry and Investment Survey found that maintaining or increasing profitability had become executives' top priority. Deloitte

Before the next AI purchase, work through this list. Most of it needs discipline more than software, and each step lifts hotel commercial performance on its own.

• Reconcile monthly. Compare bookings reported by ad platforms and GA4 against arrivals in the PMS. Track the gap as a number.

• Agree one set of definitions. "Direct booking", "net revenue" and "incremental" should mean the same thing to marketing and revenue.

• Label every metric's source. Your team should know whether a figure came straight from a system, was calculated, or was estimated.

• Judge every lever in one currency. Paid media, rate changes and promotions all end up as net margin. Measure them that way, and your hotel marketing strategy starts speaking the same language as revenue.

• Name one owner for the numbers. One person should be accountable for the full reconciliation across marketing, revenue and distribution.

• Hold one weekly decision meeting. Marketing and revenue in the same room, looking at the same numbers.

• Buy AI last. Once the above holds, AI in the hospitality industry has something reliable to work with.

A hotel commercial strategy built on this checklist makes every future tool more useful, including the ones you already pay for.

Where pulse. Fits

This is the problem pulse. was built to solve, and the reason AI in hospitality needs a connected foundation first.

pulse. by dhi Hospitality brings marketing and revenue signals together in one commercial brain. Every commercial signal contributes to every commercial decision.

• It reads from the systems you already run: PMS, booking engine, ad accounts and GA4.

• It reconciles marketing-reported bookings against actual stays in the PMS.

• It expresses marketing and revenue decisions in the same measure: net margin.

• Every figure carries a label showing where it came from.

For owners, hotel commercial strategy becomes one conversation, built on one set of numbers. For marketing managers, hotel marketing strategy connects directly to the revenue it produces.

Explore pulse.

Start With the Pipes

The hotels that get real value from AI in hospitality will be the ones whose data agrees with itself. Tool count matters far less.

• Connect the systems you already own.

• Put marketing and revenue on one scoreboard.

• Then let AI in hotels do what it does well.

That sequence is a hotel commercial strategy decision before it is a technology decision, and it is the fastest route to better hotel commercial performance.

Most hotels discover their biggest commercial leak before they discover their biggest growth opportunity.

Get early access to pulse. and see where revenue, marketing and distribution data stop aligning: pulse.dhihospitality.com