In Brief: Paige Lopez introduces Ernest, a conversational AI platform that lets hotel marketers query and analyze direct-channel data, enabling faster decisions and better campaign performance.
It’s Monday morning and your general manager asks a simple question: why did direct bookings dip this week? You know the answer exists. Part of it sits in your booking engine reports, part in your rate shopper and part in your personalization dashboard. Pulling it together means a request to a colleague, a spreadsheet and a wait. By the time the picture is clear, the week has moved on.
Picture a different Monday. You type the question into one workspace. The answer comes back with the data behind it, a likely cause and a suggested next step. If it helps, the email to your OTA account manager or the report for your team is already drafted.
That’s the kind of question Ernest, Lighthouse’s AI teammate, is built to answer. Here’s how he handles the ones hotel marketers ask most often, and how that differs from using a general-purpose AI tool.
Key takeaways
-
Ernest is Lighthouse’s AI teammate. You ask a question in plain language and get an answer grounded in your own data, with context from 80,000+ hotels
-
Your direct channel data sits next to rates, distribution and demand, so marketing decisions reflect the full commercial picture
-
Ernest helps you see which OTAs undercut your direct rate, find where direct bookings leak and judge which personalization campaigns pay off
-
Unlike generic AI, he’s pre-loaded with your commercial dataset, works out of the box and lives on the platform your team already uses
Meet Ernest
Ernest is Lighthouse’s AI teammate for hotel commercial teams. He creates one AI Workspace where you ask a question in plain language and get an answer grounded in your own data, with links back to the numbers behind it.
He’s pre-loaded with your commercial dataset, so your first conversation delivers answers, not generic advice. And he draws on the same intelligence that powers 80,000+ hotels across 185 countries, so every answer comes with market context. Your direct channel data, from Lighthouse Direct, sits in that same workspace.
For marketers, that means the direct channel stops being a silo. You can see it next to rates, distribution and demand, and ask questions that cross all of them.
Where do you actually stand against your comp set?
Most hotels optimize their website without knowing how it compares. Is your booking engine conversion strong for your market, or only strong for you? Is an OTA quietly beating your direct rate?
Ask Ernest, and he shows you.
Ask Ernest which OTAs are undercutting your direct rate and how much revenue is at risk, and he starts with what your disparities data shows: which channels undercut you, how often and what it’s costing you. Then he reads the situation. For example, if Price Match is enabled on your website to match OTA undercuts in real time, Ernest shows you where it’s recovering that revenue and where it falls short.
What he recommends next depends on what he finds. If disparities run deeper than your Price Match is set to cover, he’ll suggest adjusting those settings and raising it with the OTA, and he’ll offer to draft the email to your account manager with the details and frequency of the disparities. If the undercutting is occasional and already covered, he’ll tell you it can wait. The answer changes with the situation, and so does his advice.
Try questions like these:
-
Compare [Property]’s Direct Booking Index this month vs. last month
-
Which OTAs are undercutting our direct rate, and how much revenue is at risk?
-
How does our booking engine to booking conversion compare on mobile vs. desktop?
The second question matters most. Undercutting is easy to suspect and hard to size. Putting a revenue figure next to it turns “we think OTAs are cheaper sometimes” into a case you can take to your revenue manager or the OTA. The third one often settles a long-running debate about where to invest: the mobile experience or the desktop funnel.
Where are your direct bookings leaking?
Traffic is up, bookings are flat. Everyone has an opinion. Fewer people have the data to settle it.
Ask Ernest a broad question like “How can we improve our direct booking performance compared to the market?” and he starts where a good analyst would. He compares your conversion funnel against your comp set and similar hotels in the Lighthouse network, one stage at a time, so you can see exactly where you fall behind. Is the problem getting visitors to the booking engine, or getting them to finish once they’re there?
Then he keeps going. If the numbers point to a drop-off, Ernest suggests the obvious next step, such as pulling a price disparity report to check whether OTAs undercutting your rate explain it.
That’s the pattern worth noticing. One broad question locates the leak, and the follow-up tests a likely cause. No exports, no stitching together reports.
Once you know where the leak is, you can test decisions against what happened afterward. Four scenarios worth running:
-
After a rate change. “We raised our BAR 5% last Tuesday. Has the Disparities Index moved since then?” You see whether a pricing move helped or hurt your position against OTAs, without waiting for the monthly review.
-
After a contract change. “We signed a stricter parity clause with Booking.com back in June. Has disparity frequency actually dropped since?” Now you know whether the clause is doing its job.
-
When paid traffic doesn’t convert. “We’re getting a ton of visits from the new paid social campaign but hardly any bookings. Where’s it breaking down?” Instead of guessing at the landing page, you see the stage where visitors drop.
-
When you’re deciding on promos by market. “Is it worth running different promos for US vs. UK visitors, or is our booking engine conversion basically the same for both?” If the numbers match, you save the effort of building two campaigns.
Each of these used to be a mini research project. Today they’re a single question.
Is your personalization paying off?
Personalization earns its budget only if you can show what it did. That’s harder than it sounds when campaign data and booking data live in different places.
Ask Ernest which campaigns are performing best and he does the ranking for you. He pulls the personalization data, weighs revenue against conversion rate and shows you what’s working. He also goes beyond the leaderboard. The biggest earner isn’t always the strongest performer, and Ernest flags the campaigns that convert well, lift average booking value (ABV) and fill more room nights, even when they reach fewer visitors. That’s the kind of detail a revenue total alone hides.
When you need to take it to the team, he offers to turn the findings into a quick report you can share.
Ask Ernest:
-
Which of our campaigns are performing best this month?
-
Which of our messages should we pause or refresh because they’re underperforming?
-
How many bookings are our Exit messages actually saving before people leave the site?
-
How many bookings did the SUMMER25 code drive this month, and what’s the average value on those bookings?
The last two are the ones that tend to land in leadership meetings. A figure for bookings saved by Exit messages answers the question,”Is this worth running?” A figure for SUMMER25 tells you whether the promo code drove valuable bookings or just discounted ones you’d have won anyway. Hotels using personalization see direct booking conversion lifts of 15–30% on average, and this is how you show your share of it.
How do AI answers describe your hotel?
Guests increasingly start with an AI answer instead of a search results page. If your hotel is missing from those answers, or described poorly, you won’t see it in your web analytics. The visit never happens.
Two questions to start with:
The second question separates two very different fixes. If the cause is your own content, you rewrite your pages. If it’s how others describe you, you work on reviews, listings and partner content. Knowing which one saves months of effort on the wrong problem.
Why Ernest and not generic AI?
General-purpose AI tools are useful for drafting, summarizing and quick analysis, and your team will keep using them for that. Decisions about your direct channel are different. These tools reason over public internet knowledge and whatever you paste in. They haven’t seen your rates, your pace or your comp set, so they can answer a pricing or distribution question with full confidence and no data to back it up. The difference is grounding, not intelligence.
Four differences matter to a marketing team:
-
Decades of expertise, built in. Lighthouse revenue managers build and refine his skills for pricing, distribution and wider commercial workflows, so you don’t spend time teaching him what experienced commercial teams already know
-
Pre-loaded with your data. He starts with your commercial dataset and a governed data layer with fixed definitions, so the first conversation delivers answers, not generic advice, and two people asking the same question get the same answer
-
Ready out of the box, tailorable when you want it. He works from day one and adapts when you want it your way
-
Already where your team works. He lives on the platform your team uses every day. No new logins, no data migration. A general-purpose tool needs custom connections to your systems that someone has to build and keep running, and Lighthouse carries that work for Ernest
Each answer links back to the underlying data, so you can check the reasoning before you act on it. And he doesn’t wait to be asked. Set up a routine and he runs it on schedule, flagging things like a parity break or an underperforming campaign. For the full comparison, read our breakdown of Ernest vs. generic AI.
Frequently asked questions
What is Ernest?
Ernest is Lighthouse’s AI teammate for hotel commercial teams. He creates one AI Workspace where you ask questions in plain language, get recommendations grounded in your data and see the numbers behind every answer.
What can hotel marketers ask Ernest about the direct channel?
Anything that connects your direct channel to the rest of your commercial picture. Common questions cover direct rate undercutting by OTAs, where visitors drop off in the booking funnel, how your conversion compares with your comp set, which personalization campaigns perform best and how your hotel shows up in AI answers.
How is Ernest different from general-purpose AI tools?
General-purpose AI doesn’t know your data, your market or how hotel distribution works, so its answers can look right and still be commercially wrong. Ernest carries hospitality expertise that’s codified and refined continuously, comes pre-loaded with your commercial dataset and draws on the Lighthouse network of 80,000+ hotels across 185 countries.
What direct channel data can Ernest see?
Ernest works with your Lighthouse Direct data, including conversion at each stage of the booking funnel, comp set benchmarks, price disparities against OTAs, personalization campaign performance and AI visibility. He reads it alongside your rates, distribution and demand data, so you can ask questions that cross all of them.
How is asking Ernest different from using the Lighthouse Direct dashboards?
Dashboards show you the numbers once you know where to look. Ernest takes a question in plain language, pulls the data from across your direct channel and tells you what it means. He then suggests a next step, such as checking a price disparity report or drafting an email to your OTA account manager.
Can Ernest change my Price Match settings or personalization campaigns?
Ernest recommends by default. He might suggest adjusting your Price Match settings or pausing an underperforming message, but you decide. Autonomy stays off until an administrator turns it on, guardrails are set at property and portfolio level and every action is logged.
Paige Lopez
Paige is a content marketer with a passion for storytelling and helping hotel brands grow their direct channel. With a background in hospitality marketing, she explores how personalized messaging, guest-centric content, and AI-driven tools can boost direct bookings by optimizing every stage of the user journey.
About Lighthouse
Lighthouse (formerly OTA Insight) is the leading commercial platform for the travel & hospitality industry. We transform complexity into confidence by providing actionable market insights, business intelligence, and pricing tools that maximize revenue growth. We continually innovate to deliver the best platform for hospitality professionals to price more effectively, measure performance more efficiently, and understand the market in new ways.
Trusted by over 65,000 hotels in 185 countries, Lighthouse is the only solution that provides real-time hotel and short-term rental data in a single platform. We strive to deliver the best possible experience with unmatched customer service. We consider our clients as true partners – their success is our success.
Source: View the original article at Lighthouse.