Meta ads for travel: the creative advantage most hotels and OTAs waste

Travel owns more creative raw material than any other category in advertising. Every property, cabin class, season and traveller type is its own reason to buy, while a fashion brand has to manufacture that kind of variety from a narrow product set. On paper, no category is better set up for how Meta now works.

Since Meta Andromeda rebuilt how Meta matches ads to people, the platform works out who should see each piece of creative, rather than serving your creative to an audience you define. It can only do that with the material you give it. The more genuinely different your ads, the better it gets at finding the right person for each one. This is what creative is the new targeting means in practice: your creative range now does the audience selection that targeting settings used to do.

Yet most hotel, OTA and airline accounts show up to Meta’s ad auction with a handful of hero assets built around a brand line, plus a catalog feed carrying a price, a date and one functional photo of a room. The variety exists in the business. Most of it never becomes an ad, so Meta never sees it.

There are three reasons why, and they need different fixes. The first is strategic: creative gets briefed as variants of one message rather than permutations of a structure, so volume arrives without variety. The second is technical: the catalog feed sets the ceiling on what dynamic formats can produce, and most travel feeds are thin. The third is operational: creative production runs on a monthly cycle in a category where demand moves weekly and peaks are known months in advance. This guide covers all three, then sets out how to build an account that gives Meta the range it is asking for.

Key findings

• Meta now finds the right person for each piece of creative rather than the right creative for a defined audience, so the range of creative you supply sets the ceiling on delivery.
• Travel owns more natural creative variety than any category in advertising (every destination, property, season and traveller type is a distinct asset), yet most accounts flatten it to a few hero ads and a thin catalog feed.
• Volume only works when it’s structured. Travel can brief across at least five dimensions (inventory, calendar, booking context, customer, message), each one multiplying the rest, and every permutation keeps its coordinates so a win tells you where to build next.
• Dynamic Ads for Travel are only as good as the feed behind them. Enriching the catalog improves every ad it will ever generate, which makes it the highest-return work in the account.
• The UK booking calendar (January peak, Easter, summer, October half-term, Christmas) is a creative-refresh trigger you can see coming, yet most accounts brief seasonal creative as the peak arrives, when CPMs are highest.
• Travel’s ~60-day booking cycle breaks standard attribution windows; measuring on last click alone defunds the activity that creates demand.
• Search referral to travel sites fell 20% year on year (SimilarWeb, 2025); Meta is a highly incremental channel to recover it (Haus’s meta-study shows +19% average lift to primary KPIs).
• Producing creative weekly instead of monthly turns 12 optimisation cycles a year into 52.

Permutations, not variants: brief across the dimensions travel already owns

The fastest route to creative volume in travel is combination, not more production. But what you combine matters more than how much. A variant is a copy with a tweak: same idea, new headline. A permutation is a distinct arrangement drawn from a structured set, and Meta rewards the second, because genuinely different inputs are what its matching runs on.

The structured set is the point. Most categories have two or three dimensions they can genuinely vary. Travel has at least five:

But the inventory grid multiplies the surface of your ads, not the reasons to buy. Rotate destinations under one message and you’ve made many versions of the same argument in different photography.

Personas multiply every other dimension

Trip type describes the holiday. A persona describes the person buying it, which makes it a separate multiplier: the same beach week is a different ad for the couple taking it and the family taking it.

The persona evidence already exists in your booking data. Party composition, lead time, length of stay, rate band and repeat behaviour describe your customers more precisely than any research exercise. In most travel businesses the clusters resolve to something like the short-break couple, the school-holiday family, the solo traveller and the premium upgrader, and the ones specific to your business are worth more than the generic set.

As Nest’s lead creative strategist Maddie O’Reilly puts it, a persona strategy starts with the who: “you want someone to look at an ad and say, ‘that’s me.'” A product strategy is different, treating the creative as “an answer to a question”, built around use case. Most travel brands need both, dialled up or down by season.

Run each persona’s creative broad and let the platform do the segmenting. People who recognise themselves engage, Meta gets sharper with every signal, and the creative ends up doing the audience selection for you.

The UK booking calendar is a refresh trigger

Every travel marketer knows the pressure points of the UK booking year: the January peak, Easter, the summer window, October half term, Christmas. The demand curve is visible months out, which makes travel one of the few categories where creative can be ready before it’s needed.

Very few accounts use that visibility. The usual pattern is that new creative gets briefed as the peak arrives, produced during it, and lands as demand is already falling away, while CPMs sit at their highest and the account is defending them with ads that have been in market for weeks.

Treat those dates as production deadlines, not campaign dates, and bank seasonal creative ahead of them. You go into the peak with depth instead of a thin set of tired assets, and the platform has room to find the right message for each person while the demand is actually there.

Dynamic Ads for Travel: the catalog feed is the creative bottleneck

Dynamic Ads for Travel generate creative automatically from your catalog, which makes them the most scalable format in the category and also the most misunderstood. Because the system produces the ads, the quality of the feed decides the quality of the ads. In most accounts the feed is maintained as a data file by whoever owns the booking-engine integration, and it’s the weakest creative asset in the account.

Most hotel and OTA catalogs carry the operational minimum: property name, price, availability, one exterior or room photo. Every advertiser is drawing on similar imagery from similar sources, so the dynamic inventory converges on the same look. The format can produce thousands of variants, and it produces thousands of variants of nearly the same thing. A campaign running on that feed is doing exactly what the material allows.

Enriching the feed improves every ad it will ever generate

So what does a good feed look like? Treat the catalog as a creative surface and work through it in this order:

1. Replace functional imagery with lifestyle imagery on the properties carrying the most spend. Room and exterior shots tell a customer what they’re buying; lifestyle imagery tells them why.
2. Add seasonal image sets so the same property presents differently in February and July. One property, several looks, no new brief each season.
3. Map creator and UGC content to properties. Travel has more customer footage available to it than almost any category, and most of it never reaches the feed.
4. Add short video where the property or destination supports it.
5. Apply generative variation across the long tail: destination-specific backgrounds, price and urgency treatments, image enhancement. This is how properties outside the top tier get creative depth at all.

Do the feed work before you produce anything new. Enriching the catalog improves every dynamic ad it will ever generate, across every campaign, with no per-property brief. It is the one job in a travel account you do once and benefit from everywhere.

Creative AI at catalog scale

Enrichment used to be limited by production cost, which is why it rarely happened beyond the top few properties. That constraint has moved. Generative tools can now produce destination-specific backgrounds, price and urgency treatments and image enhancement across a whole catalog. For an OTA carrying ten thousand properties, one usable image per property becomes five to ten contextual variants. That is a scale of creative diversity you could not reach manually at any budget.

Travel gets more from this than other verticals do. A brand with a narrow product range can generate variants and still be showing the same object. Travel’s inventory is genuinely different property to property, so the variants are genuinely different ads.

Measuring travel Meta ads when the booking cycle runs to 60 days

Travel buys slowly. Consideration runs across weeks of comparison, price-checking and negotiation inside a household, and the click that gets credited is rarely the interaction that created the demand. Platform attribution windows were not designed for a 60-day average buying cycle, and in travel they systematically under-credit upper-funnel activity.

Report against those windows alone and the outcome is predictable. The channels that touch people early look inefficient, budget moves to the channels that catch demand at the end, and the demand being caught slowly stops arriving. You need incrementality measurement to see what the last click cannot show. Geo-lift and MMM are the usual options, layered onto the reporting you already run.

There is a second reason this matters now. Search referral traffic to travel sites fell 20% year on year (SimilarWeb, 2025) as AI answer engines resolve the research queries that used to end in a click through to an OTA or hotel site. That demand has moved earlier and onto other surfaces, and social is one of the places it is going. Meta is well placed to catch it: Haus’s meta-study of incrementality tests found an average +19% lift to primary KPIs. If you are watching search volumes soften, build the demand through always-on prospecting instead of waiting to intercept it. That puts more weight on creative that works without an existing intent signal.

Why Advantage+ adoption lags in hotel and OTA accounts

Consolidated, broadly targeted campaigns generally beat granular manual structures in travel, delivering lower CPA. Adoption in the category is still low, and the reason is rarely a failed test. Travel accounts are built around destination-level budget control, and consolidation asks you to give that up.

The concern is legitimate. Destination budgets are how travel teams manage inventory, margin, contracted rates and the properties that need volume this month. Handing that to an algorithm feels like handing over commercial decisions, not media ones. Consolidation still wins on performance, though, so the control has to move somewhere else in the account. The build sequence below ends at exactly that point.

Building the account: nine steps from audit to weekly cycle

Travel’s creative advantage is real, and the operating model most teams run cannot express it. The sequence below fixes that. Steps one to three are strategy; four to six are the build; seven onwards is the operating rhythm, and it decides whether the rest holds.

1. Audit what the account is actually showing people. Count distinct ideas live, not ads live. Sort the last 90 days of assets into concepts, where two ads are the same concept if a customer seeing both wouldn’t register them as different propositions. Most accounts land far below what they expected, because destination imagery rotating under a fixed brand line reads as variety in reporting and as repetition in the feed. Run the same check on the catalog.
2. Build the personas from booking data, three to five of them, each a page: the motivation, the barrier, the creative cues, and the recognition test, the moment the customer thinks “that’s me.” Map them with the brand team, who will correct a persona faster than the data will.
3. Build the multiplication grid (destination × trip type × season, run through the personas from step two) and prioritise on commercial need instead of trying to fill every cell.
4. Fix the feed before producing anything new. It’s the highest-return work in the account and it’s almost always skipped because it doesn’t look like creative work.
5. Template the personas so production stops being linear. One template per persona holds the visual strategy constant while destination, price and date sit in dynamic slots underneath; generative tooling varies text and background inside the template. If templating collapses the personas into one look with different copy, you’ve rebuilt the original problem with better tooling.
6. Bank the season before it arrives, treating the start of each peak as the production deadline.
7. Consolidate last, after steps two to five. Once each persona has real creative depth behind it, you steer the account through what you build and how much of it. A shoulder season that needs filling gets its own persona treatment and creative weight, and the platform spends against it because the creative is finding people who respond. You keep control of commercial priorities without paying for it in efficiency. Do not consolidate a thin account; broad targeting with four hero assets removes the structure and adds nothing, which is where the category’s scepticism about consolidation came from.
8. Set measurement up for the long cycle before you scale spend: an incrementality method, a booking-window view held alongside your reporting cadence, and a prospecting read that doesn’t depend on last click.
9. Run the weekly cycle. Read what ran, convert the read into the next distinct brief, generate against it through the persona templates and enriched feed, sign it off, ship, and start the next read.

A team producing creative monthly gets twelve chances a year to learn something and act on it. A team producing weekly gets fifty-two. In travel the gap is wider than the arithmetic suggests, because twelve cycles means roughly one learning cycle per peak, and it lands too late to change that peak. Fifty-two means you are reading January while it is still open, and building Easter from what January actually taught you.

How Nest closes the gap

Travel already owns more natural creative variety than most categories in advertising. The work is building an operating model that turns it into ads, and that is the work Nest does.

Nest is a technology-led performance and creative agency working with brands like Urban Outfitters, Hims and Damson Madder. Our approach has three parts: a senior strategist who bridges brand and performance and turns customer research into the angles creative is built on; aggressive creative volume and variety across studio, creator and AI-assisted formats; and Nest Creative AI, our end-to-end performance and creative solution, whose automated ad analysis and creative insight close the loop so what performs shapes what gets made next, with human sign-off built into the workflow at every critical production stage. The final call stays with your team.

None of the three parts is the answer on its own. Run as one loop, the gains compound, because every cycle starts from a higher base than the last, and the account gets faster every week the loop runs.

FAQs