Why travel retail media operations stop scaling (and it's rarely strategy)

Ask an OTA why their retail media programme isn't scaling and you'll usually get a strategy answer. Not enough supplier budget, the wrong ad formats, or a sales team that hasn't found its pitch yet.

Sit in enough of these conversations and the same operational issues keep surfacing. The strategy is often fine. What tends to slow things down is that the retail media platform's CRM, ad server, and campaign reporting all sit in different places, each maintaining its own version of campaign information, held together by manual updates across all three.

We've written before about why OTAs need to modernise their co-marketing — the strategic case for why standing still costs you partners and revenue. This is the part that piece didn't get into: what fragmentation actually looks like inside the operation, and what fixes it.

What fragmentation looks like in reality

The warning signs aren't dramatic. They're usually businesses that have done the hard part (migrated onto a proper CRM, built a media kit, signed real supplier-funded media budget) and still can't tell you, without a manual pull, how a live campaign is pacing.

A platform moves its sales process into a modern CRM, and the CRM genuinely works. But inventory control, campaign management, and reporting were never connected to it, so every deal that closes in the CRM still has to be manually re-entered somewhere else before it can actually run. That re-entry step hasn't gone anywhere, it has simply shifted to just after the deal closes instead of before it.

Elsewhere, the automation exists but only covers half the job. Sponsored listings can be booked and served through the ad server automatically, but the reporting layer was bolted on afterwards, as a separate project, months later. Sales can sell self-serve. Nobody can yet show a client a live dashboard without exporting something first.

A simple test is this: if someone asks how a live supplier campaign is performing, how many systems does your team need to open before they can answer confidently? If the answer is more than one, your operating model is probably doing more work than it should.

None of this is a strategy failure. It's what happens when a retail media programme grows faster than the operating model underneath it.

"By the time people come to us they've usually proved there's demand. They've signed suppliers and built the proposition. The frustration comes when they realise the operational side still depends on people updating three different systems by hand," says Roy Stratford, Director of Retail Media Services at Platform 195.

Why the cost is bigger than it looks

If you answered ‘more than one’ to that test, here's what it's actually costing you.

The direct cost is obvious enough: someone's afternoon spent reconciling a spreadsheet against a CRM export and an ad server report. But that's not the expensive part. The expensive part is what fragmentation does to confidence.

A sales team that can't show live performance without exporting and reassembling it first ends up underselling what they actually have, because the story of  ‘amazing platform, slightly awkward reporting’ is harder to tell than either half on its own.

A supplier who gets a manually assembled report every month, instead of a live dashboard, quietly recalibrates how sophisticated they think the operation is, and prices their renewal conversation accordingly.

Suppliers don't renew because they received another dashboard, they renew because they have confidence the investment can be measured, explained and improved. Fragmentation doesn't just slow reporting, it makes that confidence harder to build.

That's manageable at ten campaigns but at two hundred campaigns it becomes the operating model.

And the same disconnect that slows day-to-day reporting is exactly what makes new inventory hard to launch. If turning on sponsored listings for a new category means someone manually configuring three systems instead of flipping a switch in one, the business will keep finding reasons why now isn't quite the right time to launch it.

More often than not, the real obstacle is the unbudgeted cost of a manual launch, not a flaw in the idea.

What actually fixes it

The instinct is usually to build more connective tissue, an integration here, an export-import script there. That helps at the margins, but it only treats the symptom.

The problem is that the systems are often working from different definitions of the same thing. Sales thinks a campaign starts when a deal closes, ad operations thinks it starts when creative is approved, finance thinks it starts when invoicing begins, and campaign reporting thinks it starts when impressions are delivered.

Until those definitions line up, integrations don't remove the inconsistency. They just move it around.

The businesses that scale cleanly usually make one surprisingly simple decision early. They decide which system becomes the authoritative record of campaign state, then make every other platform consume that information rather than maintain its own version.

In our experience, that authoritative record is often the ad server rather than the CRM. The CRM records the commercial agreement, but the ad server is where the campaign actually runs, making it the logical place to determine whether it's live, paused or complete.

Once that hierarchy is accepted, a CRM update triggers a change in the ad server rather than creating another independent record. Finance and reporting then read campaign status from the ad server instead of maintaining their own separate version of it.

The CRM records the opportunity and commercial agreement, the ad server owns campaign state, the reporting layer reflects campaign delivery, and finance invoices against completed activity.

That single decision removes far more manual work than another layer of integrations ever will. It's a smaller decision than it sounds like, and a much bigger one than it looks like from the outside — the difference between an integration that needs babysitting and one that just works.

"Most businesses don't have a technology shortage. They have too many systems all trying to manage the same campaign. Once one platform becomes the source of truth, a lot of the manual work disappears surprisingly quickly," adds Roy.

Building on what's already there

Fragmented retail media operations don't usually announce themselves. They just quietly set the ceiling on how fast a platform can grow, because every new supplier, every new format, every new market is another thing that must be manually reconciled across systems that were never meant to talk to each other in the first place.

The good news is that the hard part, the supplier relationships and the commercial model, is usually already there. What's missing is the connective layer that lets the rest of the business see and trust what's already working.

None of this requires replacing every system overnight. Most businesses already have much of the technology they need. The challenge is deciding which system owns campaign truth, and integrating around that rather than layering more manual process on top.

The goal isn't a perfect technology stack. It's an operating model that allows retail media revenue to grow without operational complexity growing at the same rate.

Platform 195 works with OTAs, tour operators, and airline holidays businesses to bring retail media operations into a single connected system through Uplift. To talk through what that would look like for your platform, get in touch.

‍

‍