Executive Summary
Your ad platforms, like Google and Meta, are all reporting credit for the same booking. This makes it hard to know your actual return on ad spend, and often leads to tricky conversations with the finance team.
You might be spending money on channels that seem to be working but aren't. It's difficult to put forward a budget you can stand behind when the numbers come from the platforms themselves.
The answer isn't usually another attribution tool. It's about changing the technical setup to bring all your event data into one place. From there, you can build a single, agreed-upon set of rules for attribution that you control.
The common problem: your ad platforms all claim the same sale
It's a familiar scene. You report that the £50k spent on Google Ads and the £40k on Facebook last month both generated over £100k in bookings. The CFO looks at the profit and loss account and sees only £130k of total new bookings. The numbers don't match. A quiet settles over the room.
This isn't a small discrepancy, it's a sign of a deeper problem in how the data is structured. Even with a good team and modern tools, this happens. Sometimes, moving to the cloud just means you're making the same mistakes, only faster.
In my experience, this is a common challenge for businesses with a high volume of transactions, like those in travel or hospitality. Once you start spending across several channels, relying on the dashboards inside each platform becomes problematic. They are built to show their own value, which helps encourage you to spend more with them.
Why this happens: each platform uses its own rules
The issue isn't usually with the marketing team, but with the technical setup. Each ad platform tracks users on its own. It uses its own rules, like a 28-day window for clicks, and has a clear interest in claiming credit for any conversion it had a hand in. So instead of one clear picture, you're looking at several different versions of the story, each from a platform selling its own services.
Without a single, central place to see the whole customer journey, you have little choice but to use their numbers. This is what leads to the same booking being counted several times. This makes it very difficult to get a clear view of Channel Performance, which is the foundation for proper Data-driven Marketing.
A more reliable approach: building your own attribution model
From what I've seen, buying another attribution tool doesn't usually fix the underlying problem. It often just adds another set of numbers to the debate. A better way to resolve the arguments is to build your own Single Source of Truth for how you measure marketing. This is more of an engineering task than a marketing one.
Step 1: Bring all your raw event data together
The first step is to move away from relying on the reports inside each ad platform. This means collecting all the raw data: the clicks, impressions, page views, items added to a cart, and bookings from every source, and putting it all into your own data warehouse. We then piece these events together to build a single timeline of each customer's journey. This gives you a complete record of every touchpoint, separate from what the ad platforms report.
Step 2: Define your business rules in a semantic layer
With all the data in one place, the next step is to build a Semantic Layer. This is simply a place to define your business rules in code, so they are consistent and can be tracked over time. We'd work with your marketing and finance teams to agree on a single definition for an 'attributed booking'. Whether that's first-touch, last-touch, or something else is less important than the fact that everyone agrees on it, and it's applied automatically and consistently. This is how you build real Data Trust in your reports.
It's about a defensible model, not a perfect one
It's worth being clear: no attribution model is perfect. The aim isn't to find some absolute truth about why a sale happened. It's to build a consistent and defensible model that helps you make better decisions.
I've found that putting this in place is as much about people as it is about technology. For instance, if you move away from a simple last-click model, the results for different teams will change. The reported ROAS for performance marketing might look lower, while the contribution from the content team might become clearer. Making this kind of change requires support from senior leadership, as it prioritises a clear, shared picture over familiar, but separate, metrics.
The result: confident budget decisions
Once this setup is working, those weekly arguments about whose numbers are right tend to fade away. The conversation can shift from questioning the data to deciding how to act on it. You can plan a multi-million-pound budget with more confidence, because you own and understand the logic behind the numbers. You can present a single, consistent view of marketing performance to the board and explain the reasoning for your spending. It puts you back in control of your own data.