Every ads dashboard you open has good news in it. Conversions up, cost per result down. A green line heading the right way. Nobody at the platform is faking the numbers. But there's a detail worth sitting with. The company selling you the ads is also the company telling you how well they worked. No other purchase in your business works like that. Your supplier doesn't send you a report on how much you enjoyed the delivery.
Why the dashboard always looks decent
Platforms count conversions in the way that flatters them, because the flattering methods are easy and the strict ones are hard. View-through conversions credit you with a sale when someone saw your ad and bought later without clicking, even if they were always going to buy. Attribution windows stretch far enough to sweep in sales that had other causes. Every conversion the platform can plausibly claim, it claims.
None of that is lying. It's measuring in the seller's interest, and the seller's interest is that the campaign looks healthy enough to keep funding. The tell is that nearly every campaign looks like it's working. A number that almost never goes down isn't a measurement. It's a sales tool with a chart on it.
The number that can't be flattered
Your revenue is the number that can't be talked up. Actual sales, in your own system, recorded where the platform can't reach. When the dashboard says conversions doubled and revenue is flat, you didn't buy customers. You bought attribution. When the dashboard looks modest but enquiries keep arriving, the ads are probably doing more than they admit. The only way to know which world you're in is to compare spend against what the business actually took in.
The trap is biggest with existing customers. Someone who already knows your name searches for it, clicks the ad at the top, and buys. The platform records a conversion. You paid to sell to a customer you already had. It shows up as a win in the report and as a margin leak in the accounts, and only one of those places tells you the truth.
Habits that keep the report honest
You don't need a data team for this. Four habits close most of the gap:
- Decide what a conversion is worth before you launch, then measure it in your own system. Platform numbers are the forecast. Your sales data is the weather.
- Watch branded search. If paid clicks on your own name climb while organic and direct traffic fall by the same amount, you're renting customers you already owned.
- Run a holdout when you can afford it. Pause a channel, or switch it off in one region for a few weeks, and watch total revenue rather than the platform dashboard.
- Re-report the same campaign under a shorter attribution window and see how much of the magic survives. The drop is usually educational.
This is not a case against paid ads
Paid ads work, and we run them every day. Good campaigns bring in customers who had never heard of you, and the results for those campaigns are real. The point is knowing what you're reading. The platforms automate more of your ads every year, and automation optimises for the platform's definition of success. That's fine as long as your definition is in the room too. Your definition is revenue, and revenue doesn't live in the seller's dashboard.
The honest part
The platforms aren't villains. Attribution is genuinely hard, and they make an honest effort at a fuzzy problem. But the effort is made by the seller, and you'd never accept that arrangement from anyone else. Trust the platform's trend. Distrust its total. When the two disagree, the bank account wins.
If your ads look like they're working but the revenue doesn't agree, that's worth pulling apart properly. Our paid team spends its life reconciling dashboards with bank accounts. Show us the gap and we'll tell you where the money is actually going.
