Direct answer
A tracking setup that's broken doesn't just fail quietly — it produces reports that look fine. Traffic looks healthy, engagement looks strong, the invoice looks justified, and the one number that actually matters — how many visitors became customers — sits uncounted or miscounted one layer down. The cost isn't a mysterious gap. It's every month you kept paying because nothing in the report gave you a reason to look closer.
Most business owners assume the risk is a bad marketing report. It isn't. The risk is a report that looks fine — because a good-looking number is the one thing nobody in the building has a reason to question.
That's not an accident of a badly built dashboard. It's the direction the incentive runs. The person building the report is judged on the report looking good. The person reading it wants it to look good too — that's easier than reopening the last six months of spend. Nobody in that arrangement gets rewarded for finding out the number is wrong. So it stays wrong, for as long as the invoices keep clearing.
Why the number that gets checked never fails
Ask most marketing agencies how a campaign performed and they will show you a dashboard they built, filled with numbers the ad platform calculated, describing spend the ad platform benefits from you increasing. That's not dishonesty. It's structure. Nobody in that chain has a professional reason to go looking for the number that would make the invoice harder to justify.
That's the actual mechanism behind this piece: a broken tracking setup doesn't just hide a problem. It actively produces the report that keeps the arrangement going. Strong traffic. High engagement. A healthy click-through rate. All of it can be completely true, and completely beside the point, if the number underneath it all — how many of those visits actually became paying customers — was never being counted correctly in the first place.
| What the report shows | What it actually confirms | What it doesn't tell you |
|---|---|---|
| Website traffic | People arrived at your site | Whether they were the right people |
| Engagement (time on page, pages viewed) | People stayed and looked around | Whether staying led anywhere |
| Click-through rate | People clicked what you asked them to click | Whether that click ever became a customer |
| Conversion rate | Some visitors completed the counted action | Whether that action is really a sale — or counted correctly at all |
A B2B company, excellent engagement, and a 0.3% conversion rate
Here is what that looks like in practice, from a diagnosis I ran directly — not a benchmark to compare yourself against, just what one broken layer produced in one account.
From a diagnosis I ran: a B2B company selling a customized product had strong website traffic and excellent engagement data — time on page, pages per visit, the numbers an agency report likes to lead with. Its conversion rate was 0.3%. The cause wasn't the traffic, the creative, or the media spend. It was a 16-field contact form asking corporate buyers to specify exact product details before they had even decided whether they were interested. Once the form was cut down to what a sales callback actually required, conversion improved immediately — no new budget, no new campaign, just fewer fields standing between an interested buyer and a phone call.
The engagement numbers in that report were not wrong. People really were spending time on the site. They really were reading multiple pages. Every metric an agency report typically leads with was, on its own terms, accurate. The problem was sitting one layer beneath what was being reported — in a form nobody had thought to test from the buyer's side of the screen.
Good engagement numbers and a 0.3% conversion rate can both be true at the same time — and only one of them was ever going to show up in the monthly report.
The three places marketing actually breaks — and why this is the one nobody checks
When a marketing budget stops producing results, the cause is almost always one of three things. First, the tracking itself is broken or misleading — the numbers being reported don't match what actually happened. Second, a decision further upstream was never properly made — who you're actually selling to, what you're offering them, or why they should pick you over the next option, decided by no one and defaulted into whatever the last campaign assumed. Third, it's a genuine execution problem — the media buying, the creative, or the sales follow-up simply isn't good enough.
All three are real, and any one of them can be true in a given account. But the first is structurally different from the other two, because it's the one that hides itself. A bad decision about who you're targeting eventually shows up as a business owner's gut feeling that something is off. Weak execution shows up as underwhelming results everyone can see. Broken tracking shows up as nothing at all — or worse, as a result that looks good enough to keep funding it. That 16-field form sat inside a report that, on the surface metrics, looked like a mild success. It took someone actually walking through the form as a buyer would to find the real number underneath it.
What checking your own numbers actually looks like
You don't need a marketing background to do this. You need about forty-five minutes, your own systems, and a willingness to compare two numbers that are supposed to match.
- Pick one real, recent sale or signed deal. Trace it backward through your own CRM or invoice log to the first marketing touchpoint that produced it.
- Ask what your agency counts as a "conversion" — a form fill, a call booked, a signed contract — and check whether that matches what you'd personally call a sale.
- Compare the conversion number in the agency's dashboard against a number from a system you control — your CRM, your call log, your bank statement.
- Open the form, page, or call script that sits just before a "conversion" is counted, and fill it out yourself as a buyer would. Would you actually finish it?
- Ask what happens to the report if next month's number is worse — and whether the person answering that question is the same person who gets paid if it stays good.
If the two numbers don't match
That gap is not automatically proof of dishonesty. Sometimes it's a tracking script that broke after a website update and nobody noticed. Sometimes it's a form nobody tested from the buyer's chair. Either way, the fix is rarely more budget. In the case above, it was fewer fields.
This is the same discipline covered at more length in Reports, Not Revenue — the difference between a report that looks like progress and a number that actually moved. It connects to the wider question of what an unchecked marketing budget is actually costing you across a business, not just inside one report.
If you'd rather have someone else run that comparison — trace your own numbers against what's being reported, and tell you plainly what they find — that's most of what happens inside a Commercial Immersion Diagnostic, before another euro of new spend is committed.
Marc Wajsberg — Senior Marketing Strategist, X8 Agency. 30+ years across buyer psychology and commercial strategy, 150+ businesses guided. More about Marc.
Checking your tracking once doesn't fix an incentive that runs the wrong way every month after.
The Commercial Immersion Diagnostic costs €2,900, fixed, no retainer — and tracing your own numbers against what's being reported is part of it. It surfaces something concrete and actionable, or you don't pay for it.
Book an introduction call See what's included