Signs Your Marketing Tracking Is Broken

Most owners expect broken tracking to look like an error message. It doesn't. Broken tracking produces reports that arrive on the same day every month, in a consistent format, with figures that are technically accurate. That is what makes it expensive: nothing about it asks to be questioned.

Here is a plain definition worth keeping. Broken tracking is reporting that measures its own activity rather than money moving toward the business. The arithmetic can be perfect while the meaning is absent. Below are the signs, all of which you can check yourself, without knowing a single marketing term.

One: the numbers only ever go up

Real commercial figures move in both directions, because real markets do. A report where every line is up on last month, month after month, is usually not describing a business — it is describing a set of counters that only increment. Cumulative totals, follower counts, all-time impressions. Ask to see any figure that has fallen. If nobody can produce one, you are reading a scoreboard, not a measurement.

Two: nobody can name a customer who came from it

This is the fastest test there is. Point at any reported result and ask for one name — a company, a person, an order number — that arrived through it. Not a percentage. A name. If the answer is that the system doesn't work like that, then the system does not connect to your customers, which was the entire point of building it.

Three: the good months don't match your bank statements

Put twelve months of reported performance next to twelve months of actual revenue and look at the shape. They will never match perfectly — there are lags, and some marketing pays off slowly. But if the strongest marketing months are not visible anywhere in the money, even later, something in between is not connected. This one takes half an hour and needs no software.

Four: every channel claims the same sale

When separate reports each take credit, add up what they claim. If the total credited revenue exceeds what the company actually invoiced, the reporting is double-counting — usually because each channel is measured in isolation and every one of them touched the buyer somewhere. This is not necessarily dishonest. It is arithmetic nobody has been asked to reconcile.

Five: the report changes shape when results are poor

Watch what happens after a weak quarter. If the metrics that were headline figures quietly move down the page, or a new measure appears that has never been mentioned before, the report is being managed rather than produced. A report designed to measure will keep the same shape whether the news is good or bad — that is what makes it a measurement.

Six: “leads” has never been defined out loud

Ask what counts as a lead in your reporting. A form submission? A phone call of any length? A download? Then ask who checks whether existing customers, suppliers and spam are excluded. Where nobody can answer, the number is a count of events on a website, and the word “lead” is doing work the data cannot support.

Seven: the person who built the tracking is the person it judges

This is the structural one. If the same agency or employee designs the measurement, produces the report and is paid partly on how it looks, that is not fraud — it is an incentive nobody set out to create. Every business separates the person spending from the person counting somewhere in its finances. Marketing is often the last place that separation arrives.

What to do when a number won't trace

An untraceable number is not proof that someone is deceiving you. Very often the tracking was simply never built to connect spend to sales, and no one decided to fix it. But it should never be allowed to stand in for a result. The rule to adopt is one sentence long: a figure that cannot be walked to money does not get reported as if it were money.

That single rule does more for a non-marketer owner than any course would, and it is the same instinct behind checking whether your marketing numbers are real in the first place.

• Ask for one figure in the last year that went down. Note how long the answer takes.
• Ask for one customer name that arrived through a reported result.
• Lay twelve months of reporting beside twelve months of revenue and compare the shape.
• Add up what every channel claims. Compare the total with what you actually invoiced.
• Ask for the written definition of a “lead”, and who excludes existing customers and spam.
• Check whether the person producing the report is paid on how it looks.

None of this requires you to become a marketer. It requires the question you already ask about every other cost in the business, which is what not checking your tracking quietly costs you, and the argument behind Reports, Not Revenue.