The audit made the numbers look worse. That was the point.

"My leads are flat. I just need better keywords."

An owner-operator who sells hard-to-find electronic components came to us with a clear and reasonable ask. Find better keywords, advertise more parts, maybe earn some free ranking so he could stop paying for every click. He had run his own advertising for years, and he had run it carefully. On paper the account looked alive, with hundreds of reported successful actions and years of steady investment behind them.

Underneath the request was the feeling every owner knows. Working the same effort, with a brand and a catalog he believed in, and watching the results sit flat anyway. Because the dashboard looked healthy, the natural conclusion was that the machine worked and just needed better fuel.

That belief feels safe because nearly everyone shares it. When the number on the screen looks good, you reach for more keywords, more parts, more spend. The catch is that more fuel poured into a gauge that reads wrong does not move the business. It just runs up the bill while the real result stays hidden. Safe is not the same as true.

What we found when we looked at the scoreboard

So before touching a single keyword, we opened the screen that decides what the account counts as a win. That is where the story was. He did not have a leads problem. He had a measurement problem he could not see past.

The account was treating phone taps, "get directions" clicks, and a nearly dead purchase action as if each were a sale, and several were being counted two and three times over. Nothing connected a dollar of spend to a dollar of revenue, because the link from the ad to his quote system had never been wired together. The number he had trusted for years was built almost entirely on actions that were never customers.

You already know this in your own work. A bin labeled "in stock, 400" is worth nothing if most of those parts are the wrong size or already broken. The label was confident. The bin was nearly empty. He had been pricing his whole business off the label instead of the contents, and no relabeling of the keywords would have changed what was actually on the shelf.

The intervention

We did not tune the keywords. We fixed the count first. We stripped the phantom actions out and kept only one thing as a win: a real quote request from a real buyer. That single change dropped the reported rate from the double digits to under one percent, and that lower number was the deliverable. It was the first honest reading the account had ever produced.

Then we wrote the prescription. The remaining half was a closed loop, a small wiring job so that every click could be traced through to the quote system and back, turning invisible outcomes into ones he could finally see and grade. That half was structurally the owner's to authorize. When he chose not to build it, the machine had a missing part that no operator on the outside could supply, so we left the full written prescription, held our price, and parted cleanly rather than keep working an account no one could measure.

The result, in context

This is a teaching case, so we will be plain about what changed and what did not. There is no performance lift to report here, and we are not going to invent one. What changed was knowledge. He went from a confident false number to an honest one, and from an invisible business to a visible one on paper.

The drop is not a defeat and the under-one-percent is not the point. The point is that he could finally see his own business clearly for the first time. A scary honest number beats a comfortable false one, because you can act on the honest one and you can only be fooled by the other. He understood the diagnosis and chose not to build the loop, so we left the prescription with him, did not discount our work to keep the relationship, and ended it well. The bridge stayed standing.

The rates above are values reported inside the advertising account during the engagement, not third-party audited revenue.