The five-minute tracking audit before launch

Every dollar of ad spend that launches before tracking is verified is a gamble, not a test. The difference between the two is whether you will know the answer when the money is spent.

Before any campaign goes live, I run a short audit that takes about five minutes and catches the failures that quietly ruin the numbers afterward. This is the checklist. The full setup version lives in the tracking checklist guide.

Why five minutes is enough

Tracking usually fails in a handful of predictable ways, and each one is visible before launch if you look in the right place. The audit is not a rebuild. It is a confirmation that the plumbing is connected, so the spend that follows produces data you can trust.

The checklist

1. One container, one source of truth. Confirm the site loads a single tag manager container. If two containers fire, every conversion can double-count, and the report will flatter you while the bank disagrees.
2. The conversion event actually fires. Trigger the real action on the site: submit the form, click the call button, land on the thank-you page. Watch it register in preview or debug mode. If it does not fire when you do it, it will not fire when a customer does it.
3. The thank-you page is reachable and tracked. Form redirects that clear too fast, or pages that reload before the tag fires, are invisible to the container. Load the page directly and confirm the conversion tag triggers on it.
4. Calls are counted once. If you use call tracking, confirm a call longer than your minimum duration registers as one conversion, not one per ring, and that browser and server events are not both counting the same call.
5. The numbers land where you will read them. Confirm the conversion appears in the account that runs the ads, with the right name and the right value. A conversion that fires but never reaches the bidding account is the same as no conversion at all.

What happens when something fails

If any step fails, the launch waits. This is not negotiable, and it is not a delay for its own sake. Spending against broken tracking produces confident numbers that are wrong, and wrong numbers are worse than no numbers, because they spend the next month’s budget in the wrong direction.

The fix is usually small: a missing trigger, a doubled container, a redirect that clears too fast. Five minutes to find, sometimes an hour to fix, and the alternative is a proving period built on sand.

The rule this enforces

No spend until the audit passes. One checklist, one verdict, and the launch only happens when the answer is clean. That discipline is what lets the Ledger at the end of the month say something the owner can act on, instead of something we both have to apologize for.

I count conservatively. Every untracked job is your upside, not my credit.