How Google Ads conversion tracking works

What a conversion action actually is#

Ask Google what it is measuring and the answer is disarmingly simple. Google Ads Help, 2026 defines a conversion action as a specific customer activity with value to your business: a purchase, a sign up, a phone call, an app install, even an imported offline sale. Everything else in the measurement stack, the columns, the statuses, the automated bids, hangs off that one specific setting.

Under the bonnet the mechanics are old fashioned and honest. You add the Google tag, a small code snippet, to your site, and when someone interacts with an ad the tag tells Google Ads which click deserves the credit. The tag talks, the linker lays cookie crumbs, and the account keeps count. No tag, no trail, no numbers.

You will find the den where all of this lives at Tools and settings > Measurement > Conversions, the classic path most guides still quote. In the current interface the same screen sits under the Goals icon, via Conversions and then Summary (Google Ads Help, checked 11 August 2026).

Each action carries a status, and reading it is the first, fastest test. The four labels practitioners memorised are Recording conversions, Inactive, Unverified and No recent conversions (Search Ads 360 Help, 2026). Google has since regrouped the same signals into five troubleshooter statuses: Active, Needs attention, Misconfigured, Awaiting conversions and Removed (Google Ads Help, checked 11 August 2026). The table below maps the old scent to the new.

Statuses are the pulse check, not the full diagnosis, and reading them is only the first move in how to test Google Ads conversion tracking properly. For the wider anatomy of the platform, from campaign types to bidding, start at the folkfox Google Ads hub and burrow outward from there.

Primary versus secondary: the goal split#

Every conversion action now carries one of two roles in its conversion action settings: primary or secondary. The split does real work. Primary actions are reported in the Conversions column and used for bidding, so long as the standard goal they belong to is used for bidding. Secondary actions are observation only, reported in the All conversions column and never bid against (Google Ads Help, 2026).

That is the whole primary vs secondary conversion actions decision in one line: primary powers Smart Bidding and the Conversions column, secondary sits and watches. Get it backwards and the bidding brain optimises towards newsletter sign ups while your purchases watch politely from the sidelines.

There is one cunning exception. Actions placed inside a custom goal are used for reporting and bidding regardless of their primary or secondary flag (Google Ads Help, 2026). It is the sly trapdoor under many misconfigured accounts, and worth checking before you blame the algorithm.

If several actions all claim primary status, your cost per acquisition is an average across unequal events. Getting the role right is step one of how to test Google Ads conversion tracking, because the rest of the audit means nothing if the bidding brain is reading the wrong column. Our PPC service starts nearly every account audit right here.

Count Every versus One, and the value settings#

Next comes counting. The count every vs one conversion setting decides what happens when a single click converts twice. Every counts each conversion that happens after an ad interaction; One counts a single conversion per ad click, however many times the thank you page fires (Google Ads Help, 2026).

The fit follows the business model. Retailers and ecommerce brands want Every, because every extra sale from one click adds real revenue. Lead generators want One, because a visitor submitting the same form five times is one lead, not five, and Every would flatter the figures.

Then the value settings. Google offers three: use the same value for each conversion, use different values fed in per transaction, or use no value at all. The same value option is easier to set up but less representative of individual transactions, while different values mirror reality at the cost of more setup work (Google Ads Help, 2026). Choosing no value quietly starves value based bidding: strategies such as target ROAS exist to maximise conversion value, and they cannot weigh what you never weighed.

So the measured marketing move: transaction level values for shops, sensible fixed values for lead businesses (weight a demo request above a brochure download), and no value only on actions you would never bid towards. Match counting and value to the business model correctly and you are most of the way through how to test Google Ads conversion tracking without touching a single bid.

Windows, click dates and the backfill#

A conversion window is the countdown that starts at the ad interaction. The click-through window can be set anywhere from 1 to 30, 60 or 90 days depending on the conversion source, which means a click-through window runs up to 90 days at most on Search and Display (Google Ads Help, 2026). View-through and engaged-view windows are separate settings with their own defaults, 1 day and 3 days respectively (Google Ads Help, 2026).

Here is the part that catches even patient practitioners: conversions are credited to the click date, not the conversion date. The primary conversion columns are calculated on the time of the click, so a click last Tuesday that converts today is written back into last Tuesday's row (Google Ads Help, 2026). That is why the column backfills: recent days always look soft, then fatten as delayed data arrives, up to 90 days after the click depending on your window (Google Ads Help, 2026).

The fresh figures fallacy follows directly. Judge last week against a mature month and last week loses every time, not because performance dropped but because its conversions have not landed yet. Compare mature ranges against mature ranges, and learn your own account's conversion delay: that patience is as much a part of how to test Google Ads conversion tracking as any button you click.

How to test Google Ads conversion tracking, step by step#

Now the practical part, and the reason this entry exists: how to test Google Ads conversion tracking end to end, before a misfiring setup wastes a month of spend. Five checks, run in order, catch nearly every fault a fox will meet in the wild.

The nastiest failure is the quiet one. The conversion linker detects ad click information in your landing page URLs and stores it in first party cookies on your own domain (Google Tag Manager Help, 2026). When the linker is missing, blocked by a consent misfire or lost on a cross domain hop, nothing errors. The ads run, the tag fires, and a broken linker drops conversions silently rather than erroring, which is why the damage surfaces as a mystery dip instead of a red flag.

Testing at scale, or importing offline conversions, runs into the hard edges of the Google Ads API: basic access allows 15,000 operations per day, conversion uploads are capped at 2,000 conversions per request, and a mutate request cannot contain more than 10,000 operations (Google Ads API documentation, 2026). Batch accordingly, and never assume an upload landed without reading the response.

Measurement is the scent trail every other decision follows. If your columns cannot be trusted, neither can your bids, your budgets or your board report. Run the five checks above and you know how to test Google Ads conversion tracking end to end, not just glance at a green dot and hope. For a second pair of eyes across the whole earth, from tag to table, talk to folkfox.

Five faults that fox even careful advertisers#

Run how to test Google Ads conversion tracking often enough and the same five faults keep turning up, den after den, account after account. None of them throw an error. All five just quietly starve a column of data until someone notices the numbers look thin.

Every one of those five rows traces back to a setting already covered above: the status, the linker, the primary or secondary flag, the counting choice, the value setting. Nothing exotic, no hidden dashboard, just five habits a patient fox checks before trusting a number. That is why how to test Google Ads conversion tracking is a five minute routine rather than a rare audit: cheap to run, expensive to skip.

None of this needs new tooling. The built-in troubleshooter surfaces the Misconfigured status the moment a tag or linker breaks, and the same conversion linker documentation that explains the cookie also explains how to confirm it is present. The fastest fixes are the ones you already have access to.

Frequently asked questions#