Google Ads conversion tracking: Are your numbers right?

Most people treat the Conversions column in Google Ads like a scoreboard: whatever number sits there is the truth, and the job is to make it go up. But Google Ads conversion tracking doesn’t measure reality.

It reports the output of a stack of settings you, or whoever built the account, chose: an attribution window, a counting rule, a list of which actions get to “count.” Change any one of them and the number changes, without a single extra sale happening.

That matters because you make real money decisions on this data. You scale the campaigns that look like winners, pause the ones that look like losers, and report ROAS to a client or a boss. If the underlying count is wrong, every one of those calls is wrong too, and you usually won’t feel it until the budget is already spent on the wrong thing.

Why conversion numbers drift more than people expect

Two identical accounts can report wildly different conversion counts from the exact same traffic, purely on configuration. Neither is broken. They’re just measuring different things, the way two thermometers set to Celsius and Fahrenheit disagree while both being right.

The trouble starts when you read the number in plain English. “142 conversions” sounds like 142 people did the thing you care about. What it actually means is closer to “142 configured events fired, credited under these rules, over this lookback window, with modeled data filling the gaps Google couldn’t observe.” That’s a very different claim, and the distance between the two is where bad decisions live.

The fastest way to build trust in your Google Ads conversion tracking is to know the handful of places it goes sideways, so you can rule each one in or out. Here are the seven that account for nearly every discrepancy.

7 reasons your Google Ads conversion numbers are wrong

1. Your attribution window isn’t what you assume

The click-through conversion window decides how long after a click a conversion still gets credited back to that click. Set it to 30 days when your sales cycle actually runs 60, and you’re only ever crediting the first half of the journey. Recent campaigns look weak because their conversions simply haven’t landed in the report yet.

Where to check: Tools and Settings, then Conversions, then click the conversion action and open the Click-through conversion window setting. Confirm it matches how long your buyers really take to decide, not the default value someone left untouched at setup. If you sell a considered purchase, the default is almost certainly too short.

2. You’re double-counting with “every” instead of “one”

Google Ads lets you count every conversion from a single click, or just one. For a purchase, “every” is usually correct, because a customer can genuinely buy more than once. For a lead form, “every” quietly inflates your count the moment one person submits the same form twice, which people do constantly when a page reloads or they aren’t sure it went through.

If your conversion total looks suspiciously high next to the actual number of leads sitting in your CRM, the counting method is the first place to look. It’s a two-click fix once you spot it.

3. You’re counting micro-actions as primary conversions

Add-to-cart, form views, newsletter signups, page scrolls. All useful signals, but if any of them is set as a primary conversion, it gets folded into your headline number and, more dangerously, into the target Smart Bidding optimizes toward. Suddenly you have 400 “conversions” and 12 sales, and the algorithm is happily buying you more cart-adds instead of more revenue.

The fix is to move anything that isn’t a real business outcome to secondary. It stays tracked and visible for analysis, but it’s kept out of the count that drives bids and out of the number you report. This single change often cleans up the biggest gap between reported conversions and real ones.

4. Attribution model confusion

In 2026, data-driven attribution is the default in Google Ads and is generally the right choice. If your account is still running last-click, you’re handing 100 percent of the credit to the final click and undercrediting every campaign that started the journey. Upper-funnel work looks worse than it is, so you cut it, and then wonder why your best-converting campaigns slowly dry up.

Worth knowing: Google removed the older rules-based models (first-click, linear, time decay, position-based) back in 2023, so if you’ve inherited an old account with one of those still referenced somewhere, treat it as a flag that the setup hasn’t been reviewed in a while.

5. Google Ads and GA4 will never match, and that’s fine

This is the discrepancy that eats the most hours. The two platforms use independent attribution models on different data scopes, so by design they cannot agree.

Google Ads counts conversions that originate from a Google Ads click and models the gaps aggressively, because Smart Bidding needs a complete signal to function. GA4 credits only clicks that produced a tracked session and applies its own model across every channel, paid and organic alike.

A 10 to 30% difference is normal. GA4 usually reports differently because it’s filling in modeled conversions for users who declined tracking, and it’s crediting a broader view of the journey. GA4’s April 2026 attribution restructure widened the gap further for some accounts.

The mistake is treating that gap as a bug and burning days trying to force the two dashboards to reconcile. They won’t. Document the gap, confirm it stays consistent week to week, and only investigate when it suddenly jumps.

6. Conversion lag makes recent days look worse than they are

Conversions are credited to the day of the click, not the day the conversion actually happens. So a click today that turns into a sale next week shows up in today’s row a week from now, backfilling into the past. If you judge the last three or four days too early, you’ll see soft numbers, panic, and pause campaigns that are about to look excellent once their conversions mature.

The fix is patience calibrated to your own data. Check your typical days-to-conversion and treat anything inside that window as incomplete rather than final. For many accounts that means not making hard calls on the most recent week of data.

7. Missing, duplicate, or under-powered tags

This is the one setup-adjacent check worth doing while you’re already in here. A tag that fires twice inflates the count. A tag missing from a key confirmation page loses conversions entirely.

And Enhanced Conversions being switched off is the common quiet culprit: without it, Google Ads systematically under-counts, because it can’t recover the cross-device and cookie-blocked conversions that Enhanced Conversions is designed to match back.

Use Google Tag Assistant to confirm each conversion tag fires exactly once, on the right page, and nowhere it shouldn’t. Then check whether Enhanced Conversions is on under the conversion action’s settings. Turning it on is one of the few changes that makes your count go up and become more accurate at the same time.

Also Read: Google Ads metrics: What to track and what each one tells you

How to audit your Google Ads conversion tracking

You don’t need a full teardown to trust your data. This is a focused pass you can run once a quarter, and any time the numbers suddenly look off. Work through it in order, because each step rules out a whole category of problem before you move on.

1. Review every conversion action. Go to Tools and Settings, then Conversions. For each action, check its status, category, counting method (“every” vs. “one”), and attribution window. Flag anything set as primary that isn’t a genuine business outcome, and note any window that looks shorter than your real sales cycle.
2. Reconcile one action against GA4. Pull the same conversion event, over the same date range, in both platforms. Compare conversion volume, not revenue, because revenue amplifies the gap and muddies the picture. Use a window of at least 60 days so daily noise smooths out, and you’re comparing trends, not single days.
3. Scan for suspicious patterns. Suspiciously round numbers, a sudden spike with no matching campaign change, or a drop to zero all point to a tag or settings problem rather than a real shift in the market. Trust your gut here: if a change in the data has no story behind it, there usually isn’t one.
4. Confirm conversion value is passing correctly. If every conversion shows the identical value, your tag is sending a flat default instead of the real order value, and your ROAS is fiction built on a placeholder. For ecommerce especially, dynamic value passing is the difference between useful and worthless ROAS reporting.
5. Sanity-check against the back-end. Compare your Google Ads total against the actual orders or leads sitting in your CRM or store admin. For anything you report to a client or a board, the CRM is the source of truth for volume and revenue. Google Ads and GA4 are for deciding where to spend, not for counting the money.

What “close enough” actually looks like

Perfect agreement between platforms is the wrong target. Chasing it wastes hours and teaches you nothing, because the two systems are built to disagree. Here’s the realistic bar for a healthy setup:

• A 10 to 30% gap between Google Ads and GA4 is normal and expected, not a defect.
• GA4 often reports differently because it credits the journey across all channels and models conversions Google Ads never sees.
• A gap beyond about 15% that you can’t explain is worth a look. A larger gap you fully understand and can account for is not.
• Server-side tracking improves signal quality, but it won’t make the two dashboards match, because they still run different attribution models on different data.

The healthy mindset is simple: know your gap, know exactly why it exists, and watch for the day it changes. That, not two numbers that happen to line up, is what a trustworthy setup looks like. A consistent, explainable gap is a sign your Google Ads conversion tracking is working as designed.

The real problem: your numbers live in three places

Notice how much of the audit above involves bouncing between the Google Ads interface, GA4, and your back-end, while holding date ranges and event definitions in your head as you jump. That friction is exactly why most accounts never actually get audited. It’s tedious, it’s easy to put off, and discrepancies love to hide in the gaps between tabs where nobody’s looking.

Pulling your Google Ads data into a single view makes the sanity-check dramatically faster. When conversions, cost per conversion, and conversion value sit right next to spend, clicks, and CTR on one screen, an inflated count or a flat conversion value jumps out at you instead of hiding two tabs away. You spend your time interpreting the data rather than assembling it.

ContentStudio’s Google Ads Analytics dashboard connects your account and lays these metrics out together, so the pattern-scanning steps of your audit take minutes instead of a tab-hopping afternoon. 

To be clear about what it does and doesn’t do: it won’t set up your conversion tracking, because that lives in Google Ads where it belongs. What it does is make the ongoing job of checking whether your numbers are right far less painful, which is the difference between auditing your data every quarter and never getting around to it.

Conclusion

Your conversion count is an instrument reading, not a headcount, and accuracy is something you verify rather than assume. Run the 20-minute audit once a quarter, expect a 10 to 30 percent gap with GA4, reconcile against your CRM before you report revenue, and work through the seven causes in order when a number looks wrong. Do that, and you’ll trust your data enough to act on it fast, which is the whole point.

Want the checking part to take minutes instead of an afternoon? Try ContentStudio and see your conversions, cost per conversion, and conversion value next to spend and clicks in one view, so a number that’s off is obvious at a glance.

FAQ