Your Google Ads dashboard shows 47 conversions last month. Your CRM shows 29 actual inquiries. Your sales team closed 12.
Something is off, and it is not the CRM.
Conversion tracking errors are common, quiet, and expensive. When the numbers in your account do not match reality, Smart Bidding is working off bad data. It optimises toward phantom leads, bids up on the wrong users, and the account drifts further from real results each week. In accounts I run, one of the first things I audit is whether the conversion data Google sees is a reasonable picture of what is actually happening.
The good news: tracking problems leave trails. Once you know the signs, you can find them.
The Short Answer
- A large gap between your Google Ads conversion count and your CRM or form-builder submissions is the clearest sign something is off.
- Duplicate conversions (the same form submission firing the tracking tag more than once) is the most common cause.
- Counting secondary actions like page views or button clicks as primary conversions artificially inflates your numbers.
- A tracking audit involves checking tag firing in Google Tag Manager, comparing columns in Google Ads, and cross-referencing with GA4.
- Bad tracking means Smart Bidding chases the wrong signals. Fixing it usually causes a short reset before performance improves on real leads.
Why “tracking is set up” does not mean tracking is right
Most accounts I take over have some form of conversion tracking in place. Very few have it set up correctly.
A conversion tag that fires is not the same as a conversion tag that fires once, for the right event, set as the right conversion type. The most common problems are subtle: a thank-you page tag firing on every visit to that URL, not just after a genuine form submission. A purchase event imported from GA4 alongside a separate Google Ads tag, so the same transaction counts twice. A phone call conversion that counts any call over 10 seconds, including calls from existing customers.
None of these look broken at first glance. The tags are present, the columns show numbers, the campaign status is healthy. The problem only surfaces when someone compares the Google Ads count to reality.
Most businesses notice something is wrong only when a campaign is paused and the conversion numbers barely move. At that point, the account has often been optimising against bad data for months.
Verdict: a tag that fires is a start, not a finish. Correct setup means it fires once, for the right event, counted as the right type.
Keep in Mind: Google Tag Manager’s Preview mode shows whether a tag fires and how many times. If your form confirmation page tag fires twice on a single test submission, you have a duplicate problem.
The five signs your tracking is not telling the truth
Run through this list before pulling any other lever in the account.
| Sign | What it likely means | Where to check |
|---|---|---|
| Google Ads conversions much higher than CRM leads | Duplicate tag firing or wrong event counted | GTM Preview, form-builder dashboard |
| “All conversions” much higher than “Conversions” column | Secondary actions counted alongside primary | Conversion actions settings, column comparison |
| High conversion volume but poor lead quality | Wrong actions set as primary (page views, button clicks) | Conversion action type in Google Ads settings |
| Conversions still rising after a campaign is paused | Residual tag firing from other sources, or view-through counted | Conversion settings, attribution model |
| GA4 and Google Ads show very different numbers for the same period | Duplicate tracking, attribution window mismatch, or different event definitions | GA4 vs Google Ads report side by side |
Two or more of these applying at the same time is a system issue. One alone can sometimes be explained. Two or more means something structural is wrong.
Pro Tip: In Google Ads, switch your columns to show both “Conversions” and “All conversions” side by side. A large gap between the two usually means secondary actions (short phone calls, button clicks, cross-device views) are sitting alongside your primary lead actions. That gap should be small.
Duplicate conversions: the most common silent inflator
Of all the tracking errors I find in audits, duplicate conversions are the most common and the least obvious.
The typical scenario: the thank-you page fires a Google Ads conversion tag. GA4 also fires a form-complete event. Both are imported into Google Ads. Every form submission counts as two conversions in the platform.
Another version: the conversion tag is placed directly in the site’s global snippet and again as a separate trigger in Tag Manager. Two tags, both firing, both reporting.
A third: the thank-you page URL is reachable by anyone who types it directly into the browser. People share confirmation links. Automated crawlers hit the URL. Every visit counts as a conversion.
Smart Bidding trains on these inflated numbers. If the real lead count is 25 but the platform thinks it is 55, the system is calibrated to a world that does not exist. It targets users who look like the phantom leads, spends up on traffic that produces form-page visits from crawlers, and the account looks healthy until someone cross-references the CRM.
Verdict: test it. Fire a real test submission and count the events in Tag Manager’s Preview mode. If you see more than one conversion tag fire, find the extra one.
Common mistake: Using both a native site tracking snippet and a Google Tag Manager implementation at the same time. This is one of the most frequent sources of duplicate conversions, especially on Wix, Squarespace, and WordPress sites with multiple plugins adding tag snippets.
Counting the wrong actions as primary conversions
Google Ads divides conversion actions into primary and secondary. Primary conversions go into the “Conversions” column and are used by Smart Bidding. Secondary conversions go into “All conversions” only; Smart Bidding ignores them.
The problem appears when someone marks a secondary action as primary by mistake, or when an action that should be secondary (a button click, a page view, a scroll depth event) was set as primary because it seemed like a useful signal.
Examples I see regularly:
- A “contact page visit” counted as a lead conversion, even though most visitors never submitted a form
- A phone number button click counted as a call, including people who clicked and then did not dial
- An email link click counted as a lead, when it is far upstream of an actual inquiry
- A thank-you page that is also reachable by clicking “back” from another page
Each of these inflates the conversion count with events that are not actual leads. The campaign looks efficient, the cost per “conversion” looks acceptable, and the account optimises toward users who visit contact pages or click phone buttons without ever calling.
The fix is to audit every conversion action in your account, check whether it is set to primary or secondary, and check what it actually measures. For a lead-gen business, your primary conversion should be a confirmed form submission to a genuine thank-you URL, a phone call over a meaningful duration (typically 45-60 seconds minimum), or an offline conversion imported from your CRM.
For more on which actions to count, see what to count as a lead in Google Ads.
Verdict: primary conversions drive Smart Bidding. If the wrong actions are set as primary, the whole account optimises in the wrong direction.
The GA4 and Google Ads discrepancy
If GA4 and Google Ads show materially different conversion numbers for the same period, that gap is worth investigating.
Some difference is expected. GA4 uses a different attribution model (data-driven or last-click, configurable), and Google Ads has its own attribution window (default 30 days for click, 1 day for view-through). A 10-20% difference between them is usually attributable to these window and model differences.
A gap larger than that, especially if GA4 is consistently lower, often means Google Ads is counting conversions that GA4 cannot verify. Common causes: conversions firing client-side only (so they miss users with ad blockers or cookie restrictions), view-through conversions counted in Google Ads but not GA4, or Google Ads importing GA4 events that are defined differently in each platform.
I covered the full breakdown of why these numbers diverge in the GA4 vs Google Ads conversion tracking post. The short version: GA4 running slightly lower than Google Ads is usually fine. A large or growing gap in either direction is worth investigating.
Verdict: a consistent large gap between GA4 and Google Ads is a tracking problem until proven otherwise.
How to run a quick tracking audit yourself
You do not need a developer to do this. Here is the order I work through it.
Step 1: Compare the numbers. Pull your Google Ads conversion count for the past 30 days. Pull the same period from your CRM or form-builder. Pull the same from GA4. If any two of those differ by more than 20%, keep digging.
Step 2: Check your conversion actions. In Google Ads, go to Tools and Settings > Conversions. Look at every conversion action. Note whether each is set to primary or secondary, what the counting method is (one per click or every), and what the attribution window is. For a lead-gen account, primary conversions should be confirmed form submissions and meaningful phone calls only.
Step 3: Test your tags. Open Google Tag Manager’s Preview mode, go to your website, and submit a real test form. Watch which tags fire and how many times. If your conversion tag fires twice for one submission, you have duplicates.
Step 4: Compare columns. In Google Ads, add both “Conversions” and “All conversions” to your column set. A large gap means secondary actions are inflating your reported numbers even if Smart Bidding ignores them. Know what is in that gap.
Step 5: Check view-through conversions. In your conversion settings, look at whether view-through conversions are included and what window is set. For lead-gen, view-through conversions counted as primary conversions are almost never appropriate.
If the fix involves restructuring your conversion actions or reconfiguring tags, plan for a Smart Bidding relearning period of 3-4 weeks after any significant change.
For setting up offline conversion tracking correctly once your primary actions are clean, see how to set up offline conversion tracking in Google Ads.
How I handle this for lead-gen businesses
Every account I take over gets a tracking audit before I touch any campaign settings. Bidding decisions built on bad data compound the error: a Smart Bidding strategy chasing phantom conversions will keep producing them and call it success.
In practice, I check three things first. Whether the numbers in Google Ads roughly match the CRM (within 20%). Whether every primary conversion action is an actual lead action, not a proxy. Whether a test form submission fires the tag exactly once.
If any of those are off, fixing tracking is the first job, even if it means a short reset period for Smart Bidding. A correctly calibrated account showing 25 real leads per month is more valuable than one reporting 60 and producing 20.
Ready to find out what your tracking is actually counting? If your conversion numbers look better than your pipeline feels, there is probably a gap worth looking at. I run tracking audits as part of every account review.
Performance marketing specialist for 10+ years.
LinkedIn