Google Ads conversions can differ from your actual inquiries because the two reports may count different actions, use different dates, or give credit differently. Tracking can also count the same inquiry more than once, miss an inquiry, or record a contact click that never becomes a conversation. Start by checking what each conversion action measures, then compare it with received forms and call records.
If your dashboard says 20 conversions and your team remembers nine inquiries, neither number explains the gap on its own. Before changing the budget, find out what those 20 conversions contain.
What counts as an inquiry
For this review, an inquiry means a genuine request about your services received through a form, email, phone call, or another contact channel. Decide whether you are counting individual contacts or unique prospective customers. A person who submits a form and calls about the same request may create two contact records but only one prospect.
| Measure | What it tells you |
|---|---|
| Tracked contact click | Someone clicked a phone number, email link, or contact button. Completion is not established. |
| Platform conversion | An action credited under the platform’s tracking and reporting rules. It may measure a click, submission, call, or later outcome. |
| Actual inquiry | Your business received a genuine request. Check the form entry, message, or call record. |
| Qualified lead | The inquiry meets your agreed requirements, such as service fit, location, and budget. |
| Booked business | A confirmed appointment, signed engagement, or sale, according to your definition. A booking is not automatically collected revenue. |
Keep these measures separate. A campaign can produce genuine inquiries that are a poor fit. It can also produce useful inquiries that your tracking fails to recognize.
Why the numbers can differ
The conversion measures an earlier step
An email-link click can open a draft that is never sent. A phone-link click can open a dialer without a completed call. A form-button click can happen before validation fails. Check the trigger behind the conversion name. Calling an event “lead” does not prove a lead was received.
A GA4 key event is an event marked as important to the business. It is not inherently a verified inquiry, and it is not limited to Google Ads traffic. Creating a Google Ads conversion from it adds advertising measurement; it does not independently verify delivery to your inbox or CRM. (GA4 key events and conversions)
The reports include different actions
In Google Ads, primary actions generally feed the Conversions column and bidding when their goal is used by the campaign. Secondary actions appear in All conversions and are generally observation-only. A custom goal is an important exception: included secondary actions can be used for bidding. Compare the actual actions and campaign goals, not just the column labels. (Primary and secondary conversion actions)
One person creates several records
Someone might submit twice, call after submitting, or trigger two separate conversion actions for the same form. A native Google Ads action and a GA4-imported action can overlap if both measure the same outcome and both are included in the total you are reviewing.
Google Ads’ One counting option applies to a conversion action after an ad click. It does not merge separate actions into one person. Every can count repeated conversions for an action. Review the setting, but do not expect it to deduplicate your CRM. (Conversion counting options)
The dates and attribution do not line up
Standard Google Ads conversion columns generally report against the ad interaction date. Your CRM may record the date an inquiry arrived. A September 30 ad click followed by an October 2 inquiry can therefore appear in different months. Compare the relevant by-conversion-time column with your inquiry dates, align time zones, and allow for processing and later updates. (Conversion reporting and dates)
Also compare like-for-like traffic. An all-channel form total includes inquiries that Google Ads may not receive credit for. Attribution models and windows can allocate credit differently, including fractional credit. For video or display activity, check whether the report includes engaged-view or view-through credit rather than assuming every conversion followed a click. (Data discrepancies and attribution)
Tracking and delivery are separate systems
A form may save successfully while its notification email fails. A CRM integration may miss a saved entry. Conversely, tags may fail because of consent settings, browser restrictions, or a broken implementation. Modeled conversions can fill some measurement gaps in eligible setups, but they do not supply a person-by-person CRM list. Investigate these possibilities rather than assuming any one explains your account. (Conversion modeling)
A practical reconciliation checklist
Start with an agreed reporting period and a record of the current setup. Investigate first; changing goals halfway through the review makes the comparison harder.
1 Define the outcome and scope
Write down what qualifies as a received inquiry, a unique prospect, a qualified lead, and booked business. Specify whether existing customers and repeat requests belong in the acquisition total. Separate all-channel inquiries from the subset with evidence of a Google Ads connection.
2 Break the Google Ads total into actions
Segment the report by conversion action. Record each action’s source, trigger, count setting, primary or secondary status, campaign goal, attribution model, and conversion window. Note whether you are reviewing Conversions or All conversions. An account-level summary and a campaign report can have different scopes. (Primary and secondary conversion actions; Conversion counting options; Data discrepancies and attribution)
3 Put the reports on a comparable calendar
Export the relevant Conversions (by conv. time) or All conv. (by conv. time) column for the same period as your business records. Match action scope as well as dates. Record both systems’ time zones and the export date. Treat recent totals as provisional and revisit them after processing. (Conversion reporting and dates)
4 Check the records behind the inbox
Review saved form entries, received emails, CRM records, and call logs. Mark spam, repeat requests, existing-customer contacts, and delivery failures separately. For phone conversions, establish whether the action measures a click, a call meeting a duration threshold, or an imported business outcome. A long call alone does not prove qualification. (Phone call measurement)
5 Match what you can and label the rest
Within your authorized business systems, match records using reliable inquiry IDs, call IDs, timestamps, and captured attribution information. Keep contact details out of public reports and ordinary analytics event parameters. Mark each item as matched, duplicate, excluded with a reason, or unresolved. An inquiry close to an ad click in time is not enough to prove attribution.
6 Investigate the specific tracking failures
If a conversion has no matching record, inspect its trigger and the form or CRM delivery path. If a received inquiry has no conversion, inspect tag coverage, attribution capture, consent behavior, and import processing. Check whether a thank-you page can trigger again on refresh or direct access. Use existing records and configuration first. Arrange any test submission with the business before sending it.
7 Produce a usable result
Report platform conversions, verified inquiries, unique prospects, qualified leads, and booked outcomes separately, with clear definitions. List explained differences and unresolved items. Assign each fix to an owner and date. Preserve the original baseline so later changes can be evaluated.
The goal is an explainable relationship between the reports. Modeled and attributed totals will not always reconcile one-for-one with identifiable people. Do not invent matches to make a spreadsheet balance.
A worked example
Illustrative example only. These figures are invented to explain the process and are not results from Frank Fay or any client.
Suppose an action-level review shows 20 conversions: eight form actions, five phone-call actions, and seven email-link clicks. For this simplified example, all 13 form and call actions have matching records in the aligned reporting period; fractional credit and modeling are not involved.
| Step | Evidence and treatment | Remaining count |
|---|---|---|
| Platform total | 8 forms + 5 calls + 7 email-link clicks | 20 actions |
| Separate contact clicks | The 7 email clicks do not establish sent emails. No matching received emails were found. | 13 form and call actions |
| Review forms | Of 8 form records, 1 repeats a request and 1 is spam. | 6 genuine form inquiries |
| Review calls | Of 5 calls, 1 is an existing-customer service request excluded from new acquisition, and 1 repeats a form inquiry. | 3 additional prospects |
| Deduplicate prospects | 6 form prospects + 3 additional call prospects | 9 unique prospects |
| Assess business outcomes | 5 of the 9 meet the qualification rules. Of those 5, 2 book. | 5 qualified; 2 booked |
Twenty reported actions and nine unique prospects can both describe the same period. That does not make all 20 equally useful for campaign optimization. The next decision is whether the selected goals reflect what the business wants more of.
What to fix before spending more
If contact clicks are being treated as completed inquiries, clarify the reporting and review the bidding goals. If the same submission feeds overlapping actions, identify which action should represent the outcome. If inquiries are saved but not reaching the team, repair the handoff. If genuine inquiries are arriving but rarely qualify, review targeting, search terms, the offer, and landing-page expectations.
When reliable qualification and booking records exist, assess whether those later outcomes can be fed back through an appropriate offline conversion setup. Define ownership, data quality, and the intended bidding outcome first. Avoid counting several stages as interchangeable successes. (Offline conversion imports)
Need help finding the gap
If your advertising report and your team's inquiry count tell different stories, I can review what is being measured and where the records diverge. Send me your website and a brief description of the discrepancy. We can agree on a focused review and the access it needs. Explore Google Ads consulting
Frank Fay is an independent marketing consultant whose work connects campaigns, website content, customer follow-up, and measurement.