Google Ads reports 60 conversions. The CRM contains 43 leads. The advertising team suspects a sales-process problem; the sales team suspects inflated marketing numbers.
Neither conclusion follows from the totals alone.
Google Ads conversions and CRM leads can differ because they count different events, use different dates and apply different attribution rules. A technical fault is possible, but the first job is to make the comparison fair.
This guide is a diagnostic workflow. It does not assume that every gap is harmless or that every account should reconcile perfectly.
1. Write down exactly what each number means
Start with one conversion action, one account and one reporting period. Ask what causes the event to be recorded.
A form submission, a telephone click, a connected call and a qualified prospect are four different events. If Google Ads combines them while the CRM counts unique contacts, the totals answer different questions.
Segment the advertising report by conversion action. In the CRM, separate contacts, enquiries, opportunities and sales. Decide whether a returning contact who submits a second enquiry creates a new lead record or updates an existing one.
2. Put both reports on the same calendar
Suppose someone clicks an ad on 30 September and enquires on 2 October. A report organised by the ad interaction can place the conversion in September, while the CRM records the October enquiry.
Google explains that its reporting date can differ from an external system's event date. Use the relevant “by conv. time” columns when reconciling against conversion timestamps, and align time zones. Upload date is another date again; it is not the date the buyer converted.
Also distinguish a lead's creation date from its qualification date. A CRM report of newly qualified leads is not directly comparable with a report of newly created contacts.
3. Compare the same population
A CRM often contains enquiries from search, social ads, organic traffic, referrals and property portals. A Google Ads account reports conversions it attributes to that account's advertising interactions.
Start with a clearly defined Google-related cohort, but do not assume a CRM's last-click source field recreates Google's attribution model. A person may discover the business through one channel and return through another.
Keep “unknown source” visible. Manually assigning unknown enquiries to Google to close the gap makes the report look tidy while making the evidence worse.
4. Inspect counting and duplicate handling
Google offers One and Every counting settings for conversion actions. One is a rule tied to an ad interaction, not a universal deduplication system for every person in your database.
Inspect what happens when someone submits twice, refreshes the confirmation page or enquires from another device. Check whether both a direct Google Ads tag and an imported analytics event measure the same business action as separate conversions.
Use a stable event identifier where the integration supports it. Do not create artificial new timestamps to make a rejected duplicate count again. Preserve the real event time and resolve the cause.
5. Check which columns and goals you are comparing
“Conversions” and “All conversions” are not interchangeable. Review the action's optimisation settings and the goals actually used by the campaign.
A record can be collected without being part of the metric you are looking at. Conversely, a combined report can count multiple stages of one customer's journey. Review Google's reporting-discrepancy guidance before editing the tracking code.
