Start with ten minutes of triage, then follow the evidence. A click can be billed, reach your website, create a session and still fail to produce a useful enquiry. Those are four different questions. No bounce rate, IP address or visitor score answers all of them.

Choose one campaign and a comparable baseline. Record the reporting dates, account and analytics timezones, network, landing page, device and business outcome. Allow for conversion delay; keep interaction-date and conversion-date reports separate. Ten minutes can find the next investigation, not certify all traffic.

Ten checks, in order

1. Inspect native filtering and billing separately

Add Invalid clicks to the campaigns table. Keep filtered interactions separate from later Invalid activity credits. A third-party flag does not establish that Google charged the click; do not multiply every flagged visit by average CPC and call it a refund.

Inspect native filtering and billing separately →

2. Check search intent

Review reported search terms against what you actually sell. Irrelevance can indicate targeting waste without automation. The report omits some queries for privacy, so missing terms are not proof of concealed bots. Record candidates for negative keywords before changing them.

Check search intent →

3. Check geography against the service area

Compare location settings and available location reports with qualified enquiries. Location is an estimate; VPNs and shared networks complicate it. A data-centre city label alone does not justify blocking every visitor or an entire region.

Check geography against the service area →

4. Compare device and time patterns

Look for a change against the same campaign and comparable weekdays. Check mobile failures, loading and scheduled demand before treating a night-time spike as hostile. Keep sample size and lead delay beside the rate.

Compare device and time patterns →

5. Separate networks and placement evidence

Compare available network reports and business outcomes. Placement reports can describe delivery without proving which individual placement caused a bad lead. Use the free placement tool for reviewed exclusions; its output does not apply changes to Google Ads.

Separate networks and placement evidence →

6. Test the landing and attribution path

Open the destination directly for a controlled test, without clicking your live ad. Check redirects, preserved query parameters, consent behaviour and tag diagnostics. The URL checker compares pasted URLs locally; it does not fetch the site or certify tag execution.

Test the landing and attribution path →

7. Interpret engagement as behaviour

In GA4, review engagement alongside the configured session threshold, key events and page views. A spam submission marked as a key event can look engaged. Short visits and a clicks/sessions gap are investigation leads, not a fraud percentage.

Interpret engagement as behaviour →

8. Count log observations carefully

Review a short, access-controlled log window. A repeated request, repeated click ID and distinct click IDs are different counts. Refreshes and copied URLs can repeat IDs. Missing identifiers stay unknown; shared IPs and hosting ranges do not identify people or prove intent.

Count log observations carefully →

9. Verify what the conversion represents

Compare the action source, count setting, goal inclusion and duplicate routes with the outcome your business values. A form-button click is not server acceptance or a qualified lead. The local action audit finds configuration questions; it cannot observe your tags or change bidding.

Verify what the conversion represents →

10. Close the loop with business records

Track received, contactable, qualified and paid outcomes using stable internal record IDs. Unanswered is not automatically fake. Test one normal submission end to end and check duplicate handling before changing exclusions. Retain the evidence supporting both your diagnosis and its alternatives.

Close the loop with business records →

Review conversion configuration locally

Use the sample first or paste the supported conversion-action export. Select the intended business outcome and review the findings against your account. The read-only export script is linked below; running it requires your own Google Ads access. Neither the export nor this browser audit applies account changes.

Runs in this browser. Your action names and settings stay here.

Use the read-only export script below. Keep the header row. CSV and tab-separated exports with the same headers are accepted.

Check campaign goals to confirm bidding eligibility. Custom goals can include secondary actions. Match transaction or lead IDs to verify repeated events.

Review addresses and click-ID counts locally

Paste a minimal, redacted extract into the existing IP builder. Duplicate log lines should not become extra distinct click IDs. The builder prepares a proposed list; provider hints are network evidence, not a decision to exclude. Keep raw identifiers and personal information out of public reports.

Free tool

Build your IP exclusion list

Paste addresses from a log, a CSV export or a report, in any layout. You get a clean list in the format Google Ads accepts, with repeats removed and the most active addresses first. Paste a server log to count distinct GCLIDs from each address.

addresses share the first three parts

0 entries

Runs in your browser. Nothing you paste is sent anywhere.

Both embedded tools process pasted inputs in your browser. They do not connect to your ad account, apply exclusions, prove identity or decide billing. Verify scope, legitimate traffic and the saved destination setting before applying any reviewed change.

Leave the audit with a decision record

  1. Scope: campaign, network, dates, timezones, landing page and business outcome.
  2. Observation: metric, denominator, baseline, source and collection gaps.
  3. Alternative explanations: tracking, consent, page failure, duplicate events and lead delay.
  4. Decision: no change, repair measurement, review targeting or investigate suspicious activity.
  5. Follow-up: owner, review date, qualified outcomes, actual cost/credits and rollback condition.

Download the audit worksheet

Download the audit worksheet
Leave the audit with a decision record

Choose one campaign and a comparable baseline. Record the reporting dates, account and analytics timezones, network, landing page, device and business outcome. Allow for conversion delay; keep interaction-date and conversion-date reports separate. Ten minutes can find the next investigation, not certify all traffic.

Scope: campaign, network, dates, timezones, landing page and business outcome.
________________________________________

Observation: metric, denominator, baseline, source and collection gaps.
________________________________________

Alternative explanations: tracking, consent, page failure, duplicate events and lead delay.
________________________________________

Decision: no change, repair measurement, review targeting or investigate suspicious activity.
________________________________________

Follow-up: owner, review date, qualified outcomes, actual cost/credits and rollback condition.
________________________________________

Both embedded tools process pasted inputs in your browser. They do not connect to your ad account, apply exclusions, prove identity or decide billing. Verify scope, legitimate traffic and the saved destination setting before applying any reviewed change.

Sources checked 3 November 2026
https://adprotektor.app/blog/check-if-google-ads-clicks-are-real/

Sources checked 3 November 2026. Google Ads: invalid traffic; Google Ads: search terms; Google Ads / Analytics: clicks and sessions; GA4: engagement; AdProtektor: Google Ads integration; AdProtektor: conversion delivery.

Where AdProtektor fits

AdProtektor adds visit evidence and configurable protection. Eligible Google Ads IP exclusions require an active connection and applied sync status; configured conversion checks can withhold blocked traffic events while protection is active. Review those decisions alongside legitimate outcomes. Google determines invalid-traffic filtering and credits; third-party detection guarantees neither a charge nor a refund.

Start the free audit