When the same IP address keeps clicking your Google Ads, it is worth a look, but it is not proof of fraud. One address can stand for many people or one honest buyer, and Google may already have filtered some of those clicks before billing you. Three questions sort it out: how many times the address arrived from an ad, over what period, and what each visit did.
First, check what Google already filtered
The Invalid clicks column in your campaign reports shows traffic Google's systems filtered, and Google doesn't charge for those clicks. Google also notes that third-party tools may flag traffic, such as duplicate IP addresses or bot activity, that Google had already filtered and excluded from your bill. So a repeat address in your logs is not automatically a repeat charge.
Clicks that weren't filtered are different. If you believe you were billed for invalid activity, you can ask Google to investigate the last 60 days. Our guide to invalid click refunds covers how.
Why one address isn't one person
- Many people, one address. Google's own help page explains that most networks, such as universities, offices, coffee shops and mobile carriers, use network address translation, which lets hundreds or thousands of devices share one public IP address. Repeat clicks from such an address can come from different people.
- One honest buyer, many visits. Someone comparing quotes or products returns several times over days, often from your ad each time.
- One source, many addresses. The opposite case: automated traffic and click farms rotate through addresses, so the worst sources may never repeat an address at all.
How to read the pattern
- Count arrivals from ads, not page views. Twelve pages after one click is one arrival. Twelve separate arrivals from your ad is the pattern that matters.
- Look at the window. Five arrivals in ten minutes means something different from five over a fortnight.
- Look at what each visit did. Did anyone read, scroll, call, fill in a form or buy? Repeated arrivals that leave within seconds every time are the ones to worry about. A zero-second visit on its own proves little: Google notes that many analytics tools need a second click to record a visit's duration, so someone who read only your landing page can show zero seconds.
- Check the network. Datacenter and hosting addresses rarely belong to shoppers. Mobile carriers and home broadband usually do.
What to do about it
- A datacenter or hosting address that only produces quick ad clicks: exclude it in Google Ads. Our free IP exclusion list builder turns addresses from your logs into a list Google Ads accepts, within the 500-entry limit.
- A mobile or home broadband address: be careful. Those addresses are shared and reassigned, so an exclusion can hide your ads from real customers later, and a range never belongs on one.
- Billed activity you believe is invalid: request an investigation within 60 days. Google's checklist asks for the IP addresses that generated repetitive clicks or clicks without conversions, taken from your web server logs, along with click IDs and dates.
An exclusion is not a refund
Excluding an address stops your ads from showing to it from then on. Google still charges for the clicks that revealed it, unless it filters or credits them.
Source: Google Ads Help, About invalid traffic.
Where AdProtektor fits
- It follows the visitor, not just the address. Each visitor keeps a persistent identity that belongs to one browser on one device and survives IP, VPN and network changes, so repeat clicks stay visible when the address changes, and different people behind one address stay separate.
- Click-frequency rules act on the pattern. You set a threshold, a window in hours or days, and a block duration. The rule counts arrivals from your Google Ads, not page views, and acts when a visitor exceeds the threshold.
- Paranoia Mode is the strict option. It acts when arrivals from one IP reach your threshold within the window, with a minimum of two. It can catch returning customers, so it suits periods when you are under attack.
- The evidence travels. Blocked sources go on your Google Ads IP exclusion lists, and the Refund Report exports each flagged click with its GCLID for an investigation request.
Set any threshold above what a genuinely interested customer would do in the same window. A considered purchase involves several visits over a week; an impulse buy doesn't.
The address is where the question starts, not the answer.
To see who is behind the repeat clicks on your ads, start a free trial and add the script to your site.