The honest answer is that nobody can tell you what percentage of your Google Ads budget is wasted on fake clicks without looking at your account — and every widely-quoted figure on this topic comes from a company selling click-fraud protection software. That is not a reason to dismiss the problem. It is a reason to measure your own account rather than adopting someone else's number, and this article is how you do that.
Why the published percentages are not worth much
Search this topic and you will find confident claims: a fifth of spend, a quarter, a third. Trace any of them back and you land on a vendor's own report. Those vendors are not necessarily lying, but they have three structural problems you should know about before you plan a budget around their figure.
First, they define the term broadly. Most count every click their system flags as suspicious, and "suspicious" is a threshold they chose. Move the threshold and the headline number moves with it.
Second, their sample is self-selected. The businesses who install click-fraud software are disproportionately the ones who already suspected they had a problem. That is like measuring average illness severity in a hospital and reporting it as the national average.
Third, and most importantly, fraud rates vary enormously by industry, geography and campaign type. A legal-services advertiser bidding ₹800 a click in a competitive city has a completely different exposure from a local bakery on a ₹12 click. Averaging them produces a number that describes neither.
Worth being clear about our own position: we sell software that includes traffic quality scoring. So take the paragraphs above as applying to us too — which is exactly why the rest of this article shows you how to measure without buying anything.
What Google already refunds, before you do anything
Google runs its own invalid-click detection and does not charge you for what it catches. In your Google Ads account, add the Invalid clicks and Invalid click rate columns to any campaign report — they are available under the Performance column set and are off by default, which is why most advertisers have never seen them.
That figure is real money already credited back. It is also, crucially, only what Google's systems caught with high confidence. The clicks worth worrying about are the ones that looked human enough to bill you for.
Measuring your own waste in an afternoon
You do not need a tool for the first pass. You need to compare what you paid for against what actually arrived. Three checks, in increasing order of usefulness:
1. Clicks against sessions
Compare Google Ads clicks with the analytics sessions for the same campaign over the same period. A gap is normal — people bounce before the tracking fires, some block analytics — but a large and growing gap on one campaign while others stay stable is a signal. Clicks that never became a page view were charged for something that never loaded your site.
2. Sessions against form submissions
A campaign whose traffic arrives but never engages — near-zero time on page, no scroll, no form starts — is either badly matched to the ad or not human. Both cost you the same money. This is the step where the two problems look identical, which is why so many advertisers pause a campaign for "fraud" that was actually a targeting mistake.
3. Form submissions against real, contactable people
This is the one that matters and the one almost nobody does, because it requires connecting your ad data to what happened after the lead arrived. Of the enquiries a campaign produced last month, how many answered the phone? How many had a working number at all? A campaign generating cheap leads that never answer is worse than an expensive one that does, and the cost-per-lead column will never tell you.
The number that actually decides anything
Once you can see, per campaign, the cost of leads that turned into real conversations, the fake-click percentage stops being the interesting question. You will find one of two things.
Either the waste is spread evenly, in which case it is a cost of doing business and the right response is to optimise toward what converts rather than to hunt individual bad clicks. Or it is concentrated — one campaign, one placement, one geography — in which case you have found something specific and fixable, and excluding the source is worth the effort.
In our experience the second case is far more common than the first, and it is why a single account-wide percentage is such a poor guide to action. The average across your account can look tolerable while one campaign quietly burns a third of the budget.
Where a CRM comes into it
Step 3 above is the hard one because it requires the ad click and the eventual outcome to be attached to the same record. If your leads arrive in an inbox or a spreadsheet, the click ID is lost the moment they land, and you can never get back to "which campaign produced customers".
That is the specific gap Claudphic Ads was built to close: every lead carries its click ID through the pipeline, so cost-per-customer by campaign is a report rather than a research project. You do not need our product to do the first two checks above, and if those already show your waste is concentrated somewhere obvious, fix that first — it is free.