Bad bots can make wasted media spend look like a successful campaign. The damage goes deeper when their activity starts shaping where the next dollar is invested, argues Tumbleturn Marketing Advisory partner Dan Johns.
The impressions have been delivered. The clicks are coming in. Leads are up and the cost of acquiring them is falling. The campaign looks like it is working but some of the customers it appears to be finding may not even exist.
Ad fraud can turn wasted money into reassuring performance numbers. A malicious bot can load an ad, click on it and fill in a form. Put enough of that activity together and you get a very reassuring report, with delivery up, costs down and the campaign tracking nicely to plan.
The advertiser pays. The fraudster gets paid. The dashboard records another result.
What a billion impressions told us about bot traffic
Over the past 18 months we have analysed campaigns representing more than one billion paid media impressions using FouAnalytics. Of those, 35 per cent were confirmed bad bots and a further 21 per cent were flagged as suspicious. Only 30 per cent were confirmed human.
Put simply, more of those impressions were served to malicious bots than to people.
The wider internet offers little comfort. The Thales 2026 Bad Bot Report found that automated traffic made up 53 per cent of global web traffic in 2025, and that bad bots alone accounted for 40 per cent of all traffic. Marketers are now buying media in an environment where machines outnumber people.
How the ghost economy works
Inside that environment, a whole ghost economy has grown up. Fake and low-quality sites pull in automated traffic, programmatic advertising fills their ad slots, impressions pile up and money changes hands.
The advertiser pays for activity that will never turn into a customer, and in doing so helps fund one of the world’s largest criminal industries.
And the damage doesn’t stop at the media invoice.
Bad bots are teaching your campaign what to buy next
The first cost is the money spent on fraudulent traffic. The bigger risk is letting that fake activity shape your next investment decision.
If you optimise towards clicks, a source pumping out cheap automated clicks will look like a winner. If you accept fabricated form submissions, your cost per lead improves on paper while your sales team wastes hours chasing people who don’t exist.
Campaigns learn from the activity they receive. Unless your signals can separate real customer behaviour from manufactured activity, optimisation will keep sending more money to the wrong sources.
Left unchecked, the fraud starts writing the business case for your next budget allocation.
Any CMO asked to defend marketing’s contribution to growth should find that uncomfortable, because you can only trust the dashboard as far as you can trust the activity underneath it.
A reassuring fraud percentage in a verification report is where the questions should start. What inventory did the measurement cover? What could it actually detect? And what happened to the activity it couldn’t classify?
Here are five ways to take the buying seat back
1. Ask where your money went
Ask for the sites and apps your ads ran on, how much each one received and what each one produced. If an audience extension took the campaign beyond the environments you approved, understand the trade-off and decide whether it deserved the money.
2. Buy from inclusion lists you have assessed
Build inclusion lists of publishers and environments you have actually assessed and keep them up to date. Investigate new supply before you scale into it. An ever-growing blocklist leaves you chasing the next problem after it has already landed in your plan.
3. Connect media reporting to commercial results
Tie media reporting back to what happens in the business. Are the leads contactable? Do they meet the brief? Do the sales survive cancellations and refunds? Does the source that looks brilliant in a platform report produce anything of value once it reaches your business?
4. Add independent ad fraud measurement
Use independent scrutiny alongside your existing verification and make sure you understand what each of them covers. Write reporting access and the treatment of invalid traffic into your commercial arrangements and agree upfront who investigates discrepancies and what happens when delivery falls short.
5. Reward your agency for finding real customers
Most of all, give your agency permission to challenge the apparent bargain. Reward for finding customers and improving the quality of delivery and make room for a recommendation that reduces reported volume because it removes activity you should never have bought.
The cost of looking away
The tools to expose ad fraud exist today. Once the waste is visible, continuing to fund it becomes a decision that marketers and their partners need to account for.
Every fraudulent impression removed is money that can work harder elsewhere. Every false signal removed gives the next campaign a better chance of finding a real customer.

