What it means
Two motives explain most of it. Competitors click to exhaust a rival's budget and push their ads off the page, while owners of low quality sites carrying ads generate clicks to earn a share of the advertising revenue.
The damage is not only the wasted spend. Fraudulent clicks distort every metric built on click data, so conversion rates look worse than they are, cost per acquisition looks inflated, and automated bidding trained on that data starts making poor decisions.
Ad platforms filter a great deal of invalid traffic themselves and credit it back, often before the advertiser ever sees the charge. What gets through tends to be the more sophisticated kind: residential proxies, real people paid to click, or software that imitates human mouse movement and dwell time.
Advertisers spot it in patterns rather than in any single click. Sudden spikes from one network or region, high click volumes with almost no time on site, repeat clicks from the same address and traffic arriving at implausible hours are the usual signals.
Practical defences are mostly about narrowing exposure. Excluding suspect placements and locations, capping daily budgets, shifting spend toward conversion-based bidding and asking the platform for a credit when the evidence is clear will all reduce the bill.
In practice
Real-world examples.
Example
A local law firm bidding on expensive keywords finds its daily budget exhausted by 9am every morning. The clicks come from a narrow cluster of addresses in its own city and none last more than two seconds, which points to a competitor rather than genuine demand.
Example
A software company running display ads across a large network notices one publisher delivering 40% of its clicks and none of its trials. It excludes that placement, spend falls by $18,000 a month, and trial signups do not change at all.
Example
A travel brand switches from manual cost-per-click bidding to bidding on completed bookings. Fraudulent clicks still occur, but because the platform now optimises toward paying customers, the amount spent chasing worthless traffic falls sharply.
Formula
Calculation
Wasted spend = total clicks x invalid click rate x average cost per click
True cost per acquisition = (total spend - wasted spend) / conversions
A retailer's search campaign records 120,000 clicks in a month at an average cost per click of $2.40, so total spend is 120,000 x $2.40 = $288,000. The campaign produced 2,400 orders, an apparent conversion rate of 2,400 / 120,000 = 2.0% and an apparent cost per acquisition of $288,000 / 2,400 = $120.
Log analysis suggests 15% of the clicks were invalid. That is 120,000 x 0.15 = 18,000 clicks costing 18,000 x $2.40 = $43,200, leaving 120,000 - 18,000 = 102,000 genuine clicks.
Measured against genuine traffic, the real conversion rate is 2,400 / 102,000 = 2.35%, noticeably healthier than it looked. If the fraudulent clicks were eliminated entirely, the same 2,400 orders would cost $288,000 - $43,200 = $244,800, a cost per acquisition of $244,800 / 2,400 = $102 rather than $120, a 15% improvement for no extra marketing effort.Case study
Seen in the real world.
The following is an illustrative and entirely fictional example. Redgrave Tools, an invented online supplier of workshop equipment, spent $288,000 a month on paid search and could not understand why its cost per order had drifted from $95 to $120 over two quarters while the sales team insisted nothing had changed about the product or the pricing.
An analyst joined the ad platform data to the web server logs and found that 18,000 clicks a month, about 15% of the total, produced sessions of under one second with no page scroll, almost all of them arriving between midnight and 4am from three internet service providers the business had never seen before. At $2.40 a click that was $43,200 a month, or roughly $518,400 a year.
Redgrave excluded the offending networks and countries, tightened its frequency capping and submitted an invalid traffic claim to the platform, which credited part of the spend. In this fictional case the reported conversion rate rose from 2.0% to 2.3% without a single change to the website, which was the clearest sign that the problem had never been the website at all.
Watch out
Common mistakes.
- Assuming the advertising platform catches everything, when its filters are good but not complete and the residue can still be a meaningful share of spend.
- Blaming a falling conversion rate on the website or the offer without first checking whether the click volume itself is real.
- Reacting by cutting the whole budget rather than excluding the specific placements, regions or hours where the invalid traffic is coming from.
Questions
People also ask.
Can an advertiser get a refund for fraudulent clicks?
Often yes, since the major platforms credit invalid traffic automatically and will review a documented claim, though they rarely repay everything an advertiser believes was fraudulent.
Is click fraud illegal?
Deliberately clicking to drain a competitor's budget or to inflate publisher earnings is generally treated as fraud, but proving who did it and pursuing them across borders is usually impractical.
Does switching to conversion-based bidding solve it?
It reduces the harm considerably, because the platform stops paying for clicks that never convert, but the fraudulent traffic itself does not disappear.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related