Spam does not just waste your time. A fake enquiry is counted as a lead, attaches to whatever source the visit came from, and pulls that channel’s cost per lead down. The channel producing the most spam starts to look like the channel producing the most leads, so budget moves toward it. The arithmetic is right. The input is wrong.
The cost nobody counts
The obvious cost of form spam is time. Someone opens the enquiry, reads two lines, realises it is junk, deletes it. Thirty seconds, forty times a month. Annoying, survivable, and the only thing most anti-spam tools talk about.
The expensive cost is the one that never announces itself.
A fake enquiry does not sit quietly in an inbox waiting to be deleted. On its way in, it gets counted. It attaches to whatever brought that visit to your site, adds to that channel’s lead total, and divides into that channel’s spend to produce a cost per lead that is lower than the real one.
Then you look at the report and move money toward the channel that looks like it is working.
Worked through
Take a month with a $2,700 budget split across three channels. Illustrative figures, but the shape is common.
| Channel | Spend | Leads recorded | Cost per lead |
|---|---|---|---|
| Google Ads | $1,800 | 46 | $39 |
| Meta | $900 | 38 | $24 |
| Organic | $0 | 14 | $0 |
Meta looks like the obvious place to put more money. Half the cost per lead of Google Ads, and climbing.
Now separate the real enquiries from the junk.
| Channel | Spend | Real leads | Spam | Real cost per lead |
|---|---|---|---|---|
| Google Ads | $1,800 | 41 | 5 | $44 |
| Meta | $900 | 9 | 29 | $100 |
| Organic | $0 | 13 | 1 | $0 |
Meta was not cheap. It was noisy. The spam was concentrated in one channel, as it usually is, and it made the worst performer look like the best.
Nothing in the first table was calculated incorrectly. Every number followed from its input. The input was wrong.
Why spam concentrates
Junk submissions are rarely spread evenly. They cluster, and where they cluster tells you something.
| Where spam concentrates | Why |
|---|---|
| Paid channels with broad targeting | Automated traffic follows the ads, and loose targeting buys more of it |
| Display and partner networks | Less control over where the ad appears and who clicks it |
| Landing pages with an exposed form | An open form on a high-traffic page is a bigger target than one behind a click |
| Direct and unknown | Scripted submissions often arrive with no referrer at all |
That concentration is what makes the distortion dangerous. If spam were spread evenly it would inflate everything equally and the ranking would survive. It is not, so it does not.
The version of this that is worse
If the channel is paid search and you count form submissions as conversions, the distortion does not stop at your report. It goes back to the ad platform, which uses it to decide who to show your ads to next. That loop is covered separately in why spam leads make your Google Ads worse over time.
What fixes it
Two things, and both have to be true.
The junk has to be identified before it reaches your numbers. Not in your inbox, where you catch it after it has already been counted, but at submission.
And the enquiries have to be attached to a source individually, not counted in aggregate. Session-level analytics can tell you paid search produced 46 conversions. It cannot tell you which five of them were fake, because it has no idea who submitted anything. The difference between the two layers is the subject of marketing measurement vs attribution.
Lead Source does both. Spam is quarantined at submission, excluded from lead counts and from attribution, and reviewable if you want to check. That filtering is covered in full on form spam protection. Every enquiry that passes carries the source that brought it and the pages the person read before submitting, on the individual record rather than in a channel total. See how the attribution works, or what the reporting looks like once the junk is out of it.