Dead clicks: the cheapest conversion fix most sites never look at
People are arriving and not converting, and you have been optimising copy rather than checking whether the interface responds.
Short answer
A dead click is a click on something that produces no response: a non-interactive element that looks clickable, a control that fails silently, or a handler that runs too slowly to feel connected. It is an interface defect wearing a conversion problem’s clothes, it is usually fixable in an afternoon, and most teams never measure it.
What counts as one
Behaviour analytics flags a click that produces no detectable change: no navigation, no DOM update, no visible response. Sometimes the element was never interactive and merely looked it. Sometimes it was interactive and failed. Sometimes it worked and took too long to say so.
The three causes, in order of frequency
Things that look clickable and are not: styled cards, icons, headings, images. Things that are clickable and fail: a JavaScript error kills the handler and nothing tells the user. Things that work slowly: the response arrives after the person has decided nothing happened.
Why it hides from you
It never appears in your funnel. The visit is recorded, the page loads fine, no error surfaces server-side, and the person leaves. In aggregate it reads as a copy problem or a pricing problem, and teams rewrite the headline while the button underneath it stays broken.
Finding yours
Any behaviour analytics tool reports dead clicks and where they happened. Read it by element, not by page: the same styled component repeated across a template produces the same dead click everywhere, which means one fix resolves a large number of them.
Fixing the look-clickable case
Either make it do the obvious thing, or stop it looking interactive. Cards with hover states that go nowhere, icons without handlers, underlined text that is not a link. Users are reading your visual language correctly; the language is lying.
Fixing the silent-failure case
Read your JavaScript error log against the funnel. Errors at volume are silent conversion losses: a user hits one, the interface stops responding, and nothing is reported anywhere you look. This is the highest-value log nobody reads.
Fixing the too-slow case
Acknowledge the interaction within a frame, then do the work. This is the same fix as responsiveness, because it is the same problem: the gap between action and feedback is where the person decides your product is broken.
Questions
- Is a dead click always a bug?
- No. Some are users clicking text or images out of habit. The signal is concentration: one element producing a large share of them is a defect, while a thin scatter across a page is normal behaviour.
- What rate is normal?
- There is no published benchmark worth trusting, because it depends entirely on interface density. Compare your pages against each other instead: the outlier page is the finding.
- Where does this rank against SEO work?
- Usually above it, when the rate is high. Traffic you already have and are losing at the interface is cheaper to recover than traffic you have not earned yet.
Measured, not asserted
On ihatepdf.cv, 37.16% of 132,748 sessions contained a dead click on the way to 100,542 users, alongside 1,942 recorded JavaScript errors and 6.20% of sessions ending in a quick back-click. Recovering a fraction of a rate that size is worth more than a quarter of link building, and it takes an afternoon.
Free tool for this: Log File Analyzer. No account, nothing uploaded.
Where this goes deeper
Every number on this page comes from one complete dataset: one product taken from zero to 100K+ users on search alone, with nothing spent on advertising. The full argument is Chapter 19 of the book. Five chapters are free to read.