What is Visibility index?
A visibility index bundles many rankings into one value. The tool maintains a fixed set of search terms, checks your position for each of them regularly, and weights it by estimated search volume and the expected click-through rate of that position. The result is a curve. Sistrix, Semrush and Ahrefs each calculate with their own formula and their own keyword set, so the absolute numbers are not comparable between tools. Only the trend within a single tool carries meaning.
The real benefit is early warning. A kink in the curve often shows a change days before it lands in your analytics as a drop in visitors, and it shows it independently of season and campaigns. As a target, though, the index is weak: the keyword set belongs to the tool, not to your business. It contains terms you can safely ignore, and it may not contain your most important queries at all.
The bigger limitation is a recent one. The index measures positions in the classic list of links. Citations in AI answers, answers that never produce a click, and product details taken from a data feed rather than your page do not appear in it. A stable index alongside falling traffic is therefore no longer a contradiction but a common normal state. The more honest addition is your own list of 20 real customer questions that you re-run in the AI systems yourself.
Why does Visibility index matter?
In March 2026 Ahrefs measured across 863,000 keyword SERPs and 4 million citation URLs that only 37.9 per cent of citations in Google's AI overviews still come from pages ranking in the top 10 for the same query. In July 2025 the figure was 76.1 per cent. An index that only aggregates positions therefore reflects a shrinking share of the visibility that actually produces enquiries.
Visibility index in practice
- 01An online shop sees its index hold steady for months while losing a fifth of its traffic - the positions are intact, the clicks move into the answer box above them.
- 02An agency compares its Sistrix value against a competitor's Semrush value and derives a ranking from it - the two numbers come from different keyword sets and are not comparable.
- 03A software vendor uses a kink in the curve as a trigger to investigate and finds an accidentally set noindex directive, two days before the drop would have shown up in visitor numbers.
