Indexation rate: the share of pages Google has indexed

A directory listing does two things: it puts your product in front of the directory's visitors, and it gives your site a link. The second only counts for search ranking if Google indexed the page the link is on. A listing on a page Google never indexed still gets you the visitors who browse the directory itself, and nothing else.

Directories rarely publish this number about themselves, which is why we measure it. Two directories with the same domain rating can differ widely in how much of their content Google keeps in its index.

Search Google for site:example.com/your-page with the exact URL of the page. A result means the page is indexed; an empty result means that exact URL is not. This is the same kind of check our measurement runs, one sampled URL at a time, on a schedule.

The check answers indexed or not for a single page. It says nothing about ranking, and an empty result for one URL says nothing about the rest of the site.

A directory needs 30 checks on record before its rate is published, or one check for every live listing it has when it has fewer than that. Thirty is the conventional floor below which a proportion says nothing; the second half of the rule exists because a directory with twenty listings can never reach thirty checks, and a count of every listing it has is a census rather than a short sample. Below 5 checks nothing is published either way.

Under the floor the rate stays absent rather than provisional. The directory's page prints how many checks are on record and how many are needed, so a reader can tell a directory we have never checked from one we have checked twelve times and will not report on yet.

The sample is rolling. Each listing is rechecked at most once every 30 days, and a published rate is computed over the last 90 days of checks, with the number of checks behind it printed beside it. The rate moves as the sample turns over; it is not a single measurement taken once.