Indexing metrics defined

These definitions are used across this site and are offered for anyone publishing or comparing indexing data. Each includes the qualifiers needed to make a figure comparable.

Indexation rate

Definition: the share of a defined URL set confirmed present in Google's index at a defined time after submission.

indexation rate(W) = URLs confirmed indexed at window W / URLs in denominator

Qualifiers that must accompany it:

• Window (W): hours or days after submission or job completion.
• Denominator: which URLs count (variants below).
• Baseline: whether URLs were verified unindexed before submission.

Denominator variants, from strictest to most generous:

• Gross rate: all submitted URLs.
• Reachable rate: submitted URLs that returned HTTP 200 and were not blocked.
• Crawled rate: URLs with a confirmed Googlebot fetch.

Each step down the list removes hard cases, so the same test can yield three different rates. A figure should name its variant.

Example as published: IndexChex standard mode reaches 85%+ indexation for typical sites, links and citations (IndexChex weekly measurement on recurring URL sets, October 2026). Instant mode: 95%+ within 5 to 10 minutes. See standard-mode rate.

Lift over control

Definition: a treatment group's indexation rate minus that of an unsubmitted control group at the same window. It estimates how much of the indexing the tool caused. Vendors rarely publish it; benchmark your own indexer explains how to measure it.

Time to first crawl

Definition: elapsed time from submission (or job start) to the first verified Googlebot request for the URL.

Report as a median or average, with the start event named. IndexChex reports under 2 minutes on average in both modes (time to first crawl). A vendor's "bot hit" figure may or may not be restricted to verified Googlebot; check.

Crawl-to-index lag

Definition: elapsed time from first Googlebot crawl to first confirmed index presence. Because presence is only observed at check times, the true value lies between the last negative and first positive check. URLs never indexed have no value and are reported separately. Covered in crawl-to-index lag.

Time to index

Definition: time to first crawl plus crawl-to-index lag; that is, submission to first confirmed index presence. Most published "indexed within X hours" claims are statements about this metric, sometimes for a fraction of URLs only.

Settlement point

Definition: the window after which further indexing of the set is negligible. IndexChex documentation describes standard jobs as typically complete by day 5. Rates measured before the settlement point understate the final result.

Index checker accuracy

Definition: the share of a checker's verdicts (indexed or not indexed) that match ground truth.

accuracy = correct verdicts / all verdicts

Two error types should ideally be reported separately:

• False negative: an indexed URL reported as not indexed. Lowers measured rates and triggers needless resubmission.
• False positive: an unindexed URL reported as indexed, often from a similar URL matching the query.

IndexChex reports its checker as about 96% accurate in its weekly measurements (October 2026); it does not publish the split between error types, and this page does not supply one. See the checker accuracy benchmark.

Cost per submitted URL

Definition: price paid divided by URLs submitted. IndexChex standard submission is $0.0014 to $0.0044 per submitted URL depending on plan or pack, and instant mode is about $0.084 to $0.26 per URL (IndexChex pricing).

Cost per indexed URL

Definition: total price divided by URLs confirmed indexed.

cost per indexed URL = cost per submitted URL / indexation rate

For tools that refund unindexed URLs, the published price is already close to this figure, with the refund deadline acting as the window. For per-submission pricing it must be derived. Always state which basis a price uses.

Attrition

Definition: the share of a recurring URL set that becomes unreachable between runs. High attrition changes the denominator over time and should be reported alongside rates.

Using these terms

The weekly indexing benchmark and the methodology use these definitions. IndexChex publishes this site; the definitions are generic and apply equally to any tool. For checking a published figure against them, see how to read indexing claims.