Result Completeness is the relationship between the depth you asked a SERP collection to reach and the number of usable, de-duplicated URLs it actually recovered once bounded scraping finishes. A Top 20 request that yields 17 clean URLs is 85% complete.
| Term | Result Completeness |
|---|---|
| Category | Search and SERP Data |
| Also known as | Completeness, SERP Completeness |
| Where it appears | Activity log, Details sheet, and saved reports |
What it means in RankGear
When RankGear collects a search results page it aims for a requested depth — the top 10, top 20, or however many positions a run targets. The provider rarely returns exactly that many usable organic URLs: some positions are ad blocks or SERP features rather than pages, some links are dead, and some duplicate a URL already captured. After RankGear finishes its bounded collection and drops those, Result Completeness records how much of the requested depth survived as usable URLs. You see it surfaced in the Activity log while a run executes, in the Details sheet for a finished collection, and carried into saved reports so a run’s coverage travels with its data.
How to interpret it
Read completeness as a sample-quality flag rather than a performance score. A run that recovers 20 of 20 requested results is a full sample; a run that recovers 12 of 20 is thinner, and any average, gap, or distribution computed across it rests on fewer pages. Before you compare two runs, confirm they share provider, country, language, device, location, and requested depth — and that both reached similar completeness. Comparing a near-complete run against a sparse one, or runs collected under different localization, means comparing unlike things.
| Completeness | What it tells you |
|---|---|
| Full (recovered = requested) | Complete sample; safe to compare against other full runs on matched settings. |
| Partial | Usable set is thinner than requested; note the smaller sample before trusting averages. |
| Low | Few usable URLs survived; treat downstream metrics as indicative, and check provider, localization, and depth for the cause. |
Example
A Top 20 localized run for “commercial roofing” set to United States, English, mobile, and a specific ZIP code may take several provider calls to page through the results. If two of the returned positions are ad units and one URL duplicates another already captured, RankGear records 17 usable URLs against a requested depth of 20 — a completeness of roughly 85%. Every one of those calls keeps the same localization settings, so the 17 recovered URLs form one coherent set rather than a mix of different searches stitched together.
Important considerations
- Completeness describes the quality of the collected sample, not a page’s ranking. It is a comparative collection indicator, not a Google score, and a page being present or absent does not mean RankGear caused it to rank.
- Search results change over time and by location, device, provider, and personalization context, so two runs minutes apart can differ in what they return and how complete they are.
- Low completeness thins every metric derived from the run; check the recovered count before trusting averages or gaps built on top of it.
- Only compare runs that share provider, country, language, device, location, and requested depth — different settings make completeness figures non-comparable.
Related terms
Part of the RankGear glossary · how RankGear measures · the 870 factors.