i10-index

The i10-index is a Google Scholar-specific author metric: the number of a researcher’s publications that have each individually accumulated at least 10 citations. Google introduced it in 2011 alongside the launch of Google Scholar Citations (the self-managed author-profile feature, in limited beta from July 2011 and opened to all researchers on November 16, 2011) as a deliberately simpler companion to the older h-index.

What makes something an i10-index

Operationally, a figure counts as an i10-index when all of the following hold:

• Fixed threshold, not self-scaling. A paper either has 10 or more citations or it doesn’t — there is no adjustment for how many papers clear the bar, unlike the h-index’s h-papers-with-h-citations construction.
• Computed from Google Scholar’s own index. The figure is generated automatically from the publication list and citation counts attached to a Google Scholar Citations profile, using Google Scholar’s own web-crawled citation data — not a value a researcher calculates or reports independently of that platform.
• Displayed in “All” and “Since [5 years ago]” versions. Google Scholar Citations profiles show both a lifetime i10-index and a rolling 5-year i10-index, the same two-window convention it applies to total citations and h-index.
• Not produced by Scopus or Web of Science. The i10-index is a Google Scholar feature. Scopus Author Details and Web of Science ResearcherID pages report citation counts and h-index but do not compute or display an i10-index — it is not a cross-database bibliometric standard.

Why Google introduced it

The h-index requires understanding a self-referential condition (the largest h such that h papers have at least h citations each) that many researchers and evaluators find non-intuitive on first encounter. The i10-index has no such condition: it is a plain count against a single fixed threshold, which makes it trivial to compute by eye from a sorted citation list and easy to explain to a non-specialist audience. Google Scholar presents it alongside, not instead of, the h-index and total citation count — the three figures are shown together on every public profile, each surfacing a different view of the same underlying citation data (a fixed-threshold productivity count, a self-scaling productivity-and-impact hybrid, and a raw total).

i10-index vs. h-index

Both metrics are computed from the same citation list and both increase (or stay flat) over time, but they answer different questions:

• The h-index scales with an author’s own output — a highly prolific author with modest per-paper citation counts can post a high h-index by accumulating many papers just above a rising bar.
• The i10-index uses one bar for everyone. It rewards any paper that clears 10 citations equally, whether it has 10 or 10,000, and does not credit papers that fall short of 10 no matter how close they are.

Because the threshold is fixed rather than scaling, the i10-index is more sensitive to field-level and career-stage differences in citation volume than the h-index is — a researcher early in their career, or in a low-citation-density field, may have very few papers that ever clear 10 citations even with a respectable publication record.

Known limitations

Because the i10-index is computed from Google Scholar’s own citation data, it inherits the same coverage, deduplication, and curation characteristics documented for every Google Scholar Citations figure — see the Google Scholar Citations profile entry for the full detail on broader-but-less-curated coverage, potential duplicate-citation counting, and the absence of a published, auditable indexing algorithm. Two limitations are specific to the i10-index itself, on top of those inherited ones:

• An arbitrary round-number threshold. Ten citations has no bibliometric justification beyond being a memorable round number; Google Scholar’s own documentation does not claim otherwise. A paper with 9 citations and one with 10 are treated as categorically different despite the practical difference between them being a single citation.
• No normalization for field, discipline, or career stage. Like a raw citation count, the i10-index is not adjusted for citation-density differences across fields or for how long a researcher has been publishing, which limits its validity for cross-field or cross-cohort comparison.

Responsible-assessment guidance that already cautions against relying on the h-index for hiring, promotion, or funding decisions — including DORA and the Leiden Manifesto — applies at least as directly to the i10-index: it is a single, uncontextualized citation-count metric, not a substitute for qualitative expert review of research quality.

Why this matters for research administration

Research administrators most often encounter the i10-index self-reported in a CV, biosketch, or tenure dossier, usually alongside an h-index figure and a link to the applicant’s public Google Scholar Citations profile. Because both figures come from the same self-managed, non-peer-reviewed index, treat an i10-index the same way responsible-assessment frameworks recommend treating any single citation-based metric: as one data point that requires context (field, career stage, database source), not as a standalone ranking criterion. Where a policy or funder requires a specific citation database (for example, a Scopus- or Web of Science-sourced h-index), note that the i10-index has no equivalent in either of those systems and cannot be substituted for it.

References

• Google Scholar Blog, “Google Scholar Citations Open to All,” scholar.googleblog.com, November 16, 2011.
• Google Scholar Citations help documentation, “Citation metrics,” scholar.google.com.