Citation counts vary across academic search tools because each platform indexes different collections of publications, uses distinct counting methodologies, and updates its data at different intervals.
Different Database Coverage
Each search tool crawls and indexes a unique subset of academic literature. Google Scholar casts the widest net, including journal articles, conference papers, preprints, theses, and even some technical reports from institutional repositories. Web of Science focuses on peer-reviewed journals in its curated collection, while Scopus covers a broader range of journals but still maintains selectivity. Specialized databases like PubMed only track citations within biomedical literature. A paper cited frequently in computer science preprints might show high counts on Google Scholar but lower numbers on Web of Science if those preprints aren't indexed there.
Counting Methodology Differences
Platforms apply different rules for what qualifies as a citation. Some count only formal citations in reference lists, while others include mentions in footnotes or supplementary materials. Self-citations—when authors cite their own previous work—may be filtered out by some databases but included by others. Version control also matters: if a paper exists as both a preprint and a published article, some tools count citations to both versions separately while others merge them.
Update Frequency and Timing
Citation databases refresh their counts on different schedules. Google Scholar updates continuously but may take weeks to process new publications. Commercial databases like Web of Science and Scopus typically update weekly or monthly. This lag means a paper published yesterday won't immediately show citations even if other papers have already referenced it. When conducting literature searches, WisPaper's Scholar Search covers 360M+ papers with 500K+ daily updates, helping you find recently published work that might not yet appear in slower-updating indexes.
Metadata Quality Issues
Incorrect or inconsistent metadata—misspelled author names, wrong publication years, or multiple DOIs for the same paper—can fragment citation counts across records. Duplicate entries split citations between versions, artificially deflating the count for each individual record.
Practical tip: When evaluating research impact, check multiple sources and note which database you're citing to maintain consistency in your comparisons.

