--- title: "AI Citation Log Template: Fields to Capture Per Engine" description: "Copy the AI citation log template fields to capture per engine: keep the brand name apart from the source URL, and withhold a rate until the sample holds." url: "https://articles.flowcrews.com/ai-citation-log-template-fields-to-capture-per-engine" category: "Measurement" date_published: 2026-10-16 reading_time_minutes: 7 word_count: 1746 --- # AI Citation Log Template: Fields to Capture Per Engine *One observation row per run, a child row per URL, and no rate until the sample can carry it.* ![AI Citation Log Template: Fields to Capture Per Engine](https://pub-07fb5e4955ba485b822d6b388be96d9a.r2.dev/2e474d51-a072-472e-ab2f-1a43a8990566/ai-citation-log-template-fields-to-capture-per-engine/hero-dcff42c0-0c91-456a-aa61-882190f51bf3.jpg) **TL;DR:** - Log one observation per prompt, engine variant, region, language, and run, including empty answers. - Store each cited URL on a child row, and keep brand mentioned separate from URL cited. - Record the citation format the engine showed, not a yes/no link flag. - Do not blend engines: mention and citation rates invert between Gemini and ChatGPT. - Withhold a citation rate until the sample can carry a week-to-week comparison. ## Which Fields Belong on an AI Citation Log Template ## Should a Brand Mention and a Source URL Be Separate Columns? ## Which Fields Change by Engine? ## How Many Repeats Before You Publish a Citation Rate ## How to Log Competitor URLs on the Same Prompt ## Can Google Analytics or Server Logs Replace a Citation Log? ## How a Citation Log Differs from a Mention Tracker or Share of Voice ## Conclusion - Observation grain — Log one observation per prompt, engine variant, region, language, and run, put each cited URL on its own child row, and keep empty runs so missing citations stay visible. - Separate binaries — Keep brand mentioned and clickable citation as separate fields; ghost citations and mention-only runs are different events and must not share one yes/no cell. - Unblended engines — Record engine, variant, and the citation format actually shown, treat ChatGPT browsing or Search as its own variant, and never average engines that cite different domains. - Publishable rates — Do not publish a citation rate until repeats can carry it; one capture misses domains, and a tight interval can take on the order of 100 runs per engine per prompt. - Competitor rows — Give every competitor URL its own child row with domain and a competitor flag, and prefer a finished weekly sample over unfinished daily checks. - What the log is not — GA4, Search Console, and crawler hits cannot replace a citation log, and mention rate or share of voice must not be written into a citation column. Start the next run on this template, freeze the engine, query hash, timestamp, region, language, cited URL, position, and format, and hold every rate until the sample can carry it. ## Frequently Asked Questions ### What is the minimum set of fields for an AI citation log? At minimum, store date, engine, exact prompt, whether the brand was named, position, whether it was cited or only mentioned, sentiment, sources cited, and competitors. A sound log also pins engine variant, region, language, and a stable query identity, and puts each cited URL on its own record rather than in one cell. ### Why shouldn’t brand mention and citation share one field? They rarely co-occur. Of 3,981 appearances in the June 2026 Semrush and Kevin Indig study, 61.7% were citations with no brand name, 25.1% were mentions with no citation, and only 13.2% were both. The IAB’s August 2026 visibility framework treats mention rate and citation rate as distinct Presence metrics for the same reason. ### Can I average ChatGPT, Gemini, and Perplexity into one visibility number? No. In that same study, Gemini named brands in 83.7% of appearances but cited sources in 21.4%; ChatGPT did close to the reverse, naming brands 20.7% of the time and citing sources 87%. Superlines’ analysis of 34,234 responses, as reported by BlueJar, found citation rates differing by a factor of 615x across platforms, with ChatGPT at 0.59% in that set. ### How many runs do I need before a citation rate is worth comparing? A citation inventory is readable sooner than a rate. One capture surfaced only 49.6% of the domains five captures found; three repeats recovered 83.1% of that set. At an observed 40% appearance rate, about ±10 points takes roughly 100 runs per engine per prompt. Recommendation lists also repeat less than 1% of the time across runs, so one test is nearly useless. ### Can GA4 or AI crawler logs replace the citation log? No. Google does not surface AI Overview impressions or citations in Search Console or GA4, and those clicks appear as google / organic. ChatGPT-User and Perplexity-User prove a live fetch on a user’s behalf; GPTBot and PerplexityBot are index or training crawlers. Fetch is not evidence that an answer quoted the page. ### Should the log only track my own domain? No. Airefs’ October 2026 analysis of more than 300,000 citations found 98.6% of ChatGPT citations and 98.2% of AI Overview citations pointed at third-party sites, not the brand’s own site. Log the URLs that carried the answer, including earned-media sources, not only pages on your domain. ## Sources - [https://www.pepper.inc/blog/mentioned-vs-cited/](https://www.pepper.inc/blog/mentioned-vs-cited/) - [https://huggingface.co/datasets/ThatDeveloperGuy13/aeo-citation-tracking-schema](https://huggingface.co/datasets/ThatDeveloperGuy13/aeo-citation-tracking-schema) - [https://consultevo.com/ai-citation-tracking-tools/](https://consultevo.com/ai-citation-tracking-tools/) - [https://www.amicited.com/blog/ai-brand-mention-tracking-template/](https://www.amicited.com/blog/ai-brand-mention-tracking-template/) - [https://bluejar.ai/blog/track-ai-citations-chatgpt-perplexity-gemini-claude/](https://bluejar.ai/blog/track-ai-citations-chatgpt-perplexity-gemini-claude/) - [https://cloro.dev/blog/llm-citations/](https://cloro.dev/blog/llm-citations/) - [https://cloro.dev/research/ai-citation-tracker/](https://cloro.dev/research/ai-citation-tracker/) - [https://getairefs.com/blog/ai-search-citation-sources-statistics/](https://getairefs.com/blog/ai-search-citation-sources-statistics/) - [https://www.scrapeless.com/en/blog/ai-visibility-measurement-sample-size](https://www.scrapeless.com/en/blog/ai-visibility-measurement-sample-size) - [https://www.ansvisor.com/blog/best-ai-search-citation-tracking-tools-for-websites-tested-compared](https://www.ansvisor.com/blog/best-ai-search-citation-tracking-tools-for-websites-tested-compared) - [https://getintel.ai/blog/best-ai-citation-tracking-tools/](https://getintel.ai/blog/best-ai-citation-tracking-tools/) - [https://www.geoscorehq.com/blog/track-ai-citations-ga4](https://www.geoscorehq.com/blog/track-ai-citations-ga4) - [https://crawlytics.app/blog/how-to-track-ai-citations](https://crawlytics.app/blog/how-to-track-ai-citations)