Hasty SEO · Guide + free measurement
Measure your visibility in ChatGPT — citation rate, share of voice and entity recall, step by step.
Measuring brand visibility in ChatGPT means running your target prompts on a schedule and tracking three numbers per prompt: how often you're cited, how big a share of the cited brands you own, and whether the answer actually describes you correctly. Hasty SEO does this across all four AI engines and gives you the first ten prompts for free.
- 3 metrics covered: citation rate, share of voice, entity recall
- Daily sampling cuts ChatGPT answer variance by ~40%
- Free starter measurement of up to 10 prompts
Definition
What it means to measure visibility in ChatGPT.
It's the AI-search equivalent of a rank tracker. Instead of measuring SERP position, you run a fixed prompt library against ChatGPT on a schedule and record — for each prompt — whether your brand was named, which competitors were named instead, and whether the description of you is accurate. The output is three numbers per prompt over time, not a screenshot of one good answer.
How it works
Four steps to a stable measurement.
(1) Build a prompt library of 20+ real buyer questions per ICP. (2) Run each prompt daily against ChatGPT (and ideally Perplexity, Claude and Google AI Overviews in parallel — answers diverge). (3) Parse every answer for brand mentions and outbound citations. (4) Aggregate over a rolling 7-day window to remove daily variance and surface real movement.
Metrics
The three numbers that matter.
Citation rate = share of answers that cite your brand for a given prompt. Share of voice = your share of all cited brands for that prompt. Entity recall = whether the description of you is factually correct (positioning, geography, products). Vanity dashboards show only citation rate; serious measurement tracks all three because share of voice surfaces competitor moves and entity recall catches hallucinations early.
Free measurement
Start with ten prompts, free.
Hasty SEO ships a free starter measurement: add up to ten target prompts and we run them daily for 14 days against ChatGPT, Perplexity, Claude and Google AI Overviews. You get a baseline report with citation rate, share of voice and entity recall per prompt — no card, no auto-upgrade. Continue on a monthly plan only if the numbers warrant it.
Next
Run your baseline measurement.
Open seo.hasty.se, paste up to ten prompts and get a daily ChatGPT visibility report within 48 hours. Free for 14 days, EU hosting, no annual lock-in.
Frequently asked
Measuring ChatGPT visibility — FAQ.
What does it mean to measure visibility in ChatGPT?+
Running a fixed library of buyer prompts against ChatGPT on a schedule and tracking, per prompt, whether your brand was cited, which competitors were cited instead, and whether the description of you is correct. The output is three trended metrics, not a single screenshot.
How do you actually measure it?+
Build a prompt library of 20+ real questions per ICP, run them daily against ChatGPT (plus Perplexity, Claude and Google AI Overviews — they diverge), parse each answer for brand mentions and outbound URLs, and aggregate over a rolling 7-day window to remove variance. Manual measurement works but breaks at scale; that's where a tool helps.
Which metrics should you track?+
Three: citation rate (share of answers citing your brand), share of voice (your share of all cited brands per prompt) and entity recall (is the description of you correct). Tracking only citation rate is the most common mistake — share of voice surfaces competitor moves and entity recall catches hallucinations.
How often should you measure?+
Daily sampling, aggregated over a rolling 7-day window. ChatGPT answers vary significantly day to day — Hasty SEO data shows daily sampling cuts variance by roughly 40% versus weekly sampling. Weekly is acceptable for low-traffic prompts; never less often than that.
Is there a free way to measure ChatGPT visibility?+
Yes. Hasty SEO offers a free starter measurement of up to ten prompts run daily for 14 days across ChatGPT, Perplexity, Claude and Google AI Overviews — no card, no auto-upgrade. Manual measurement is also possible (paste prompts daily, log answers in a spreadsheet) but doesn't scale past a handful of prompts.
Does it work for Swedish-language prompts?+
Yes — Hasty SEO is built Swedish-first. Prompt libraries, entity recognition and competitor sets are tuned for Swedish B2B, the UI is in Swedish and data residency is EU under Swedish/EU law.