Free Copilot Rank Tracking: Manual, Bing Report and Trial Routes Compared

Free Copilot rank tracking comes down to two routes that work together: signed-in Microsoft Copilot sessions run by hand, which show the full answer and its sources, and Bing Webmaster Tools' free AI Performance report, which counts how often your pages were cited. Neither reports a rank. One-shot checkers and trials add speed, not control.

Free Copilot rank tracking: the short answer

1. Manual sessions show you the full answer at no cost. Microsoft's consumer terms require a Microsoft account and bar bots and scrapers, so the free method that fits those terms is a person with a signed-in account and a spreadsheet. It costs time, and Microsoft describes the free version as limited and subject to capacity.
2. The Bing AI Performance report is Microsoft's free citation data. It counts citations of your pages across Microsoft Copilot, AI-generated summaries in Bing and select partner integrations. Microsoft says the data is a sample and does not show individual answers, exact prompts or why a page was referenced.
3. Copilot's terms rule out automating a free account, and the Bing Search APIs are retired. Microsoft retired the Bing Search APIs in August 2025. The paid Grounding with Bing Search tool answers with a model you deploy, not with Copilot, so it measures something related rather than what a buyer saw.

Which Copilot, and which kind of rank?

"Copilot" names several products, and searches for this phrase mix them. Some of the results are forum threads asking for a free tool to compare GitHub Copilot with other coding assistants, others are classic SEO rank trackers, and others are tools that record what Copilot answers say about brands.

This page covers one meaning: how your brand appears in the answers of consumer Microsoft Copilot, the assistant on the web, in Windows, in Edge and on mobile. It does not cover:

• GitHub Copilot, a coding assistant. Comparing coding assistants is a developer evaluation, not brand measurement.
• Copilot signed in with a work or school account. Microsoft's Copilot terms state that they do not apply to users who sign in with a work, school or organizational account. That surface runs under your organization's settings, so measure it separately if your buyers use it.
• Classic Bing positions. If you want to know where a page ranks in Bing's blue links, a normal rank tracker or Bing Webmaster Tools' search performance data does that job.

Even within consumer Copilot, "rank" is loose language. Copilot writes an answer; it does not publish a numbered position for brands. What an answer exposes can be observed and counted: whether your brand is named, whether your website is linked, which source links were shown, whether the answer recommends you rather than just listing you, which competitors appear, and where you sit if the answer is a list. Prime AI Visibility's published metric definitions use four of those by name: Brand named, Website cited, Source links and Explicit recommendation.

What a free route has to capture

A multi-engine baseline can combine manual Copilot rows with Prime AI Visibility records from its supported engines; do not label those records as Copilot results.

Rate any free method against the same seven checks you would apply to a paid tracker. A method that fails them produces numbers you cannot defend later.

1. Prompts. You choose the exact questions, and the wording stays frozen between runs.
2. Answer. The full answer text is kept, not a score or a summary.
3. Mentions. Brand named is recorded separately from links.
4. Links. Website cited is recorded when the answer links to your domain.
5. Sources. Every link the answer displayed is listed, including other sites.
6. History. Earlier answers remain readable next to new ones.
7. Limits. The method says how many questions, how often, and what happens at the limit.

A competitor column is useful too, but it follows from the answer: if you keep the full text, you can always count rivals later.

The free routes compared on 11 October 2026

Method note: this comparison rates routes, meaning ways of collecting Copilot evidence without paying, not named products. Free tiers change without notice and can only be confirmed by using them, so the ratings come from vendors' public pages rather than trials. Ratings are words. "Verify" means the public pages did not settle the question, so test it before you rely on it.

First-party and manual routes

Check	By hand	Bing report
Prompts	strong	none
Answer	strong	none
Mentions	strong	none
Links	strong	partial
Sources	strong	none
History	partial	partial
Limits	partial	strong

Vendor routes

Check	Checker	Trial
Prompts	verify	strong
Answer	verify	verify
Mentions	verify	verify
Links	verify	verify
Sources	verify	verify
History	none	partial
Limits	strong	verify

Notes on the ratings. Manual sessions rate strong on everything visible because you read the whole answer yourself, and Microsoft's transparency note says web-grounded answers carry hyperlinked citations listed below the text. History is partial because it depends on you pasting each answer into a sheet. Limits are partial because Microsoft describes the free version as limited and subject to available capacity without publishing a question count.

The Bing report rates none on prompts and answers because Microsoft says it shows grouped "grounding queries", which are not full user questions, and no individual answers. It rates partial on links because it counts citations of your own pages from a sample, and it cannot see an answer that names your brand without linking you. History is partial because the dashboard offers 7-day, 30-day, 3-month and custom ranges within available data, so monthly exports are how you keep a longer record. Limits rate strong because Microsoft documents exactly what the data leaves out.

One-shot checkers rate strong on limits because the limit is obvious: one report. They rate none on history unless you save the report. Trials rate strong on prompts because you enter your own questions, and partial on history because the record usually ends with the trial. Export before it ends.

Treat the first month of any route as a baseline, not a verdict. Our page on the limits of measuring an LLM like a rank tracker explains why a single position number misleads for every answer engine, not only Copilot.

Route 1: signed-in Copilot sessions by hand

Microsoft states that it offers a limited free version of Copilot subject to available capacity and limits, and that when you hit a limit you can upgrade or come back later. Its Copilot terms, effective 18 August 2026, add three points that shape free Copilot rank tracking:

• You must be signed in. The terms say you must be logged in with a Microsoft account to use Copilot.
• No bots or scrapers. The terms say not to use tools or computer programs, like bots or scrapers, to access Copilot. A free account is a manual instrument.
• Personal use. The same section says you can only use Copilot for your own personal use. If you are checking on behalf of a company, read that clause yourself before building a routine around it.

Three settings change what you measure, so record them on every run:

• Personalization and memory. Microsoft lets signed-in users turn this off under Memory, then Personalization and memory, on copilot.com, the Windows and macOS apps, the mobile app and the Edge sidebar. Use a dedicated measurement account with it off, so your own past conversations about your brand cannot shape the answer.
• Microsoft usage data. When this setting is on, Copilot uses data from Bing, MSN, Edge and other Microsoft products to personalize. Turn it off on the measurement account and note that you did.
• Grounding. Microsoft's transparency note says Copilot grounds certain information-seeking conversations in web search results, centering the response on high-ranking web content and listing hyperlinked citations. Not every answer is grounded, so record whether citations appeared. An answer with sources and one without are different measurements.

Conversations are saved by default and kept for 18 months unless you delete them, according to Microsoft's privacy FAQ. That is convenient as a backup, but do not treat the app's history as your record: paste each answer into your own sheet the moment it finishes.

Route 2: Bing's free AI Performance report

Bing Webmaster Tools describes itself as free, and its AI Performance report gives site owners first-party Copilot citation data. According to Microsoft's help page, read on 11 October 2026, it covers Microsoft Copilot, AI-generated summaries in Bing and select partner AI integrations, and shows:

• Total citations, the number of times your content was visibly referenced as a source;
• Cited pages and average cited pages per day;
• Grounding queries, short grouped phrases the AI used when retrieving content it cited;
• Page-level citation activity, with a mapping between grounding queries and pages.

Data refreshes daily with a short processing delay, and exports are available in CSV and Excel.

The limits matter as much as the features. Microsoft says the report does not measure rankings, authority or importance; it shows which content was cited. The data is a sample, very low activity may not surface, totals can differ between views, and it does not show individual answers, exact prompts or why a page was referenced. The help page lists three surfaces together and does not describe a way to isolate Copilot from the other two. And it only sees your own site: it cannot tell you which competitors were cited, or whether an answer named you without a link.

Use it alongside manual sessions, not instead of them. The report tells you which pages Copilot-family answers are citing and roughly how often; your sheet tells you what the answers said. Grounding queries can suggest wording for new manual questions, but they are grouped phrases, not real buyer questions, so write the final question yourself. Because the report reflects only content eligible for Bing's index, it is also a prompt to check Bing coverage; our comparison of how Copilot and ChatGPT answer business questions covers the Bing side in more depth.

Route 3: one-shot checkers and trials

These are readings of public pages on 11 October 2026, not trials, rankings or endorsements. Prices are left out because they change; check each pricing page.

• Ahrefs' free AI Visibility Checker says it needs no signup and queries ChatGPT, Gemini, Perplexity, Copilot and Google AI Overviews with prompts derived from real searches, and describes its report as a limited preview. You do not choose the prompts.
• Keyword.com offers a 14-day free trial with no credit card on its Copilot tracker page. Its FAQ says Microsoft does not expose a public API that returns the exact consumer Copilot answer.
• AIclicks offers a 3-day trial on its Copilot tracker page, with CSV export, and says you pay nothing if you cancel before day 3.
• ProRankTracker has a "Start for FREE" button on its Copilot tracker page that leads to its AI plans page, and reports "N/R" when a brand is neither mentioned nor cited.
• Orchly is named in a vendor roundup from Omnia as offering a free Copilot checker. Orchly's own home page offers a free analysis for ChatGPT, Gemini, Perplexity "and more", says Microsoft Copilot is tracked, and says all plans include a free trial.
• Searchable, in its own guide to Copilot tracking tools, says truly free Copilot tracking barely exists.

Before you rely on any checker or trial for Copilot, ask five questions: who chose the prompts, was the answer collected signed in and with personalization off, can you read the full answer, are all source links listed, and how does the vendor collect consumer Copilot answers given Microsoft's terms. Our guide to judging a visibility checker before trusting its score applies the same questions across engines.

What Microsoft's APIs can and cannot record

For some engines a pay-as-you-go API gives you history without manual pasting. The Microsoft APIs closest to that job do not return consumer Copilot answers:

• Microsoft retired the Bing Search APIs on 11 August 2025 and says existing instances were fully decommissioned.
• Grounding with Bing Search, the Azure AI tool that lets your own agent answer from Bing results, is available only on paid and pay-as-you-go Azure subscriptions and can incur costs.
• Which storage rules apply depends on how you access it. Microsoft's standard terms of use for the tool, last updated November 2025, prohibit copying, storing, caching or archiving its output, or creating a database of it. The same page says those terms do not apply to enterprise integrations such as Azure Foundry, which fall under separate enterprise terms instead. If you plan to use the tool through Foundry, check those enterprise terms before keeping any output.

The storage terms are not what rules this tool out for Copilot tracking. It answers with the model you deploy, not with Copilot, so even a saved run would be a related measurement rather than what a buyer saw in Copilot. For a record of Copilot itself, the routes that work are your own sheet and Bing report exports.

A spreadsheet workflow for free Copilot rank tracking

To choose the prompts and keep every whole answer at no cost, build the record yourself. This sheet captures the checks above and nothing else.

Column	What to record
Question	Exact wording, frozen; version it if it changes
Run	Date, time, country, account used
Settings	Personalization and usage data on or off
Grounded	Whether citations appeared under the answer
Full answer	The complete text, pasted
Brand named	Yes or no
Website cited	Yes or no, plus the exact URL
Source links	Every displayed link, in order
Recommended	Yes only when the answer clearly recommends you
Competitors	Every rival brand named
Notes	Limit messages, refusals, follow-up prompts

The run protocol:

1. Write ten questions a buyer would ask before choosing a product in your category: discovery, comparison and use-case questions.
2. Use the measurement account, start a new conversation for each question, paste it with nothing added, and copy the whole answer into the sheet as soon as it finishes.
3. Run the set twice in the same week. Answers vary between runs, and two runs show how much before you read anything into a change.
4. If the free version hits a limit, stop and finish later. Log the unanswered question as missing, not as a no.
5. Never overwrite a row. Next month adds rows.
6. Export the Bing AI Performance report for the same month and file it next to the sheet.

Compute rates against usable answers, not questions asked, using the requested, returned, usable and missing vocabulary from Prime AI Visibility's metrics page:

requested = questions x runs
returned  = answers received
usable    = on-topic answers
missing   = requested - returned

Brand named rate
  = answers naming the brand
  / usable

Website cited rate
  = answers linking your site
  / usable

A worked example with a fictional payroll app: ten questions run twice is 20 requested. A capacity message stopped two runs, so 18 returned; one answer asked a clarifying question instead of answering, so 17 are usable. The brand is named in 5 answers and linked in 2, a Brand named rate of 29 percent and a Website cited rate of 12 percent. With 17 usable answers, one answer moves a rate by about 6 points, so a month-on-month change of one answer is noise. Our editorial rule of thumb, not a statistical law, is three months of rows before a movement deserves a sentence in a report. To turn these counts into a share against competitors, the AI search share of voice measurement guide explains the formulas.

Is there a best free Copilot rank tracker?

The best free Microsoft Copilot rank tracker depends on the job, because the routes are not comparable on equal terms. Match the route to the job:

• Ten or fewer questions, monthly. Signed-in manual sessions, the sheet above, and a monthly Bing report export.
• You only need to know which pages Copilot cites. The Bing AI Performance report alone, read as a sample.
• A quick outside read before committing time. A one-shot checker, saved with its date and never compared directly with your own counts.
• You are evaluating paid trackers anyway. Load your frozen questions into the trial, ask how it collects Copilot answers, and export before it ends. Our evidence-based comparison of platforms with Copilot coverage sets out what to check.

The same pattern holds for other engines; our guide to free Claude tracking routes applies it to an engine that also offers a pay-as-you-go API route.

Where Prime AI Visibility fits, and where it does not

Prime AI Visibility does not support Microsoft Copilot on any plan.

What it does cover: the free Flash plan checks up to five buyer questions across ChatGPT, Claude and Perplexity in one completed check per calendar month, with up to 15 answers and no payment card. Gemini and an AI Overview-style preview built from Google-grounded answers are paid features, starting on Starter at $69/month, which also adds more saved questions and credits for repeat checks. Each report states requested, returned, usable and missing answers, with Brand named, Website cited, Source links and Explicit recommendation counted separately. Prime AI Visibility does not report a rank, estimate AI traffic or claim that a content change caused a movement. The full table is on the pricing page. The Prime AI Visibility features describe saved-answer checks for its supported engines, not Copilot coverage.

If Copilot matters to your buyers, keep the manual sheet and Bing exports for Copilot and measure the supported engines separately. Do not average the two into one number.

References

1. Microsoft, Microsoft Copilot Supplemental Terms of Use (effective 18 August 2026). https://www.microsoft.com/en-us/microsoft-copilot/for-individuals/termsofuse
2. Microsoft Support, Transparency Note for Microsoft Copilot (checked 11 October 2026). https://support.microsoft.com/en-us/microsoft-copilot/transparency-note-for-microsoft-copilot
3. Microsoft Support, Microsoft Copilot privacy controls and Privacy FAQ for Microsoft Copilot (checked 11 October 2026). https://support.microsoft.com/en-us/microsoft-copilot/microsoft-copilot-privacy-controls and https://support.microsoft.com/en-us/microsoft-copilot/privacy-faq-for-microsoft-copilot
4. Microsoft, Copilot pricing plans for individuals, FAQ "Is Copilot free?" (checked 11 October 2026). https://www.microsoft.com/en-us/microsoft-365-copilot/pricing/individuals
5. Bing Webmaster Tools Help, AI Performance in Bing Webmaster Tools (checked 11 October 2026). https://www.bing.com/webmasters/help/ai-performance-9f8e7d6c
6. Bing Webmaster Tools, About (checked 11 October 2026). https://www.bing.com/webmasters/about
7. Microsoft Lifecycle, Bing Search APIs retiring on August 11, 2025. https://learn.microsoft.com/en-us/lifecycle/announcements/bing-search-api-retirement
8. Microsoft Learn, Grounding with Bing Search (classic) (checked 11 October 2026). https://learn.microsoft.com/en-us/azure/foundry-classic/agents/how-to/tools-classic/bing-grounding
9. Microsoft, Terms of Use for Grounding with Bing Search and Grounding with Bing Custom Search (last updated November 2025, checked 11 October 2026; states it does not apply to enterprise integrations such as Azure Foundry). https://www.microsoft.com/en-us/bing/apis/grounding-legal
10. Ahrefs, Free AI Visibility Checker (vendor page, checked 11 October 2026). https://ahrefs.com/ai-visibility-checker
11. Keyword.com, Copilot tracker (vendor page, checked 11 October 2026). https://keyword.com/copilot-tracker/
12. AIclicks, Copilot visibility tracker (vendor page, checked 11 October 2026). https://aiclicks.io/trackers/copilot-visibility-tracker
13. ProRankTracker, Copilot Rank Tracker (vendor page, checked 11 October 2026). https://proranktracker.com/copilot-tracker
14. Omnia, Best Copilot rank tracking tools (vendor article, checked 11 October 2026), and Orchly, home page (vendor page, checked 11 October 2026). https://www.useomnia.com/blog/rank-tracking-tool-copilot and https://orchly.ai/
15. Searchable, Best Copilot tracking tools (vendor article, checked 11 October 2026). https://www.searchable.com/blog/best-copilot-tracking-tools
16. EyeFi Forum, Any recommendations for the best free Copilot rank tracker tool (thread dated 2 March 2026, checked 11 October 2026). https://eyefi.com/t/any-recommendations-for-the-best-free-copilot-rank-tracker-tool/2558
17. Prime AI Visibility, Pricing and Metrics and methodology (checked 11 October 2026). https://primeaivisibility.com/pricing and https://primeaivisibility.com/metrics

Next steps

If your saved answers reveal gaps in third-party coverage, digital PR services from Percepture are an implementation option, separate from a Copilot tracking tool. Percepture is commercially connected to Prime AI Visibility.

1. Compare platforms with Copilot coverage on evidence and fit once your manual sheet shows which questions matter.
2. Read Prime AI's metric definitions so your sheet and any tool report count the same things.
3. For the engines Prime AI Visibility supports, create a Prime AI Visibility workspace and run a free Flash check across ChatGPT, Claude and Perplexity.