Best LLM SEO tracking tools in 2026, and when the free AI visibility reports are enough

For one brand on a paid plan, SE Visible’s $99 Basic tier buys the most prompts per dollar (200, on five engines) of the self-serve LLM SEO tracking tools whose vendors explain how they collect answers,¹ and Otterly.AI at $29 a month is the cheap way to spot-check four engines.² Agencies should look at ZipTie, which charges a cent a check and never bills for seats. Semrush and Ahrefs only make sense if you’re already paying for the suite, and Profound doesn’t publish a brand price at all, so that one starts with a sales call. Before any of that, most small sites should open the free AI reports in Search Console and Bing Webmaster Tools, check GA4’s AI Assistant channel, and run a short list of prompts by hand for a month, which covers part of the same ground for nothing.

I don’t sell a tracker, and there are no affiliate links in this post. I didn’t contact any of these vendors or run a hands-on, head-to-head test, so the comparison sticks to what I could check for each LLM visibility tool on October 18, 2026: the vendor’s price, prompt allowance, engine list, and stated collection method, plus what 16 AI answers recommended when I asked them logged out.

19 trackers compared from their own vendor pages, Oct 18, 2026

8 of 15 top-ranking listicles that put the publisher’s own tool first

16 AI answers captured logged out on Oct 18, 2026

1 of 7 AI answers that priced Profound and matched Profound’s own page

The best LLM SEO tracking tools by situation, and when not to pay for one

These picks come from published specs and the run-to-run variance research, so treat them as a shortlist for trials.

SE Visible’s Basic plan comes to about $0.50 a prompt by my arithmetic (200 prompts for $99), the lowest of any self-serve plan whose vendor documents its method, and SparkToro’s research says a visibility percentage becomes a reasonable number across “dozens to hundreds of prompts run multiple times.”¹⁵

Otterly.AI Lite’s 15 prompts are well short of that SparkToro threshold, and adding AI Mode takes the bill to $38 a month.² I’d use it to check whether five or six money prompts mention the brand at all.

Claude needs its own line if you sell to businesses, since Goodie’s brand-averaged B2B panel put it at 18.5% of measurable AI referrals.¹⁶ Otterly.AI, Ahrefs, Peec AI (on Enterprise), Rankscale, and Keyword.com say Claude runs through the API, while Profound and AIclicks list Claude among the engines they query through the consumer interface,¹⁴¹⁷ and none of the interface-versus-API studies further down tested Claude, so I’d ask which method produced a Claude number before comparing it with the ChatGPT number beside it.

LLM SEO tracking tools compared: price, prompts, engines, and method

Every cell comes from the vendor’s own pricing, docs, or FAQ pages on October 18, 2026, and where a vendor’s pages contradict each other, the cell says so.

Hall isn’t listed. Its homepage is now an acquisition notice (“Hall is now part of Tracksuit,” dated 21 July 2026) and its pricing and docs pages return 404,³¹ though two of the ranking pages for this query still list it.

Which engines each tool tracks: ChatGPT, Gemini, Perplexity, and the rest, at its entry price

At minimum, LLM visibility tracking should cover ChatGPT, plus AI Overviews and AI Mode as two separate lines, and Claude too if you sell to businesses; most entry plans cover three to five engines, and Claude is enterprise-only, an add-on, or missing on most of them.

Ahrefs compared 540,000 query pairs in a December 2025 study and found only 13.7% of citations overlapped between AI Overviews and AI Mode, even though the answers scored 86% on semantic similarity, which is why the two need separate lines.³² Previsible found ChatGPT drove 92.4% of trackable LLM referral traffic (session share) across 166 GA4 properties,³³ and Goodie’s B2B panel has it at 62.6%¹⁶ (different populations, measured different ways).

Ahrefs’ March 2026 update found only 37.9% of URLs cited in AI Overviews also appeared in the top 10 for the same query, down from about 76% in its July 2025 study (Ahrefs says its parsing improved in between),³⁴ so the rank tracker you already have covers part of Google’s side, and you need a separate tracker to see how ChatGPT, Gemini, Perplexity, Copilot, and Claude answer.

ZipTie prices by engine count (“Price follows how many engines you run, never which ones”), so “Yes” there means available, and the $42.75 preset runs only ChatGPT, AI Overviews, and Perplexity; Profound’s row reflects a free trial on ChatGPT, Gemini, and AI Overviews, with the rest on Enterprise.

What a tracked prompt and a tracked answer cost

At list price the entry plans on most LLM visibility trackers land near $2 per tracked prompt a month. Among plans that state an answer allowance, one tracked answer costs anywhere from about 0.33 cents (SE Visible) to 8.2 cents (AthenaHQ), about 25 times as much, because the plans differ in how many engines each prompt runs on and how often it runs.

SE Visible Basic

0.50 5 engines

Peec AI Starter

1.90 3 engines

Surfer AI Search Analytics

1.90 5 engines

Otterly.AI Lite

1.93 4 engines

Radarkit Lite

1.93 6 engines, every 72 hours

AIclicks Starter

1.97 3 engines

Writesonic GEO Starter

1.98 3 engines

Scrunch AI Core

2.00 4 engines

Goodie AI Core

2.68 5 engines, billing cycle not stated

Semrush AI Visibility Toolkit

3.96 3 engines in Prompt Tracking

Peec AI’s FAQ works its own allowance out as “25x3x30 = 2250 AI answers,”²¹ and SE Ranking says 200 prompts on 5 LLMs daily is “approximately 30,000 checks per month.”¹ ZipTie’s unit is “One prompt, one engine, one check: $0.01,”¹² AthenaHQ’s is “1 credit = 1 AI response,”²⁸ Keyword.com charges 0.25 credits per chat-app run and 1 per API-model run,¹¹ and Ahrefs sells extra checks at $50 for 2,500, $100 for 7,000, and $250 for 25,000.²⁵

SE Visible Basic

0.33¢

ZipTie (published rate)

1.00¢

AIclicks Starter

1.27¢

Rankscale Pro

2.06¢

Peec AI Starter

2.11¢

Radarkit Lite

3.22¢

Scrunch AI Core

5.00¢

AthenaHQ Starter

8.19¢

The same arithmetic on plans that state no answer allowance is rougher. Otterly.AI comes to about 1.61 cents, Goodie AI 1.79, Surfer 1.27 at most because it runs several queries a day, and Keyword.com about 12.25 cents at 0.25 credits a chat-app run, which puts it above AthenaHQ. Nightwatch states 1,500 AI answers a month but prices in euros, which comes to about 6.6 euro cents an answer with a rank tracker included. Peec AI drops to 1.78 cents on annual billing. Semrush, Writesonic, and Ahrefs don’t make the engines per prompt, or the base plan behind the packages, clear enough to work out.

Semrush is $60 per 50 extra prompts and $99 per extra user or domain,⁶ one extra model on Peec AI Starter is $30 a month on annual billing,²¹ and Surfer’s content plans sell extra prompts in packs of 10 at $18 a month.³⁵ Tracking 75 prompts on Semrush comes to $159 a month.

Rankscale’s 4,800-answer ceiling assumes consumer-interface engines at 0.25 credits a run; API models cost 1 to 4 credits, so a prompt set weighted toward API models gets far fewer answers for the same $99.²⁶³⁶ AIclicks’ 4,650 responses also don’t reconcile with 30 prompts on 3 models for 30 days, which is 2,700. The page doesn’t explain it.

Prices you’ll see quoted that don’t match the vendor’s page

Every quoted figure in the left column below came from one of the 15 ranking pages or the 16 AI answers and differed from the vendor’s own page that day. Profound drew the most wrong guesses because it publishes no brand price, and among the prices vendors do publish, the biggest under-quotes were AthenaHQ Starter, $200 under list, and Otterly.AI Standard, $100 under.

Otterly.AI

100% 14 of 14

Semrush AI Visibility Toolkit

100% 12 of 12

Peec AI

90% 9 of 10

Profound

14% 1 of 7

How these were checked

I read every vendor’s pricing, docs, and FAQ pages on October 18, 2026 (in a browser, since some prices only load by JavaScript), and where a vendor’s own pages disagree, the tables say “Pages disagree.” I read the 15 pages ranking for this query the same day; 8 put the publisher’s own product first (AIclicks, SE Ranking, Surfer, and Writesonic among them),^41424344 and none published a prompt set, run counts, test dates, or raw results, with MarketerHire coming closest (a weighted rubric, no published scores). A few of these prices have probably moved since.

The picks favor tools that document how they collect answers, report each engine on its own line, and show a self-serve price, and I’d rather not pay for a tool whose headline metric is a rank position (the rank section further down says why). I wrote up the metrics these trackers report in an earlier post, so they get a line each here. Tools are grouped by buyer situation, then ordered by cheapest published price, with ties alphabetical and unpublished prices last.

Standalone AI visibility tools for a single brand

These six are separate subscriptions for tracking one brand, from $29 to $99 a month.

Otterly.AI

Otterly.AI’s Lite plan is $29 a month for 15 prompts tracked daily on ChatGPT, Google AI Overviews, Perplexity, and Microsoft Copilot, with Standard at $189 for 100 prompts and Premium at $489 for 400.² AI Mode and Gemini are $9 add-ons on Lite and Claude is $29. Queries go “through the website of the AI search,” except Claude, which runs on the API,⁵ and the daily schedule is fixed (“You can’t speed it up or slow it down”).⁴⁵ Lite can’t buy extra prompts (team members are unlimited, though), and the trial is 7 days with no card.⁴⁶

Radarkit

Radarkit is the one I’d look at for a local business or a shop, since it runs local checks “from a residential IP in the city you pick”⁴⁷ and its pricing table lists local, shopping, and ad tracking on every plan, Lite included.¹⁸ Lite is $29 a month for 15 prompts and 900 responses, refreshed every 72 hours, and Growth at $79 moves to 50 prompts daily.¹⁸ All six engines run through “Real chat interfaces, no APIs,”⁴⁷ and API and MCP come on every plan. None of the six is Claude. The 7-day trial takes no payment upfront, and Radarkit refunds a charge if you ask within 48 hours.

AIclicks

AIclicks (which ranks itself first on its own best-tools post,⁴¹ one of the pages ranking for this query) starts at $59 a month for 30 prompts, three models picked from a menu of ten, 4,650 responses, and one country.¹⁹ It queries platforms “through their actual user interfaces, not APIs.” Results update daily.¹⁷ GA4, the API, and MCP start on Pro at $189. The pages don’t say which of the ten models (Claude, Grok, DeepSeek, and Meta AI are among them) can be picked on Starter.

LLMrefs

LLMrefs sells one plan at $79 a month, tagged “Limited time only” with no end date or regular price.²⁰ It works from keywords (“you track keywords, not prompts”),²⁰ generates prompts from them across ten engines,⁴⁸ and includes unlimited seats and projects plus CSV and API access,²⁰ which suits an SEO team that already thinks in keyword lists. Before paying I’d ask what the plan holds, since the card says 500 prompts while the FAQ implies 50 keywords and the site gives two geo counts (20+ countries and 50+). I’d also ask how answers are collected, which LLMrefs doesn’t document beyond “Respectful, compliant access to AI engines and APIs following best practices.”⁴⁸

Peec AI

For an in-house team I’d trial Peec AI next to SE Visible, because seats aren’t capped and its docs are about the clearest in the group on method (“Peec AI uses UI scraping technology,” with the platform picking the model and deciding on web search the way it would “for a logged-out user”).⁴⁹ Starter is $95 a month, or $80 billed annually, for 50 prompts on three models chosen from six, 4,500 answers a month, and one project.²¹ Enterprise-only channels such as Claude and Grok are labeled API, Perplexity is an add-on, and one extra model is $30 a month on Starter, billed annually. Peec lists no trial length. With three engines at entry, you lose Copilot and Gemini if you pick ChatGPT and both Google surfaces.

SE Visible (SE Ranking)

SE Visible is SE Ranking’s standalone tracker. Basic costs $99 a month for 200 prompts, about 30,000 answers, three projects, and five engines (ChatGPT, Gemini, AI Mode, Perplexity, and AI Overviews), and because the limit counts prompts, all five engines come included.¹ It “collects real AI responses using a browser and graphical user interface.” Results refresh daily. Geography is limited to seven countries and five languages, Copilot isn’t tracked, and Claude “will be added soon.”¹ If you’re already an SE Ranking customer, the same 200 prompts come with the AI Search add-on at $89 a month, or $71.20 on annual billing, on top of a base plan such as Core at $129.⁵⁰

If you already pay for an SEO suite, content tool, or rank tracker

Keyword.com, Surfer, Nightwatch, Semrush, Writesonic, and Ahrefs sell AI prompt tracking as part of a bigger product, and the entry prices run from Keyword.com’s $9.80 credit plan to Ahrefs Lite at $129 a month.

Keyword.com

Keyword.com mostly suits someone who already tracks Google keywords there, since 360° Visibility is $33 a month for 1,000 keywords plus 15 AI credits, while the AI Visibility plan on its own is $9.80 a month ($7.83 annual) for 20 credits.¹¹ By Keyword.com’s own calculator formula, 20 credits cover 20 prompts on one chat-app engine refreshed weekly, or under 3 prompts refreshed daily,¹¹ and unused credits expire at the end of each cycle.⁵¹ Its pricing page and an April 2026 help doc list different credit costs per engine. Users and projects are unlimited and white label is a $29 add-on, though the API “is designed for high-level reporting and analytics, not for retrieving raw AI-generated responses.”⁵¹

Surfer AI Search Analytics

Surfer sells its AI Tracker on its own as AI Search Analytics, at $95 a month or $82 billed yearly, for 50 prompts on ChatGPT, Perplexity, AI Mode, AI Overviews, and Gemini, refreshed daily, with five seats.²² It says “we run multiple queries per day per model, then average the results to cut through the noise,” which, with Profound’s daily runs, is about as much as any vendor here says about handling run-to-run variance.⁵² The Standard content plan tracks 25 prompts on ChatGPT only, weekly,³⁵ and the 7-day trial “won’t cancel on its own.”²² Surfer’s product page also calls it “the only AI visibility tool that uses real scraping rather than an API,” which other vendors’ own documentation contradicts, since Peec AI, Otterly.AI, SE Visible, ZipTie, Rankscale, Radarkit, and AIclicks all describe collecting from consumer interfaces.

Nightwatch

Nightwatch’s Starter is €99 a month billed monthly or €79 billed yearly (it prices in euros), for 500 keywords, 50 AI prompts tracked daily, and 1,500 AI answers a month with no per-seat fees.¹⁰ If you need a daily rank tracker anyway, 50 prompts in the same dashboard is cheap. Its pages disagree on engines: the pricing page lists six including Claude and Gemini on every plan, while a help doc updated in February 2026 lists four and says Gemini needs a plan with 300+ prompts.⁵³ Nightwatch says it “simulates real user interactions,” updated daily alongside rank tracking.⁵⁴ White label and API start on Professional, at €159 a month billed yearly or €199 monthly.

Semrush AI Visibility Toolkit

Semrush’s database of over 317 million prompts, sourced from “AI search clickstream data and Google’s keyword dataset,” lets you look up any domain, a competitor’s included, before writing a prompt of your own,⁷ and the AI Visibility Toolkit it sits in is $99 a month per domain with 25 tracked prompts, $60 a month for each 50 more, and $99 for each extra user or domain. There’s no free trial.⁶ Prompt Tracking runs inside Position Tracking next to classic rankings on ChatGPT Search, AI Mode, and Gemini (one KB page adds AI Overviews, another doesn’t).⁵⁵ Brand Performance updates weekly. The Starter plan of Semrush’s SEO and AI Search bundle is $199 a month or $165.17 billed annually, with 50 prompts.⁵⁶ Semrush’s methodology page for its AI Visibility Index says “Differences between months should not be interpreted as growth or decline in brand visibility on their own,”⁵⁷ and I’d apply that to the toolkit’s charts as well. I’ve written separately about the Semrush and Ahrefs MCP servers for pulling this data into an assistant.

Writesonic GEO

Writesonic sells GEO Starter at $99 a month, or $79 a month billed annually.²³ That buys 50 prompts (the card says “50 prompts, 50 answers tracked daily,” which doesn’t obviously square with three engines), one user, and ChatGPT, Gemini, and AI Overviews only, with 15 AI articles and 10 site audits a month bundled in. Everything else needs Enterprise, which the pricing page says covers 9 platforms and the tracker page says 8. It scrapes the web interfaces daily,⁵⁸²⁴ and the tracker page says its prompt volume estimates are “pulled from 2B+ AI conversations.”⁵⁸ The bundled articles and audits make it a fit for a content team that would use the writer anyway.

Ahrefs Brand Radar

Ahrefs puts Brand Radar and 5 tracked AI prompts into its Lite plan, listed at $129 a month, which Ahrefs counts as 150 checks a month, or 5 prompts on one platform daily, and it sells full database access as Brand Radar AI “from $199/mo.”²⁵⁸ I’d buy it for category and competitor research, since five prompts on Lite is a token amount for daily tracking of your own. Database prompts are “derived from People Also Ask questions from our keyword database,” run through “the web version of each AI chatbot using the default model,” and refreshed “Once a month” for the chatbots, while custom prompts run daily.⁹ Claude goes through the API at 8 checks per update.⁸ Grok data is paused.⁹ The product pages served euros, so the dollar figures here come from the pricing page.

Best AI visibility tracking tools for agencies

Rankscale sells credits and ZipTie sells single checks, so with either one you can work out what each client costs you, and both offer white-label reporting (Rankscale from its $385 Growth plan).²⁶⁵⁹

Rankscale

Rankscale documents which engines it scrapes and which go through an API,⁶⁰ and cadence runs anywhere from hourly to monthly, though on October 12, 2026 its changelog noted that “Perplexity GUI scrapes are currently hitting login walls, so monitoring is temporarily paused.”³⁶ Its Essentials plan is $17 a month billed yearly only, with limits that don’t render on the page, so the realistic entry is Pro at $99 a month for 1,200 credits, up to 4,800 answers, and 10 brand dashboards.²⁶ A run on a consumer-interface engine typically costs 0.25 credits and API models cost 1 to 4, unused credits roll over up to 2x on Pro, and REST API access starts at Pro+, which is $229 a month.²⁶³⁶

ZipTie

ZipTie has no tiers. One prompt on one engine for one check is $0.01, or $0.0087 a check on all seven engines, and the cheapest preset on the page, 50 daily prompts on ChatGPT, AI Overviews, and Perplexity, comes to $42.75 a month.¹² Seats, projects, and competitors are unlimited and never billed, and agencies get a white-label client report. It captures “from the live consumer products rather than a model API” in “fresh, neutral sessions,” and it dropped the position metric on purpose (“Because the order inside an AI answer does not hold still.”) in favor of a 0 to 100 AI Success Score per prompt.⁵⁹ Limits: API tokens are per project, and there’s no on-demand recheck and no refunds. Reports only get made when someone asks. Claude isn’t among the seven engines. ZipTie used to be ziptie.dev, which now redirects to ziptie.ai.

Profound sells client workspaces at $399 a month apiece, each with 100 prompts on ChatGPT, Perplexity, and AI Overviews, but shows no base price for its Agency Growth plan,¹³ and Scrunch AI lists Agency Core at $500 a month.⁶¹ For multi-client setups on the self-serve plans, SE Visible Basic includes 3 projects, Peec AI gives 1, 2, and 5 projects by plan, LLMrefs allows unlimited projects, and Otterly.AI Lite is one workspace.

Mid-market trackers that bundle site and content fixes

Scrunch AI, AthenaHQ, and Goodie AI start at $250 to $295 a month and sell optimization work alongside the tracking.

Scrunch AI

Scrunch AI starts at $250 a month on Core, which Scrunch’s own page describes as a plan “For teams to evaluate Scrunch,” for 125 prompts, 5,000 responses, five seats, one country, and four engines (ChatGPT, Perplexity, AI Overviews, and Copilot).²⁷ It collects through “browser automation and official platform APIs” without saying which engine gets which, daily for the first 14 days and every 72 hours after that.⁶² AXP, which serves an AI-optimized version of your site to agents at the CDN, is Enterprise only. The pricing page doesn’t label the $250 as monthly or annual, though Scrunch’s FAQ says per month.⁶¹

AthenaHQ

AthenaHQ’s Starter is $295 a month for 3,600 credits at “1 credit = 1 AI response,” across 11 models including Claude, Grok, DeepSeek, Meta AI, and Mistral, which is more than most self-serve plans here (Rankscale claims 17+ on every plan²⁶).²⁸ Prompts are listed as unlimited, but each response spends a credit, so the cap is 3,600 responses. The docs don’t say whether answers come from consumer interfaces or model APIs, and the only cadence they document is weekly or monthly.⁶³ A free Essential plan comes with 300 credits. ACE, its tool for predicting which content gets cited, is Enterprise only.

Goodie AI

Goodie AI, which publishes the B2B referral study cited earlier, also starts at $295, on Core, for 110 prompts on ChatGPT, AI Overviews, Perplexity, AI Mode, and Copilot, with an MCP server, though the page has a billing toggle and doesn’t say which cycle $295 reflects.²⁹ Seats are unlimited. Gemini doesn’t arrive until Pro at $999 (along with SKU-level shopping visibility), and collection runs daily by a method the pages don’t document.

Enterprise trackers sold through sales

BrightEdge and Profound don’t publish a price for brands, so both start with a demo or a sales call.

BrightEdge AI Catalyst

AI Catalyst’s /pricing URL returned a 404 and the only way in is a demo request, so in practice it’s an option for teams already on BrightEdge.³⁰ It has been “available now for all customers” since April 9, 2025,⁶⁴ names five engines (“ChatGPT, Perplexity, Google AI Overviews, Google AI Mode, Gemini and other AI platforms”), and doesn’t document how it collects answers or how often.

Profound

Profound offers brands a free 7-day trial (50 prompts run daily on ChatGPT, Gemini, and AI Overviews, from a recommended set you can’t edit) and an Enterprise tier marked “Custom” with up to 9 engines.¹³ It captures responses “directly from the consumer experience, not API outputs” and runs every tracked prompt daily “so your visibility score reflects your true average across all responses,”¹⁴ and its Prompt Volumes demand data comes from licensed “double-opt-in consumer panels.”³⁷ The pages disagree on which ninth engine it tracks, Grok on the product FAQ and Exa Search on the pricing table, and on panel size, “1.5 billion+” conversations on one page and “2B+” on another. Profound says it serves over 1,000 enterprise brands,⁴⁰ and its own comparison page lists a con to read before the sales call, that Profound “doesn’t replicate traditional SEO tooling like backlink analysis or keyword rank tracking.”³⁸

In late 2025 HubSpot agreed to buy XFunnel⁶⁵ and Peec AI raised a $21M Series A.⁶⁶ This year Adobe closed its Semrush acquisition on April 28,⁶⁷ Hall went to Tracksuit in July, and Profound raised $96M in February³⁹ and $180M on September 15,⁴⁰ so before signing a 12-month contract I’d ask what happens to the account and its history if the vendor is bought.

Why two trackers report different numbers for the same brand

AI answers change from run to run, and what an API returns differs from what a person sees in the app, so two trackers can report different numbers for the same brand, and the first things I’d check on any LLM visibility tool are how it collects answers and how many runs sit behind each number. I went through why answers vary at all in how retrieval and query fan-out pick sources.

Consumer interface or API

An arXiv preprint auditing AI product recommendations (the one independent study here) found mean domain overlap between API output and the consumer interface of 12.0% for ChatGPT and 14.8% for Gemini.⁶⁸ The vendor studies agree, and each comes from a company with an interest in the result. Surfer, which sells a scraping tracker, found brand overlap of 15.5% to 23.8% across 13,779 answers and domain overlap as low as 4.8% on ChatGPT.⁶⁹ Brandlight found 6% to 12% brand overlap across 2,565 matched pairs,⁷⁰ and Genezio found the GPT-5.2 API and ChatGPT.com ran zero shared search queries, with 12.3% of sources used by both.⁷¹

How many runs a visibility number needs

SparkToro’s Rand Fishkin and Patrick O’Donnell had 600 volunteers run 12 prompts through three tools a combined 2,961 times and found under a 1 in 100 chance of the same brand list coming back twice, and about 1 in 1,000 of it coming back in the same order, while human-written prompts with the same intent had a semantic similarity of just 0.081.¹⁵ O’Donnell joined Gumshoe (which sells AI tracking) in 2025, and the post discloses it. Fishkin later walked through a sampling method Gumshoe published: one pass of 10 prompts in 7 phrasings across 7 models (about 490 answers) carries about plus or minus 5 points, and it takes about 20 passes to get to plus or minus 1.⁷²

A reanalysis of five audit studies found five runs reliable across a portfolio of 250 questions (G = 0.96) and unreliable for five versions of a single question (G = 0.27), though it’s an arXiv preprint, last revised October 14, 2026, and I haven’t seen it published in a journal.⁷³ Conductor, a vendor working through APIs, found purchase prompts shared only 4 of 10 brands across two runs while comparison prompts kept the same leader 91% of the time,⁷⁴ and AirOps (another vendor) reports that “Only 30% of brands stay visible run to run.”⁷⁵ Ahrefs found AI Overviews have a 70% chance of changing between observations, last 2.15 days on average, and share 54.5% of URLs between consecutive responses.⁷⁶

Profound runs every prompt daily and Surfer averages several runs a day, while Radarkit Lite runs every 72 hours and Scrunch AI drops to that after its first 14 days; most vendors haven’t published how many runs each prompt gets per refresh.

What an AI rank tracker’s “rank” means

Semrush’s US data on October 18 had “ai rank tracker” at 1,000 searches a month, “free chatgpt rank tracking” at 720, “chatgpt rank tracker” at 590, “llm rank tracker” at 480, and “how to track chatgpt ai rankings over time” at 320, so a lot of people are shopping for a position number. SparkToro concluded that “any tool that gives a ‘ranking position in AI’ is full of baloney,” and called visibility percentage across many prompts run many times “a reasonable metric.”¹⁵ ZipTie dropped position for the same reason (“the order inside an AI answer does not hold still”), and I’d report share of answers across a portfolio of prompts instead, even in a tool like Semrush whose plan card still advertises “daily AI rankings.”⁵⁶

What six AI answer engines recommended when I asked them to pick a tracker

In this one capture, Profound and Semrush were the trackers named most often, each in 15 of the 16 logged-out answers, and a rerun minutes later could give a different count. I put three prompts into six AI surfaces on October 18, 2026, between 22:32 and 22:45 UTC, all logged out: “best LLM SEO tracking tools,” “what is the best tool to track my brand’s visibility in ChatGPT and AI search,” and “best AI visibility tracking tools for agencies.” The surfaces were Google AI Overviews and AI Mode (run with gl=us), Perplexity, ChatGPT search, Bing, and Brave. Perplexity answered once and then asked for a sign-up, and Brave’s summarizer fired once with its Ask mode covering the other two prompts, which left 16 answers (ChatGPT described its own picks as “not a controlled head-to-head performance test”).

Profound

15 on 12 of 15 ranking pages

Semrush

15 11 of 15

Otterly.AI

14 7 of 15

Peec AI

13 10 of 15

Ahrefs

10 5 of 15

SE Ranking / SE Visible

5 9 of 15

Radarkit

5 0 of 15

Scrunch AI

5 4 of 15

AthenaHQ

4 6 of 15

Rankscale

3 1 of 15

LLMrefs

3 2 of 15

ZipTie

2 4 of 15

Peec AI was named in 13 and Ahrefs in 10. Otterly.AI came up in 14 answers but on only 7 of the 15 ranking pages, though Zapier’s list, cited in 5 answers, puts OtterlyAI first and labels it for affordability.⁷⁷ Radarkit came up in 5 answers and on none of them. Google’s AI Overview for “best LLM SEO tracking tools” barely overlapped with the other engines and attributed 4 of its 7 picks to one source, Rankability. Bing named RankScope “Best overall” while citing RankScope’s own blog, and on the agency prompt four of Bing’s five sources were vendor blogs, from VisibilityStack, Profound, ZeroRank, and Semrush, and all four vendors made the list. Reddit was the most-cited source, in 8 answers, then semrush.com in 7, and zapier.com and youtube.com in 5 each.

AI Mode merged two different companies into “Peak AI / Peec AI,” and two Google answers gave Peec AI “up to 11” engines where Peec’s Enterprise card says “Up to 13.” Semrush’s database was “over 317 million” prompts on one Google surface and “over 26 million” on the other, two minutes apart.

Free ChatGPT rank tracking and the other free reports

For ChatGPT, the free options are mostly one-off checks, such as the Ahrefs AI Visibility Checker (no signup, top 5 results on six engines),⁷⁸ while Search Console, Bing Webmaster Tools, and GA4 show part of the rest for free and most paid trackers offer a trial of 3 to 14 days.

The generative AI report in Search Console counts impressions in AI Overviews, AI Mode, and Discover’s AI features, and as of August 31, 2026 it’s on every site worldwide.³ I covered what Search Console’s generative AI report leaves out in the August roundup. Bing Webmaster Tools’ AI Performance report covers Copilot, AI summaries in Bing, and select partners, and Microsoft calls the data “a sample of overall citation activity.”⁴ It added Citation Share on June 16, 2026,⁷⁹ then a Topics Portfolio beta on October 13 with the advice to “Treat the numbers as directional signals for planning, not as a precise ledger.”⁸⁰ GA4’s default AI Assistant channel picks up referrals from ChatGPT, Gemini, Deepseek, Copilot, and Grok, and it “excludes Google’s AI Overviews and AI Mode.”⁸¹

For a quick look at prompts beyond the Ahrefs checker, SE Ranking gives “5 free checks per day,”⁸² and AthenaHQ’s Essential plan is free with 300 credits.²⁸ The trials, as the vendor pages listed them on October 18:

• Otterly.AI: 7 days, no card⁴⁶
• ZipTie: 7 days, 25 prompts, no card¹²
• Keyword.com: 14 days, no card; a lapsed trial account is deleted after 30 days¹¹⁵¹
• Nightwatch: 14 days, no card, bills automatically¹⁰
• SE Visible: 10 days¹
• Profound: 7 days, 50 prompts you can’t edit¹³
• Scrunch AI: 7 days, card required, upgrades automatically⁸³
• Surfer: 7 days, won’t cancel on its own²²
• Radarkit, Rankscale Pro, LLMrefs, and Goodie AI Core: 7 days^18262029
• AIclicks: 3 days¹⁹
• Semrush AI Visibility Toolkit: none, per its knowledge base;⁶ Peec AI doesn’t say

Before paying anyone, check which AI crawlers your robots.txt allows, since a tracker will show you missing from answers even when the cause is a crawler you blocked yourself. The other free tool on this site lets you see how much of a page AI crawlers can read in raw HTML. It only checks readability, so it won’t tell you whether any engine cites you.

When a paid tracker is worth the money

I’d pay for a tracker once the brand shows up in enough answers, or enough clients ask about it, that a hand-run prompt list stops being practical. (I made a similar argument about expensive AI SEO tool suites earlier this year.) Harry Clarkson-Bennett at Search Engine Journal found “almost zero value” in tracking prompts “at a purely answer level,” though he still concluded “if the traffic converts well and you need to know more, get tracking,” and he priced a DIY tracker at about $96 a month in October 2025.⁸⁴ Backlinko says prompt tracking is “NOT worth it” when your audience isn’t using AI to find solutions in your category or you don’t publish content regularly.⁸⁵

Previsible’s industry breakdown puts LLM referrals at 0.08% of sessions for publishers and 1.71% for SMBs,³³ which I think undersells what a mention is worth. Seer’s data on 53 brands, tracked January 2025 to February 2026 and published in April 2026, showed brands cited in an AI Overview got 120% more organic clicks per impression on informational queries than when they weren’t,⁸⁶ and in Similarweb data presented by SparkToro, people were 14.2% more likely to visit Capital One when an AI recommended it.⁸⁷

Questions to ask before you sign up

1. How many runs does each prompt get, and is the reported number an average? In Gumshoe’s published method, as SparkToro described it, one pass of about 490 answers carries about plus or minus 5 points of error.⁷²
2. Which engines run through the consumer interface and which through an API? Rankscale and Keyword.com say so per engine; most vendors don’t.
3. Are answers collected logged out or logged in? Peec AI documents logged-out sessions, ZipTie uses “fresh, neutral sessions,” and Ahrefs runs prompts without stored user data.⁹
4. Which countries and languages? SE Visible covers 7 countries and 5 languages,¹ Radarkit runs residential IPs in 50+ locations for local checks,¹⁸⁴⁷ and Profound claims 150+ regions.¹⁴
5. Where do suggested prompts and “prompt volume” come from? Ahrefs uses People Also Ask, Semrush uses clickstream and Google keyword data, Profound uses licensed panels, and LLMrefs fans out your keywords, while human prompts with the same intent scored 0.081 on similarity in SparkToro’s data.
6. Can I export the raw answers? Keyword.com’s API leaves them out.
7. Will the vendor be here next year? Hall moved to Tracksuit on July 21, 2026 and its app link no longer resolves.³¹

What these trackers don’t measure

Sentiment scores have no benchmark and no agreed method behind them, so I’d read the answers on your key prompts myself before trusting a gauge. No tracker counts the people who read an answer and never click. As far as their own pages say, none of these tools sees what a logged-in customer sees either, with chat history and memory shaping the answer; the three vendors that describe their session setup run logged out (Peec AI), in “fresh, neutral sessions” (ZipTie), or “without any stored user data or prior context” (Ahrefs),⁹ and none of the others says it queries as a signed-in user with history.

Questions about LLM SEO tracking tools

Is it possible to track brand mentions in AI search?

Yes, as a sampled estimate. Paid trackers run a fixed set of prompts on a schedule and count how often you're named, Bing Webmaster Tools reports Copilot citations as a sample, and Search Console reports impressions in AI Overviews and AI Mode. SparkToro found under a 1 in 100 chance of the same brand list coming back twice, so one reading can be well off and the number is only useful as a trend across many runs.

Is there a free LLM rank tracker?

No free tool does the whole job. The closest options are Search Console's generative AI report, Bing's AI Performance report, the Ahrefs AI Visibility Checker (no signup, top 5 results), SE Ranking's 5 free checks a day, and AthenaHQ's free Essential plan with 300 credits. Most paid trackers also offer trials of 3 to 14 days, and several of them don't ask for a card.

How do you track ChatGPT rankings over time?

Run a fixed set of prompts on a schedule, several times each, and track the share of answers that mention you inside one tool, since the order of brands changes from run to run and there isn't a stable rank to follow. In Gumshoe's published sampling method, as SparkToro's Rand Fishkin described it, one pass of about 490 answers carries plus or minus 5 points, and about 20 passes get it to plus or minus 1.

What is prompt volume in AI visibility tools?

It's a vendor's estimate of how often people ask something like a given prompt. Profound licenses consumer panel data, Ahrefs models prompts from People Also Ask questions, Semrush uses clickstream and Google keyword data, Writesonic cites 2B+ AI conversations, and LLMrefs gives monthly estimates. None of the vendors says its counts come from the AI companies themselves.

What is an AI visibility score?

A composite number each vendor builds its own way, for example ZipTie's 0 to 100 AI Success Score or Profound's daily average across responses. Because every vendor defines it differently and nobody publishes a benchmark for a good score, compare a score only with itself over time inside one tool.

How do I track AI Mode in Google Search Console?

Open the generative AI performance report. It shows how often URLs from your site appeared in AI Overviews, AI Mode, and Discover's AI features, counted as impressions, and Google rolled it out to all sites worldwide as of August 31, 2026. Google says the same data stays included in the overall performance report as well.

How do I track LLM traffic in GA4?

Use the default AI Assistant channel, which groups referrals from ChatGPT, Gemini, Deepseek, Copilot, and Grok, and excludes Google's AI Overviews and AI Mode. ChatGPT will usually be most of it, since Previsible found it drove 92.4% of trackable LLM referral traffic, by session share, across 166 GA4 properties, though Goodie's B2B panel had it at 62.6%.

Which AI visibility tools offer white-label reports for agencies?

On the vendor pages checked on October 18, 2026, ZipTie includes a white-label client report, Nightwatch offers white label from its Professional plan up, and Keyword.com sells it as an add-on at $29 a month or $278 a year. Rankscale lists white-label options from its $385 Growth plan, Writesonic lists white-label reporting under its Agency Enterprise plan, and SE Visible's FAQ says white labeling isn't available yet. The other trackers in this comparison didn't state it on the pages checked.

More on measuring AI search

Earlier posts on the metrics these trackers report, why AI answers change between runs, the AI tool suites I'd skip, and the Semrush and Ahrefs MCP servers.

Measurement AI Search Visibility Metrics: What They Are and Why They Matter Now The KPIs that measure whether ChatGPT, AI Overviews, Perplexity and Gemini cite your brand. What each metric tells you, what the data shows, and which ones are worth tracking. AI Search What Is RAG (and Why It Decides Whether AI Cites You) RAG is the retrieval step behind every AI answer engine. Here is how it works, and what 1.2 million citations show about the pages it actually quotes. Content Strategy 5 SEO Trends I'm Ignoring in 2026 (and Why) Not every trend deserves your time. Five SEO tactics getting loud airtime in 2026 that I'm deliberately skipping, with the data behind each call, so you can spend effort where it moves rankings. AI Search What Is MCP (and What Is It Good for in SEO)? MCP is how AI assistants plug into your SEO tools: Search Console, Semrush, Ahrefs, analytics, crawls. What the protocol is, who supports it, how to set it up, and which parts of the hype to skip.

Sources

1. SE Ranking. SE Visible. Basic $99/month, 200 prompts, about 30,000 AI answers a month, 3 projects, five engines; browser and GUI collection, daily refresh; 7 countries and 5 languages; Claude “will be added soon”; white labeling “Not yet” in the FAQ; 10-day trial. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

2. Otterly.AI. Pricing. Lite $29/month (15 prompts), Standard $189 (100), and Premium $489 (400); four base engines; AI Mode and Gemini $9 and Claude $29 on Lite; unlimited team members. The page also shows a second price set with ChatGPT Ads Tracking (Lite $25, Standard $160, Premium $422); the figures here are the set without it. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

3. Google Search Central Blog. Introducing Search generative AI performance reports in Search Console (June 3, 2026, with an August 31, 2026 note on the rollout to all websites worldwide). ↩ ↩²

4. Bing Webmaster Blog. Introducing AI Performance in Bing Webmaster Tools (public preview) (February 10, 2026). “The data shown represents a sample of overall citation activity.” ↩ ↩²

5. Otterly.AI Help. How OtterlyAI collects data. Prompts posted “through the website of the AI search”; “we only use API for Claude tracking.” ↩ ↩²

6. Semrush Knowledge Base. AI Visibility Toolkit. $99 per month, 25 prompts, $60 per 50 extra prompts, $99 per extra domain, $99 license per subuser, no free trial for the standalone toolkit; refresh schedule by report. The AI Visibility pricing page showed “$99/mo per domain” on both its monthly and annual toggles on October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

7. Semrush Knowledge Base. Semrush AI visibility data. Over 317 million prompts; sourced from “AI search clickstream data and Google’s keyword dataset”; responses “captured from real requests and not via any APIs of LLMs.” ↩ ↩² ↩³

8. Ahrefs. Brand Radar. 481 million+ prompts; plan check allowances (Lite: “5 prompts on 1 platform = 150 checks/mo”); “Claude consumes 8 checks per update.” Prices on this page were served in euros. ↩ ↩² ↩³ ↩⁴

9. Ahrefs Help. About Brand Radar. Prompts derived from People Also Ask questions; web version of each chatbot; prompts run “without any stored user data or prior context”; chatbot data refreshed “Once a month”; Grok paused; Claude via API. ↩ ↩² ↩³ ↩⁴ ↩⁵

10. Nightwatch. Pricing. Starter €79/mo billed yearly (€948), €99 billed monthly per the page’s plan data; 500 keywords, 50 AI prompts, 1,500 AI answers a month; no per-seat fees; 14-day trial, no card, auto-bills. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴

11. Keyword.com. Pricing. AI Visibility $9.80/month for 20 credits ($7.83 annual); 360° Visibility $33/month for 1,000 keywords and 15 credits; chat apps 0.25 credit per run, API models 1; unlimited users and projects; white label $29/month or $278/year; 14-day trial, no card. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

12. ZipTie. Pricing. $0.01 per check, $0.0087 on all 7 engines; Starter preset from $42.75/mo; unlimited seats, projects, and competitors; 7-day trial on 25 daily prompts, no card; no refunds. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵

13. Profound. Pricing. Free 7-day trial, 50 prompts on ChatGPT, Gemini, and AI Overviews; Enterprise “Custom,” up to 9 engines; agency client workspaces $399/month. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

14. Profound. Answer Engine Insights. Front-end capture (“We query the front-end experiences that normal users see when they visit AI platforms,” Claude among them); “We run every tracked prompt daily so your visibility score reflects your true average across all responses”; “30+ languages and 150+ regions”; engine list on the product FAQ. ↩ ↩² ↩³ ↩⁴

15. SparkToro. New research: AIs are highly inconsistent when recommending brands or products (January 28, 2026). 600 volunteers, 12 prompts, 3 tools, 2,961 runs; research partner Patrick O’Donnell joined Gumshoe.ai in 2025, disclosed in the post. ↩ ↩² ↩³

16. Goodie. AI Search Traffic Report 2026 (May 21, 2026). Brand-averaged B2B AI referrals: ChatGPT 62.6%, Claude 18.5%. ↩ ↩²

17. AIclicks. Homepage. Ten engines listed, Claude among them; “through their actual user interfaces, not APIs”; daily updates. ↩ ↩²

18. Radarkit. Pricing. Lite $29/month, 15 prompts, 900 responses, every 72 hours; Growth $79, 50 prompts daily; six engines; local business, shopping, and ads tracking listed on every plan, Lite included; MCP and API on all plans; extra seats “No” on Lite; 7-day trial; refund within 48 hours. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵

19. AIclicks. Pricing. Starter $59/month, 30 prompts, choose 3 models, 4,650 responses, 1 country; Pro $189 with GA4, API, and MCP; 3-day trial. Checked October 18, 2026. ↩ ↩² ↩³

20. LLMrefs. Homepage and pricing. $79/month “All in One,” “Limited time only”; “With LLMrefs you track keywords, not prompts”; 500 prompts on the card, 50 keywords in the FAQ; “Unlimited team members in your organization,” unlimited projects, “Export data as CSV and API access”; 7-day trial. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵

21. Peec AI. Pricing. Starter $95/mo monthly or $80/mo annual, 50 prompts, 3 models, 4,500 answers, unlimited users, 1 project; Pro $245; Enterprise up to 13 models; extra model $30/mo on Starter (annual billing); FAQ answer-count example. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

22. Surfer Docs. AI Search Analytics (October 16, 2026). $95 monthly or $82/month yearly for 50 prompts; five engines; daily; 5 seats, 5 workspaces; 7-day trial that won’t cancel on its own. ↩ ↩² ↩³ ↩⁴

23. Writesonic. Pricing. Starter $99/month billed monthly or $79/month billed annually, 50 prompts, 1 user, ChatGPT, Gemini, and AI Overviews; Enterprise “All 9 AI platforms”. Checked October 18, 2026. ↩ ↩²

24. Writesonic Docs. Why we monitor real AI platforms, not just APIs. “We scrape the actual web interfaces of AI platforms.” ↩ ↩²

25. Ahrefs. Pricing. Lite $129/mo with Brand Radar and 5 tracked AI prompts; Brand Radar AI from $199/mo; 492 million+ prompts; check packages $50, $100, and $250. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵

26. Rankscale. Pricing. Essentials $17/mo yearly only ($204); Pro $99/mo, 1,200 credits, up to 4,800 answers, 10 brand dashboards; typical 0.25 credits per run; rollover up to 2x on Pro; REST API from Pro+ ($229); “white-label options” on the $385 Growth plan (“Remove Rankscale branding from shared dashboards”); 7-day Pro trial; pricing FAQ: “Rankscale monitors 17+ AI engines” and “All engines are available on every plan.” Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

27. Scrunch. Pricing. Core $250/mo, 125 prompts, 5,000 responses, 5 user licenses, 4 engines, 1 country; AXP on Enterprise only. Human-facing page, checked October 18, 2026. ↩ ↩²

28. AthenaHQ. Pricing. Starter $295/month, 3,600 credits, “1 credit = 1 AI response,” 11 models; free Essential plan with 300 credits. Checked October 18, 2026. ↩ ↩² ↩³ ↩⁴ ↩⁵

29. Goodie. Pricing. Core $295/month, 110 prompts, 5 engines, unlimited seats; Pro $999 with Gemini; 7-day Core trial. Checked October 18, 2026. ↩ ↩² ↩³

30. BrightEdge. AI Catalyst. Engines named on the product page; demo only (brightedge.com/pricing returned 404 on October 18, 2026). ↩ ↩²

31. Hall. Homepage. “Hall is now part of Tracksuit,” dated 21 July 2026; pricing and docs pages returned 404 and the app link did not resolve on October 18, 2026. ↩ ↩²

32. Ahrefs. AI Overviews vs AI Mode (December 15, 2025). 540,000 query pairs; 13.7% citation overlap; 86% semantic similarity. ↩

33. Previsible. AI Traffic Report (July 6, 2026). 166 GA4 properties, November 2024 to May 2026; ChatGPT 92.4% of trackable LLM referral traffic, by session share; LLM traffic about 0.08% (publishers) to 1.71% (SMB) of sessions. ↩ ↩²

34. Ahrefs. Update: 38% of AI Overview Citations Pull From The Top 10 (March 2, 2026). 863K keyword SERPs, 4M AI Overview URLs; 37.9% of cited URLs in the first 10 SERP blocks, 31.0% beyond the top 100; about 76% in the July 2025 study; Ahrefs notes it “improved our parsing methodology” since then. ↩

35. Surfer Docs. AI Tracker. Standard: “25 prompts, ChatGPT only, weekly refresh”; on Standard, Pro, and Peace of Mind, extra prompts come “in packs of 10 for $18 per month each.” ↩ ↩²

36. Rankscale. Changelog. October 12, 2026: Sonnet 5.5 at 3 credits, or 4 with web search; Perplexity GUI monitoring paused because of login walls. ↩ ↩² ↩³

37. Profound. Prompt Volumes. Licensed double-opt-in consumer panels; “1.5 billion+ real AI conversations.” ↩ ↩²

38. Profound. Profound vs Ahrefs. “2B+ real user conversations”; the con about backlink analysis and keyword rank tracking. ↩ ↩²

39. Profound. Profound raises Series C at $1B valuation (February 24, 2026). $96M Series C at a $1B valuation. ↩ ↩²

40. Profound. Profound raises $180M Series D at $1.8B valuation (September 15, 2026). Also the source for “over 1,000 enterprise brands.” ↩ ↩² ↩³

41. AIclicks. Best LLM SEO analysis tools. AIclicks is entry 1. Read October 18, 2026. ↩ ↩²

42. SE Ranking. Best LLM tracking tools (April 22, 2026). SE Ranking is entry 1. Read October 18, 2026. ↩

43. Surfer. Best AI SEO tools (last updated September 17, 2026). Surfer AI Tracker is entry 1. Read October 18, 2026. ↩

44. Writesonic. LLM tracking tools (updated March 13, 2026). Writesonic is entry 1, labeled “#1 LLM Tracking Tool.” Read October 18, 2026. ↩

45. Otterly.AI Help. Monitoring interval. Daily monitoring, “You can’t speed it up or slow it down.” ↩

46. Otterly.AI Help. Free trial. 7-day free trial, no credit card required. ↩ ↩²

47. Radarkit. Homepage. “Real chat interfaces, no APIs”; residential IP local checks; local, shopping, and ads tracking. ↩ ↩² ↩³

48. LLMrefs. AI search visibility. Ten supported engines; keyword-first fan-out prompts; method statement. ↩ ↩²

49. Peec AI Docs. Intro to Peec AI. “Peec AI uses UI scraping technology”; logged-out behavior; daily runs. ↩

50. SE Ranking. Subscription plans. AI Search add-on $89.00 monthly or $71.20 annual, including SE Visible (200 prompts); Core $129.00 monthly. ↩

51. Keyword.com Support. How AI Visibility Credits Work (April 17, 2026): “Unused credits expire at the end of the cycle,” per-engine credit costs; AI Visibility API (March 1, 2026): “The API is designed for high-level reporting and analytics, not for retrieving raw AI-generated responses”; Billing, Plans and Subscriptions: “Deactivated trial accounts are permanently deleted after 30 days.” ↩ ↩² ↩³

52. Surfer. AI Tracker. “We run multiple queries per day per model, then average the results”; the “only AI visibility tool that uses real scraping” claim. ↩

53. Nightwatch Docs. AI/LLM tracking in Nightwatch (updated February 27, 2026). Four providers listed; “Gemini is supported for plans with 300 prompts +.” ↩

54. Nightwatch. AI tracking. “Nightwatch simulates real user interactions with each major LLM platform”; updated daily alongside rank tracking. ↩

55. Semrush Knowledge Base. Prompt Tracking. Platforms selectable per campaign: ChatGPT Search, Google AI Mode, Gemini; daily updates; AI Overviews not listed. Checked October 18, 2026. ↩

56. Semrush. AI Visibility pricing, plan card “Track 25 custom prompts (daily AI rankings)”; SEO and AI Search pricing, Starter (SEO and AI Search) $199 monthly or $165.17/month billed annually, “50 prompts to track daily”. Checked October 18, 2026. ↩ ↩²

57. Semrush. AI Visibility Index methodology. ↩

58. Writesonic. AI Visibility Tracker. “We crawl the actual AI interface”; daily; “8 platforms” on this page; prompt volume “pulled from 2B+ AI conversations.” ↩ ↩²

59. ZipTie. FAQ. Live consumer products in fresh, neutral sessions; no position metric, replaced by a 0 to 100 AI Success Score; white-label client report; API tokens per project; reports made on request. ↩ ↩²

60. Rankscale. Facts. GUI engines and API model engines listed separately, as of January 2026, with “Grok Chat GUI” among the GUI engines. ↩

61. Scrunch. What is the pricing for Scrunch plans? Core $250 per month; Agency Core $500 per month. ↩ ↩²

62. Scrunch. What methods does Scrunch use to collect data from AI platforms? Browser automation and official platform APIs; daily for 14 days, then every 72 hours. ↩

63. AthenaHQ Docs. Setup. Weekly or monthly cadence; no collection method stated. ↩

64. BrightEdge. BrightEdge AI Catalyst completes the search picture (April 9, 2025). “available now for all customers.” ↩

65. HubSpot. HubSpot to acquire XFunnel (page updated October 31, 2025). ↩

66. Peec AI. We raised a $21M Series A (November 18, 2025). ↩

67. Semrush. Adobe completes Semrush acquisition (April 28, 2026). ↩

68. Uberti-Bona Marin et al. “If I Had to Buy Just ONE: Galaxy S26 Ultra”: Auditing AI-Generated Product Recommendations (arXiv preprint, September 16, 2026). Mean domain overlap 12.0% for ChatGPT and 14.8% for Gemini. ↩

69. Surfer. Scraped AI answers vs API results (last updated October 4, 2026). 1,000 prompts, 13,779 answers; brand overlap 15.5% to 23.8%; ChatGPT domain overlap 4.8%. Vendor study. ↩

70. Brandlight. UI vs API (June 2026). 2,565 matched answer pairs; brand overlap 6% to 12%. Vendor study. ↩

71. Genezio. Zero query overlaps (February 4, 2026). GPT-5.2 API vs ChatGPT.com, zero shared queries, 704 of 5,701 sources (12.3%) used by both. Vendor study. ↩

72. SparkToro. Office Hours: Can you actually track AI visibility? (May 8, 2026; webinar recorded March 11, 2026). Rand Fishkin describing Gumshoe’s published method (10 prompts x 7 phrasings x 7 models, about 490 answers a pass): one pass about plus or minus 5 points; about 20 passes for plus or minus 1. ↩ ↩²

73. Żatuchin. The Dice Roll Method (arXiv preprint, submitted September 3, 2026; v2 October 14, 2026). G = 0.96 at five runs across 250 questions; G = 0.27 for five versions of one question. ↩

74. Conductor. AI recommendation consistency analysis (updated October 7, 2026). 14,000 API calls; purchase prompts shared 4 of 10 brands across two runs; comparison prompts kept the same leader 91% of the time. Vendor study. ↩

75. AirOps. The State of AEO report. “Only 30% of brands stay visible run to run.” Vendor report. ↩

76. Ahrefs. How often AI Overviews change (November 11, 2025). 70% chance of change between observations; 2.15 days average persistence; 54.5% URL overlap. ↩

77. Zapier. Best AI visibility tools. Lists OtterlyAI first, labeled for affordability. Read October 18, 2026. ↩

78. Ahrefs. Free AI Visibility Checker. No signup; top 5 results per section; ChatGPT, Gemini, Perplexity, Copilot, AI Overviews, and AI Mode. ↩

79. Bing Search Blog. New AI visibility insights in Bing Webmaster Tools: Intents, Topics, Citation Share, Compare (June 16, 2026). ↩

80. Bing Webmaster Blog. Introducing Topics Portfolio (beta) in AI Performance (October 13, 2026). ↩

81. Google Analytics Help. Traffic-source dimensions, manual tagging, and auto-tagging. Definition of the AI Assistant channel. ↩

82. SE Ranking. AI Visibility Tracker. “You get 5 free checks per day.” ↩

83. Scrunch. What features are included in Scrunch’s free trial? “Scrunch requires a credit card to start the 7-day free trial”; users “will automatically be upgraded to a paid subscription after the end of the trial unless they cancel.” ↩

84. Search Engine Journal. Are LLM Visibility Trackers Worth It? (Harry Clarkson-Bennett, October 15, 2025). DIY tracker estimate of about $96 a month. ↩

85. Backlinko. Prompt Tracking: How to Find (and Fix) Your AI Visibility Gaps (updated August 14, 2026). ↩

86. Seer Interactive. AIO impact on Google CTR, 2026 update (April 24, 2026). +120% organic clicks per impression on informational queries when cited in the AI Overview; 53 brands tracked January 2025 to February 2026. ↩

87. SparkToro, presenting Similarweb research. How AI brand mentions influence direct visits and traditional search queries (June 29, 2026). Capital One, 14.2% more likely to visit. ↩