What AI Visibility Tracking Costs in 2026, per Answer

Self-serve AI visibility tracking costs from $29 to $599 a month on the published plans we checked, and most vendors also sell an enterprise tier by quote only. Per answer collected, the spread runs from about $2 per thousand on SE Visible's Agency plan to about $82 per thousand on AthenaHQ's Starter plan.

Neither number tells you what a usable measurement costs. AI answers vary so much from run to run that a small prompt set produces a margin of error wider than most of the changes you'd want to see, and once you size the prompt set properly, the cheapest workable plan moves up a tier or two.

*cite-met sells a managed AI visibility service, and our own plan is in the comparison below, measured the same way as everyone else's.*

Entry plans run from $29 to $300 a month, and per 1,000 answers most land between $8 and $25.

| Tool and plan | Price per month | Prompts | Engines | Answers per month | Per 1,000 answers | Collection | |---|---|---|---|---|---|---| | [SE Visible](https://visible.seranking.com/pricing/) Basic | $89 | 75 | 5 | about 11,250 | $7.90 | UI | | SE Visible Agency | $399 | 1,000 | 5 | about 180,000 | $2.20 | UI | | [Peec AI](https://peec.ai/pricing) Starter | $95 | 50 | 3 | 4,500 | $21.10 | UI, logged out | | Peec AI Advanced | $495 | 350 | 3 | 31,500 | $15.70 | UI, logged out | | [Siftly](https://siftly.ai/ai-instructions) Starter | $79 | 50 | 3 | 4,500 | $17.60 | not stated | | [Otterly.AI](https://otterly.ai/pricing) Lite | $29 | 15 | 4, without Claude | 1,800 (derived) | $16.10 | UI | | Otterly.AI Premium | $489 | 400 | 4, without Claude | 48,000 (derived) | $10.20 | UI | | [Scrunch](https://scrunch.com/pricing) Starter | $300 | 350 | 7 | about 24,500 (derived) | $12.20 | UI and API | | [Ahrefs Brand Radar](https://ahrefs.com/pricing) custom prompts, Basic | $50 | counted in checks | up to 7 | 2,500 checks | $20.00 | not stated | | [Frase](https://www.frase.io/pricing) Starter | $49 ($39 billed yearly) | 20 | 2 | 1,200 (derived) | $40.80 | not stated | | [AthenaHQ](https://www.athenahq.ai/pricing) Starter | $295 | credit-based | 11 | 3,600 | $81.90 | not stated | | [Semrush AI Visibility Toolkit](https://www.semrush.com/pricing/ai/) | $99 per domain, billed yearly | 25 | 3 to 4 | not published | n/a | not stated | | [Sight](https://docs.trysight.ai/billing/plans-pricing) Starter | $99 | 100 | 5 | not published | n/a | not stated | | [Profound](https://www.tryprofound.com/pricing) | 7-day trial, then Enterprise by quote | 50 on trial | up to 9 | not published | n/a | not stated | | [cite-met](/pricing) Growth | $700, including hosting, posts and Pulse | set per client; 71 on our own site | 2 (ChatGPT, Claude) | about 610 for 71 prompts run weekly (derived) | about $1,150, dividing the whole $700 by answers | API |

How to read it. Prices are monthly billing from each vendor's pricing page unless marked. "Answers per month" is the vendor's own figure where it publishes one; otherwise we multiplied prompts by engines by runs a month and marked it derived. "Collection" is where the vendor says answers come from: the consumer app (UI), the model's API, or not stated.

Row notes:

- SE Visible's 180,000 is its own figure. Our arithmetic (1,000 prompts, 5 engines, daily) gives 150,000, which would make it $2.70 per 1,000. - Otterly showed two price sets, $29/$189/$489 and $25/$160/$422, so confirm at checkout. Claude is an extra add-on, covered in the next section. - Scrunch runs a new prompt daily for its first 14 days and every 72 hours after that. The 24,500 uses the 72-hour rate, about 10 runs a month, and assumes every prompt runs on all 7 engines. - Ahrefs counts checks, one prompt on one platform in one location, and a Claude update uses 8 checks, so Claude answers there cost about $160 per 1,000. Ahrefs says Brand Radar works with a free account, so we haven't added a paid plan. - Siftly's figures come from its plan summary page, dated 19 August 2026. Its Scale plan is $599 a month for 150 prompts.

Read the answer counts as a best case. Answers from a logged-out app session and answers from an API are different goods. And a scheduled run for Google AI Overviews on a query that triggers no overview returns nothing, which vendors may or may not count as an answer.

The cite-met row looks expensive because it is, if answer tracking is all you want. Most of the $700 Growth plan pays for hosting, four posts a month, visitor and lead tracking and up to 10 sites. Its answer tracking covers two engines once a week, with no Google AI Overviews or AI Mode, and on our own site's 71 prompts it falls short of the credible setup described below.

Vendors sell different units under the same word. A Peec prompt runs on 3 engines every day and a Scrunch prompt on 7 engines every few days, while an AthenaHQ credit buys one AI response and an Ahrefs check is one prompt on one platform in one location.

Engines are the second difference, and Claude is the usual exclusion:

- Otterly adds Claude, through Claude's API, as a separate monthly add-on: $29 on Lite, $109 on Standard and $439 on Premium. On Premium that nearly doubles the $489 price. - Peec lists Claude only on Enterprise, and Siftly only on request. - Frase includes Claude from its Scale plan, at $299 a month ($239 billed yearly). - SE Visible lists Claude as coming soon on its pricing table, though its FAQ mentions Claude tracking on the Agency plan. - Scrunch, AthenaHQ and Sight include it on their entry paid plans, and Sight's Starter is $99.

The third is how the licence is counted. [Semrush](https://www.semrush.com/kb/1493-ai-visibility-toolkit) charges $99 per domain, billed yearly, so an agency tracking five client domains starts at $495 a month, and extra users cost from $45 a month each.

It's gone. Profound's Starter plan at $99 a month and Growth at $399 were still listed in [Semrush's July 2026 roundup](https://www.semrush.com/blog/best-ai-visibility-tools/), and [Eclincher's roundup of Profound alternatives](https://www.eclincher.com/articles/profound-alternatives-8-ai-visibility-tools-after-profounds-pricing-change) dates their removal to mid-September 2026. The pricing page now lists a 7-day trial with 50 prompts and a custom Enterprise plan. Its FAQ mentions an Agency Growth plan with no price, and Eclincher reports that agencies pay $399 per client workspace on it.

Several pages that rank for "Profound pricing" still quote $99, so check the date on any comparison you read. Our [Profound alternative](/profound-alternative) page covers what changed for agencies.

About 800 answers a period to trust a 5-point change, and at 300 answers a period only moves of around 8 points stand out from noise. That's because AI recommendations change from run to run. In a [SparkToro and Gumshoe study](https://sparktoro.com/blog/new-research-ais-are-highly-inconsistent-when-recommending-brands-or-products-marketers-should-take-care-when-tracking-ai-visibility/) of 2,961 responses from ChatGPT, Claude and Google's AI, two runs of the same prompt returned the same list of brands less than once in a hundred tries.

The arithmetic: for a single figure near 50%, the 95% margin of error is roughly one divided by the square root of the number of answers, so about plus or minus 10 points at 100 answers and 6 at 300. A change between two periods compares two noisy figures, so its margin is wider again, about 8 points at 300. And a zero isn't proof of absence: named in none of 100 answers, your true rate could still be around 3%.

Those margins assume every answer is independent, and repeats aren't. Ten runs of the same prompt share the wording, the model version and usually the location, so they behave more like a few answers than ten, and real precision is worse than the formula says. The fix is more distinct phrasings rather than more repeats. A [July 2026 preprint](https://arxiv.org/abs/2607.13304), which measured sentiment rather than mentions, projected that repeating a prompt past about five times adds almost nothing, and a [University of St. Gallen study](https://arxiv.org/abs/2604.07585) recommends pooling runs over two to four weeks. We work through the numbers in [how many prompts and runs you need](/blog/post/how-many-prompts-ai-visibility).

Between about $200 and $500 a month for a credible setup, and under $100 for a rough one.

Take three topics on three engines, reported monthly. A minimum setup is 30 prompts, 10 phrasings a topic. Even if the tool runs them daily, most of those answers repeat the same ten wordings, so the true margin is far wider than the 6 points the formula gives. A credible setup has about 100 distinct phrasings a topic, kept fixed from month to month so the trend means something: 300 prompts. Each price below is the cheapest plan that holds that many prompts on three of its included engines, at the vendor's own run frequency.

| Tool | Minimum: 30 prompts | Credible: 300 prompts | Notes | |---|---|---|---| | Sight | $99 (Starter, 100 prompts) | $199 (Pro, unlimited prompts) | Run frequency not published | | SE Visible | $89 (Basic) | $399 (Agency; Core stops at 200) | Daily runs, 5 engines | | Scrunch | $300 (Starter) | $300 (Starter, 350 prompts) | Every 72 hours after 14 days | | Ahrefs Brand Radar | about $54 (Basic, 2,500 checks plus about 200 more at $0.02) | about $290 (Scale, 25,000 checks plus about 2,000 more at $0.02) | Assumes daily checks, no Claude | | Otterly.AI | $189 (Standard) | $387 (Standard plus 200 extra prompts) | Claude add-on extra | | Semrush | $159 ($99 base with 25 prompts, plus $60 for 50 more), billed yearly | about $459 ($99 plus six packs of 50), billed yearly | Per domain, per user | | Peec AI | $95 (Starter) | $495 (Advanced) | Claude on Enterprise only | | AthenaHQ | $295 (Starter) | Enterprise; Starter's 3,600 credits cover about 4 runs of 900 answers | Credit-based | | Siftly | $79 (Starter) | Enterprise (Scale stops at 150 prompts) | |

The cheapest credible options are Sight and Ahrefs, if their run frequency suits you, and most of the rest land between $300 and $500. AthenaHQ and Siftly need an enterprise quote at this size. Frase and Profound are left out because their published plans don't hold 300 prompts at a listed price. The sub-$100 plans are still worth buying if you read their numbers as rough direction.

It leaves out the crawler fetches, the human visits and the leads. Some trackers now bundle crawler data: Peec includes bot-visit quotas on every plan, Otterly has Agent Analytics from Standard up, and Profound's Agent Analytics reads forwarded server logs, with integrations for hosts and CDNs such as Cloudflare and Vercel. Separate tools cover the rest, and some cost nothing:

- [Cloudflare AI Crawl Control](https://developers.cloudflare.com/ai-crawl-control/) shows AI crawler requests on every Cloudflare plan for traffic that runs through Cloudflare. cite-met's [AI crawler analytics](/features/ai-crawler-analytics) does the same for the sites we host. - If you'd rather not pay yet, Three Impact has a [free method for tracking AI visibility by hand](https://threeimpact.com/ai-visibility-tracking-tools), about two and a half hours a month for a small prompt set. It's published by the team behind cite-met. - [Known Agents](https://knownagents.com/pricing), where darkvisitors.com now redirects, has a free tier for crawler and agent analytics and paid plans from $29 a month. - Google Analytics 4 has an [AI Assistant default channel](https://support.google.com/analytics/answer/9756891?hl=en) for visits from sources such as ChatGPT, Gemini, DeepSeek, Copilot and Grok. It excludes Google's AI Overviews and AI Mode, which stay in Organic Search, and like every referrer-based tool it files visits that arrive without a referrer, such as clicks from some apps, as Direct.

If you need to show a return, you'll have to join the answer, the crawler fetch, the visit and the lead yourself or pay someone to. The work of acting on the data is the other missing cost. The consultant Lilach Bullock, who sells that work as we do, [puts a small business's tool](https://www.lilachbullock.com/what-ai-visibility-tracking-actually-costs/) at $50 to $200 a month and an agency's spend at $500 to $2,000 or more, and the realistic all-in cost for a small business that acts on its data at $300 to $400 a month.

Ask any vendor how many answers a month the plan collects, which engines are included at that price, and whether answers come from the consumer app or the API. On collection method, only SE Visible answers on its pricing page; Peec, Otterly and Scrunch explain it in their docs, and the other vendors in the table didn't say anywhere we looked. Then check how the number is built:

- Who wrote the prompts. Suggested prompts tend to include your brand name, which inflates visibility, so keep branded and unbranded prompts apart. - Whether the headline metric counts mentions or citations. Being named isn't the same as having your page linked. - Per-engine rates rather than a blended score, so you can weight engines by where your buyers are. - Whether model updates are marked on the chart, since a new model can move every line overnight. - Whether you can export the raw answers. Without them you can't compute your own margins or spot-check the scraper against ten prompts run by hand. - How much weight it puts on rank or "average position". These are the weakest numbers on the dashboards, and SparkToro's study recommends visibility percentage over many runs instead.

If you want the tracking, crawler, visit and lead data handled together, with someone acting on it, answer tracking starts on our [Growth plan](/pricing) at $700 a month. If you only want prompt numbers and have the time to work from them, one of the tools above will cost you less.

*Prices checked on the vendors' own pages on 8 November 2026; Siftly from its 19 August 2026 plan summary. Prices change often in this category, so check before you buy.*