Ahrefs Brand Radar tracks AI mentions inside your SEO suite, but it skips Claude, runs $828-1,148/month, and isn't a full swap for a dedicated GEO tool.
Ahrefs added Brand Radar to its SEO suite so existing customers could see how often ChatGPT, Perplexity, and Google's AI surfaces mention their brand without buying a second tool. That pitch is the whole appeal, and it's a real one. It's also not the same product as a dedicated GEO tool built to do only that job, and the gap shows up in three places: which platforms it tracks, how accurate the numbers are, and what it actually costs once you add up the pieces Ahrefs sells separately.
This post breaks down what Brand Radar covers, prices it out fully, and lays out exactly where an independent tool like Profound, Otterly, or Scrunch AI still wins.
Brand Radar is an add-on module inside Ahrefs, not a standalone product. You need an active paid Ahrefs plan to turn it on at all, and the module itself is priced separately on top of that.
Brand Radar tracks seven AI surfaces as of this writing, per Ahrefs' own product page: Google AI Overviews, Google AI Mode, ChatGPT, Perplexity, Gemini, Microsoft Copilot, and Grok. Ahrefs also runs a beta module tracking YouTube, TikTok, and Reddit mentions, bundled free into both tiers while it's in beta; Ahrefs' own page doesn't say what it will cost once the beta ends.
Prompts feed from two sources, not from a live feed of what people actually type into ChatGPT. Ahrefs' own methodology page describes drawing queries from Google's People Also Ask corpus and Ahrefs' keyword database, then expanding those into related sub-questions before running them "through the free, publicly available web interfaces of ChatGPT, Gemini, Perplexity, Copilot, and other supported platforms to reflect typical user experiences." That's a reasonable proxy for search-style questions. It's a weaker proxy for the more conversational, multi-turn prompts people actually run inside a chat app, which is where a purpose-built GEO tool's custom-prompt library tends to do more work.
Here's the number that matters most for anyone pricing this out: Brand Radar's AI Indexes cost $398/month to select individual platforms, or $699/month bundled for full coverage of all platforms, per Ahrefs' own Brand Radar pricing. That's on top of a base Ahrefs subscription, which starts at $129/month for the Lite tier and runs up to $449/month for Advanced.
Do the math and the floor is $828/month (Lite plus the full bundle), climbing toward $1,148/month if you're on a higher Ahrefs tier. Analyze AI's review puts the realistic all-in range at exactly that, $828 to $1,148 a month, and calls it roughly 2.5 times the average price of a dedicated AI visibility tool. The all-platform bundle includes 2,500 monthly checks for custom prompts, where one check equals one prompt run against one LLM in one location, per Ahrefs' own product page. Run more prompts, more locations, or more frequent checks, and that allotment goes fast.
For context on how that stacks against Ahrefs' other paid modules, see our breakdown of Ahrefs' API pricing against DataForSEO and Semrush, and for the wider GEO spending landscape, what SaaS teams actually pay for GEO in 2026.
A tool that costs more than any dedicated competitor's entry tier needs to earn that premium somewhere. Brand Radar's gaps are concentrated in exactly the three places a buyer would check first: which model it skips, how current its data is, and whether a "mention" in its dashboard means what you'd assume it means.
Claude is not one of the seven platforms Brand Radar tracks. It's absent from Ahrefs' own product page, and Profound's review of Brand Radar calls the gap out directly, noting Brand Radar is "missing some crucial LLMs like Claude, Grok, Meta, and DeepSeek." For a team specifically trying to answer "does Claude cite us," Brand Radar can't answer that question at any price, bundled or not. If Claude coverage is the reason you're evaluating a tracker, get cited by ChatGPT, Perplexity, and Claude covers what actually earns a citation from Anthropic's model specifically.
The same review flags a second gap that compounds the first: refresh rate. Non-Google platforms in Brand Radar update once a month, while Google AI Overviews data refreshes more often. A monthly snapshot is fine for a quarterly board deck. It's not fine for catching a competitor that just started outranking you in ChatGPT answers three weeks ago.
This is the part worth reading carefully before you trust Brand Radar's numbers for anything beyond a rough directional signal. Writesonic ran its own brand through Brand Radar and compared the result against its own GEO tracker. Brand Radar reported 3 global ChatGPT mentions for Writesonic, zero of them from the US, against 123 actual mentions Writesonic's own tool found, a gap of roughly 40x. Perplexity told a similar story: 6 global mentions in Brand Radar versus 212 actual.
That's not rounding error. It's the difference between "we're barely showing up" and "we're showing up constantly," and a team making a content or PR decision off the smaller number would draw exactly the wrong conclusion. The likely cause is the sampling design itself: a fixed, PAA-derived prompt set run monthly on non-Google platforms will always undercount relative to a tool that samples a broader, more current set of real user-style prompts, more often.
Give Ahrefs credit for transparency here. Its methodology page states plainly: "We do not filter out hallucinated or malformed links, as they reflect real model output." That's a defensible design choice, raw output is honest output, but it means a spike in "citations" could be the model inventing a URL that doesn't resolve, not a real referral source.
It ties into a broader distinction independent reviewer Connor Kimball raised in his Brand Radar review: "the tool often struggles to differentiate between a simple brand mention and a verifiable, clickable citation. This is a critical distinction; a mention is awareness, but a citation is authority." A dashboard number that conflates the two forces you to manually verify what actually counts, work a dedicated tool is often built to do automatically. If a model fabricating facts about your company is a live concern rather than a hypothetical, ChatGPT hallucinating about your company covers how to diagnose and fix it once you've found a case.
None of this makes Brand Radar a bad product. It makes it a specific product, built for a specific buyer, and the honest answer to "should I use it" depends entirely on which buyer you are.
If your team already pays for Ahrefs and just wants a rough sense of whether ChatGPT and AI Overviews mention you at all, Brand Radar is a legitimate "good enough for now" option. You don't add a vendor, a login, or a separate invoice, and the monthly snapshot is enough to spot a large swing, a competitor suddenly dominating a category, your own mentions falling to zero, even if it can't catch a subtle weekly trend. For that narrow use case, the $398-699/month add-on is a reasonable line item on a tool you're already paying for.
Once AI visibility becomes a real priority rather than a curiosity, the calculus flips. Otterly's entry tier is $29/month ($25 billed annually) and tracks four engines, ChatGPT, Perplexity, Google AI Overviews, and Microsoft Copilot, with a free trial to test it first. Profound's self-serve floor is $99/month for ChatGPT-only tracking, and its $399/month Growth tier adds Perplexity and Google AI Overviews. Worth flagging honestly: Claude coverage on Profound is not self-serve either, it sits behind Profound's custom-priced Enterprise tier, so a team that needs Claude specifically should confirm that before buying any tool. Both Otterly and Profound's entry prices still sit well under Brand Radar's $828+/month realistic floor, and neither requires you to already own an Ahrefs subscription to use them. We compare all three dedicated options, including Scrunch AI's alerting and Influence Score, in Profound vs Otterly vs Scrunch AI, the fuller buying guide this post hands off to once Brand Radar's gaps start to matter for your team.
The pattern across every review cited here is consistent: Brand Radar trades coverage and accuracy for bundling convenience, and it charges a premium for that convenience rather than a discount. A dedicated tool costs less at the entry tier and does more of the actual job, tracking Claude, sampling live rather than monthly, and distinguishing a mention from a citation. Brand Radar's edge is only that it's already inside a tool you might be paying for anyway.
Whichever tracker you land on, Brand Radar or a dedicated GEO tool, it answers the same narrow question: are you showing up. It won't write the page that gets you cited, and neither tool claims to. That's a content problem, and it's the other half of the equation these AI visibility tools were never built to solve.
Lyra researches the topics your buyers actually ask AI assistants about, writes each post in your blog's existing voice, verifies every fact and price against a live source the way this comparison was built, and structures pages the way answer engine optimization rewards: a direct answer early, headings shaped like real questions, and citations that hold up when a model checks them. She opens the finished post as a pull request. Nothing publishes without you merging it.
If you're already tracking AI mentions and the number isn't moving, see the plans or try Lyra directly, the content is very likely the half of the problem your tracker was never going to fix.
Brand Radar or a dedicated GEO tool can tell you whether AI assistants mention your brand. Neither one writes the page that earns the citation. Lyra does, verified and opened as a PR you review.
FAQ
Seven: Google AI Overviews, Google AI Mode, ChatGPT, Perplexity, Microsoft Copilot, Gemini, and Grok. YouTube, TikTok, and Reddit tracking is available as a beta module. Claude is not tracked, a gap flagged by multiple independent reviews and absent from Ahrefs' own product page.
Brand Radar's AI Indexes are $398/month to select individual platforms or $699/month bundled for full coverage of all platforms, and it requires an active paid Ahrefs plan starting at $129/month. That puts the realistic all-in floor at $828/month (Lite plan plus the full bundle) and up to roughly $1,148/month on Ahrefs' Advanced tier, before any custom-prompt or social-tracking add-ons.
Its own methodology says no filtering happens: hallucinated and malformed links stay in the data by design. Independent testing has also found real undercounting. Writesonic's own check found Brand Radar reporting 3 global ChatGPT mentions for its brand against 123 actual mentions from its own tracker, and 6 Perplexity mentions against 212 actual, a gap of roughly 40x on ChatGPT.
Not on coverage, accuracy, or price. Brand Radar misses Claude entirely, refreshes non-Google platforms only monthly, and costs more than any dedicated tool's entry tier. Its advantage is bundling: if you already pay for Ahrefs, Brand Radar is a directional read you don't need a second login for. If AI visibility is a real priority rather than a nice-to-have, dedicated tools close the Claude gap in different ways: Scrunch AI tracks Claude on every tier including its entry plan, Otterly offers it as a paid add-on, and Profound reserves it for its custom-priced Enterprise tier. All three offer live sampling and a lower entry price than Brand Radar's realistic floor.
Built by the tool you're reading about
Lyra finds the topics worth ranking for, writes them in your repo's voice, fact-checks every claim, and opens a pull request scored and ready to merge. You review and hit merge. Want to see what she'd write for you? Start free with three posts, no card.
Keep reading
Jasper AI pricing 2026: Pro caps at one seat, Brand Voices stay capped, and Business pricing is custom. Here is what it costs a real content team.
Sanity vs git-based blog: Sanity versions its schema in Git, but published posts skip PR review entirely. What that gap means for a SaaS blog's editorial process.
Claude vs GPT-5 vs Gemini for blog writing, compared on quality benchmarks, real BYOK token pricing, and a practical model-routing strategy for teams.