Profound alternatives, priced, now that Profound is not

So this page does the other half of that job. Every alternative below carries a number, a billing period, a currency and the date it was read, and where a vendor meters by engine the headline is worked up to what full coverage costs.

What Profound does well

Profound's product is not in doubt, and its own pages describe it in detail even where its pricing page has gone quiet. Its Answer Engine Insights feature tracks brand mentions across ChatGPT, Perplexity, Claude, Microsoft Copilot, Google AI Overviews, Google AI Mode, Google Gemini, Grok and DeepSeek, described on its features page as monitoring "all major consumer Answer Engine platforms", read from tryprofound.com/features on 16 September 2026. The same product line states it can "analyze how AI describes your brand" and "identify key themes, topics, and recurring narratives", and that it lets a customer "compare your presence against industry leaders and competitors", both quoted from tryprofound.com/features/answer-engine-insights, also read 16 September 2026.

Profound sells prompt volumes built from double opt-in consumer panels, described as covering "hundreds of millions of real user queries per month" across ChatGPT, Perplexity and Gemini, quoted from tryprofound.com/features and tryprofound.com/features/prompt-volumes, both read 16 September 2026. It reports which AI crawlers visit a customer's own site and how often, and its monitoring extends to "30+ languages and 150+ regions", both quoted from the features pages above. For a buyer whose procurement process requires SOC 2 Type II compliance and single sign-on before anything gets approved, Profound states on the same page that it has both.

What you can get without talking to sales

None of that sits behind a number any more. What a brand reaches on its own is the free trial, described on the page in six words: "10 prompts run once", "ChatGPT only". Ten prompts asked once is a demonstration rather than a measurement, and one engine cannot say whether Claude or Perplexity recommend a business. Everything past that is a sales conversation on terms neither of us can see.

Selling enterprise software that way is ordinary and plenty of good companies do it. It is still the thing a buyer on a budget needs to know before booking a demo, and it is what any comparison written before 20 September 2026 will get wrong.

Two vendors here still meter by engine, which is the pattern Profound's old Starter belonged to: Peec AI and Otterly.AI, both priced in full below. Metering is an ordinary way to sell and each of them discloses it on its own site. The problem is a comparison that quotes the headline and stops, making a metered plan look interchangeable with one that includes every engine. A comparison nobody re-reads fails one step further along, quoting a number the vendor has taken down.

The alternatives, priced

The table below prices every credible alternative to Profound at its entry tier, with the billing period and what our source says about engine coverage attached to each figure. Every figure below is in USD, and every one comes from Keylight's own price research, which reads each vendor's live pricing page in a real browser and records the date, rather than taking a figure from a blog post or a review.

Peec AI

Peec AI's Starter plan is $95 a month or $80 billed annually, both in USD, for a workspace that can pick 3 of the 6 AI models Peec offers below its Enterprise tier. A fourth model costs $35 more a month on Starter, rising to $85 on Pro and $165 on Advanced, all read from Keylight's price research on 3 and 4 September 2026. Peec's headline $80 to $95 buys 3 engines. Adding the three missing models at $35 each brings Starter to $200 a month billed monthly, or $170 billed annually at the $30 annual add-on rate. The whole of that metering worked through, tier by tier, is in Peec AI alternatives.

Otterly.AI

Otterly.AI's Lite plan is $29 a month or $25 billed annually, both in USD. Claude, Gemini and Google AI Mode are sold as paid add-ons on every tier Otterly offers, including Lite, read from Keylight's price research on 3 September 2026. The entry price is a cheap way into a recognised name in this category, provided the three gated engines are not the point of buying it.

LLMrefs

LLMrefs sells one plan, All in One, at $79 a month, with no annual billing option, in USD, read 3 September 2026. Our source records all AI search engines included at no extra fee, and notes the pricing page carries a "Limited time only" badge on that price, worth watching before treating $79 as a fixed number.

Scrunch and AthenaHQ

Scrunch's Core plan is $250 a month in USD. Its annual price could not be established: the pricing page carries a repeated banner about a coming change and the annual figure would not render, both noted in Keylight's price research on 3 September 2026. Core covers 4 engines, ChatGPT, Perplexity, Google AI Overviews and Microsoft Copilot, per the same research.

AthenaHQ's Starter plan is $295 a month or $245 billed annually, both in USD, read 3 September 2026. Starter covers 11 models, read 20 September 2026 and up from 10 seventeen days earlier, and carries 3,600 credits a month, where one credit is one AI response, so it buys more engines at its entry tier than Keylight sells at any tier. A separate Essential tier is free, with 300 credits included, according to the same source. AthenaHQ also withholds its add-on prices for API access and extra credits, answering only "contact us", per the same research.

Ahrefs Brand Radar and Semrush

Ahrefs sells two different things, and a comparison has to say which. Brand Radar is a feature bundled inside the main Ahrefs plans, starting at $129 a month on Lite, not available as a self-serve annual price. Brand Radar AI is a separate, standalone product, priced from $199 a month, and also sold as prompt packages: Basic at $50 a month for 2,500 checks, Growth at $100 a month for 7,000 and Scale at $250 a month for 25,000, with overage charges beyond that, all read 3 September 2026. A single Ahrefs figure for AI visibility, quoted without saying which of the two it is, is half a product.

Semrush runs a similar split. Its plain SEO Starter plan is $139 a month and does not include AI visibility tracking at all. The tier that does, SEO plus AI Search Starter, is $199 a month or $165.17 billed annually, and covers 50 prompts tracked daily on 1 domain, checked 4 September 2026. $199 is the figure that belongs in any AI visibility comparison. $139 does not.

Keylight

Keylight is a newer product than Profound, and its feature surface is smaller. It does not sell AI keyword research from real user query panels, content workflow automation, crawler analytics or SOC 2 Type II compliance, all of which Profound offers today. A buyer who needs any of those should not expect to find them on trykeylight.com. Keylight priced against the rest of the field, sorted by who is asking rather than ranked, is in the best AI visibility tools.

What Keylight does instead is include all 8 engines on every paid plan, with none sold as an add-on, starting with Solo at $79 a month, or an effective $65.83 a month billed annually under its two months free policy, both in USD, priced as of 3 September 2026. Solo covers 500 tracked questions a week, spread across whichever of ChatGPT, Gemini, Claude, Perplexity, Google AI Overviews, Google AI Mode, Grok and DeepSeek the customer chooses, with no engine sold as a separate add-on. One question sent to one engine once is one of the 500. Grok and Claude are each capped at 10% of an account's weekly capacity, because both are priced by usage rather than by request and cost more to run than the other six engines. A customer who wants more of either pays $39 a month to lift that engine's cap by 100 a week, inside the same 500. Every score carries a margin of error, a week-on-week move is only called a change when its whole range sits clear of zero, and the exact text the engine produced is kept behind every score. Why a score needs a margin of error at all, and what three published studies measured about how much AI answers move between identical runs, is in why AI answers change between runs. The free report card at app.trykeylight.com/report-card is the shortest way to see the first of those three on a real result: ten questions on one engine, the score with its margin beside it, and no account or email needed to read what comes back.