Profound Review: What the AI Visibility Platform Actually Tracks

This Profound review covers what the AI visibility platform actually tracks, what its three pricing tiers include as of August 2026, and where it sits next to the other tools that show up in the same comparisons. Profound (tryprofound.com) is a paid platform that runs a fixed set of prompts against AI assistants on a schedule and reports back which brands get named, in what position, and how often, the same basic job every tool in this category does, with its own choices about which engines it covers and at what price.

What Profound actually is

Profound sells recurring, scheduled tracking: you give it a set of prompts (or it generates some for you) in your category, it runs them against one or more AI assistants on a repeating basis, and it reports which brands got mentioned, where in the answer, and how that shifts over time. That is the same category of product as SE Ranking’s AI Search add-on, Peec AI, and the other generative engine optimization tools it gets compared against, distinct from a one-off checker you run by hand. The thing worth checking before buying into any tool in this category, Profound included, is exactly which assistants it checks and how that maps to your plan, because that number moves a lot between tiers.

Why engine count is the first number to check

“AI visibility” is not one product surface, it is at least five separate places a buyer can ask a question and get a different answer built from a different mix of sources: ChatGPT, Claude, Perplexity, Grok, and Google’s AI Mode. A tool that checks one of those and a tool that checks all five are not competing on depth of insight so much as on which slice of the buyer journey they can even see. Before comparing dashboards, comparing headline scores, or comparing anything else two tools report, the first question worth answering is which assistants each one actually queries, and how many of them your plan pays for. That is the frame this review uses on Profound specifically.

What each Profound plan actually covers

Pricing captured from tryprofound.com/pricing on August 25, 2026:

• Starter, $99/mo. ChatGPT tracking only, 50 prompts tracked. Built for small companies checking a single assistant.
• Growth, $399/mo. 3 answer engines tracked, 100 prompts. Aimed at growing companies actively building AEO content.
• Enterprise, custom pricing. Up to 9 answer engines tracked, dedicated Slack support, for large companies and agencies.

The gap between tiers is the part worth sitting with. On Starter, at $99 a month, you are only seeing what ChatGPT does with your brand. Whatever Claude, Perplexity, or Google’s AI Mode are saying about you in the same window is invisible until you move up to Growth at four times the price, and even Growth caps out at 3 engines. Nine engines exist on the Enterprise tier, but Profound does not publish what that costs. If the buyer question you care about gets asked across several assistants, not just one, the plan you can actually afford is the one that determines how much of the picture you get to see.

How Profound shows up next to the tools it competes with

A separate, more direct data point: a measured set of citation counts against monthly traffic across sites in the “AI search visibility tracking” category put Profound in the middle of the pack, an estimated 41,346 monthly visits to tryprofound.com against 5 recorded mentions in that comparison set. For scale, in the same measurement, Semrush (3,209,872 monthly visits) picked up 12 mentions, Ahrefs (1,513,824 visits) picked up 8, SE Ranking (320,422 visits) picked up 4, and Peec AI, with only 1,132 monthly visits, picked up 7, more mentions than a site with 283 times its traffic.

That last comparison is the useful takeaway, and it is not really about Profound at all: how often a tool gets named in roundups like this one tracks its content and outreach machine more than it tracks how much anyone actually uses it. Profound sitting in the middle of that table says its name comes up with reasonable frequency in coverage. It says nothing about whether its citation counts match what a brand would find checking the same prompts by hand, which is a separate question this article cannot answer for you, the same caution that applies to any single published AI visibility score: it measures what its formula measures, not a universal truth about your brand.

What Profound does not publish

Beyond the plan page, there is not much public detail on Profound’s exact methodology: how a prompt gets selected as representative, how a mention gets weighted by position, or how often the underlying models get re-run within a billing period. That is not unusual for this category; most vendors treat the formula as the product. It does mean the honest way to evaluate Profound, or any competitor, is the same way you would evaluate a AI visibility claim from anyone: ask what assistants were actually queried, how many prompts, and over what window, before treating the resulting number as settled.

None of that is a knock specific to Profound. It is the current state of a young category where the exact formula behind a dashboard number is usually treated as a competitive advantage rather than something published in a help document. The practical fix, whichever tool you end up evaluating, is to ask the vendor directly which prompts ran, on which models, and how recently, and to treat a refusal to answer that plainly as information in itself.

Who Profound actually fits

Profound’s tier structure fits a specific buyer well: a team that already knows ChatGPT is the assistant that matters most for its category, is fine paying $99/mo to confirm that on an ongoing basis, and can justify the jump to $399/mo once it needs a second or third engine in view. It fits less well for anyone who wants multi-engine coverage from day one without climbing a pricing ladder to get it, or an agency managing several client brands where Enterprise-only, custom-quoted access to the full 9-engine set is the only way to see everything at once.

FoundCite’s method queries 5 engines by default on every plan, ChatGPT, Claude, Perplexity, Grok, and Google AI Mode, without a separate tier gate on which assistants you get to see. The pricing page lays out what each plan tracks and publishes; there is no per-engine add-on to negotiate.

Is Profound worth trying

If a single-engine check at $99/mo answers your actual question, Profound Starter does that job, and there is no reason to pay for more coverage than the question in front of you needs. If the question is broader, whether your brand shows up across the handful of assistants a buyer might actually open before ever landing on your site, the more concrete way to find out is to run that check yourself rather than take any vendor’s word for it, Profound’s or otherwise. A review, this one included, can tell you what a tool measures and what it costs. It cannot tell you what a specific assistant is currently saying about your specific brand, and that answer changes often enough that last quarter’s screenshot from any vendor is not something to plan a quarter around.

That is what the free analysis on the FoundCite homepage runs: the same 5-engine method described above, on your own site, before you commit to a monthly plan on either end of this comparison.