Enterprise-leaning AI search visibility tracking, with engine coverage gated hard by plan tier.
TL;DR verdict
The most capable product we assessed, and the one whose pricing page is most likely to mislead you. It is also the only vendor here putting a response commitment in writing. Compare it on the middle tier, because the cheapest one watches a single assistant.
Profound tracks how answer engines describe your brand and benchmarks you against competitors, across up to nine assistants at the top tier. What separates it is the second data source: connections to CDN and hosting platforms let it read AI crawler and agent traffic arriving at your site, so it observes who actually came rather than only sampling what engines say. It is built for larger marketing teams, with agency workspaces and per-client credit allocation.
The capabilities it ships, and which of them stand out.
Standout features
What the product does not do, kept separate from what we could not establish.
Product limitations
Could not verify
Teams who need evidence, not inference
Reading agent traffic at your own edge produces a number sourced from infrastructure you control, not from sampling. Where this has to survive scrutiny from outside marketing, that difference matters.
Organisations with a procurement process
Single sign-on, a written turnaround and a named specialist are what a security review asks for, and this is the only vendor here carrying all three.
Agencies reselling visibility reporting
Credits allocate per client rather than draining one shared pool, which is what makes the economics work when you are billing several accounts.
Anyone budgeting from the advertised figure
That plan watches one assistant. The realistic entry point is four times higher, and comparing on the lower number produces a decision you will have to redo.
Small teams testing the water
There is no proportionate small plan. The cheapest thing worth running costs more per year than several rivals' top self-serve tiers.
Teams who will pipe this into their own systems
Programmatic access needs the unpriced tier, so a reporting build cannot be scoped or costed in advance.
No free plan; paid from $99/mo billed yearly
| Plan | Published price | What the tier includes |
|---|---|---|
| Starter | $99/mo |
|
| Growth | $399/mo |
|
| Enterprise | Custom |
|
Pricing watchouts
What Profound connects to, and what it commits to in writing.
Integrations
Support
Email support on every tier, with a dedicated Slack channel and a named specialist on a 24-hour SLA at Enterprise — the only published response-time commitment in this category. The integration list is the broadest here and unusual in kind: CDN and hosting platforms alongside analytics, which is what lets it read agent traffic at the edge. The API is Enterprise-only, marked "No" on both published tiers. Checked 12 August 2026.
Almost every tool in this category works the same way: it asks assistants a set of questions on a schedule and records whether you were named. That is sampling, and sampling has a known weakness. It tells you what an engine said to a synthetic asker at a moment in time, and it cannot tell you whether anybody actually arrived.
The edge integrations answer the second question directly. By reading traffic at the CDN or host, the product observes AI crawlers and agents reaching your pages, which is a record drawn from your own infrastructure rather than from a rehearsal. When someone senior asks whether any of this is real, a server log is a considerably better answer than a visibility index. We have not run this on a live account, so treat the capability score as a reading of published specification rather than of output we have seen.
The thing most likely to lead a buyer astray here is not a hidden cost, it is a published one. The figure on the marketing pages buys a tracker that watches a single assistant, which nobody choosing between vendors would actually deploy. Comparing that number against a rival's fully featured entry plan is comparing a demonstration with a product.
Run the comparison at $399 instead and the picture is honest: a strong, well-supported product at a price that reflects the buyer it is built for. That is a defensible position and the vendor would probably do better stating it plainly, because the current structure invites a shortlist to include it for the wrong reason and drop it later for the right one.
Where another tool in the category fits a reader better.
Choose this if you want the full assistant coverage on the entry plan rather than at the ceiling, plus persona modelling and page-level audits.
Read reviewChoose this if the budget must be small and knowable. Nothing like the depth, but every figure is published and programmatic access arrives early.
Read reviewChoose this if analytics and programmatic access matter more than depth, and you want both at a published price rather than through procurement.
Read reviewAimed at what the public site cannot settle.
Starter is $99 a month billed yearly, Growth is $399 a month, and Enterprise is quoted. The two lower figures are published; the tier carrying nine assistants and the API is not. Checked 12 August 2026.
ChatGPT only. It is billed annually, so the commitment is $1,188, and it covers one assistant on a ranking where seven are available elsewhere as standard. Treat it as an evaluation, not a deployment.
Yes, on Enterprise only. It is marked unavailable on both published tiers, so scoping any integration with your own systems requires a sales conversation and a quote.
To read AI crawler and agent traffic arriving at your site at the edge. That gives it a second measurement taken from infrastructure you control, rather than one inferred by sampling replies, which is a data source no rival on our ranking has.
Yes, at Enterprise: a named specialist on a 24-hour commitment, alongside a dedicated Slack channel. That is the only published turnaround we found anywhere in the category, and it is one reason the support score here is the highest of the twelve.
Scrunch gives you full assistant coverage and persona modelling on its entry plan for $250. Profound gives you the edge measurement, the identity controls and a support commitment, but the comparable plan is $399 and the deepest coverage is unpriced. Choose Scrunch for the measurement, Profound for the procurement and the server-side evidence.
From the vendor's own published pages, read at source on 12 August 2026, and scored against the five review criteria. We have not used the product, so ease of use is left unscored rather than estimated.
This is what you buy when AI visibility has become an organisational concern rather than a marketing experiment. Procurement finds the sign-on controls, the response commitment and the named specialist it expects, and the edge connections give you a measurement sourced from your own servers that nothing else here can match. Price it honestly, though. The entry plan is a demonstration with an invoice attached, the $399 tier is the real product, and anything touching your internal stack means an unpriced contract. Settle that last point before signing.
Visit ProfoundUpdated August 2026
Research-based. Everything here comes from the vendor’s own published pages and verified third-party sources, read on the date below. Where we report a claim we could not check, the review says so. How we rank
| Criterion | Weight | Score |
|---|---|---|
| Core capabilityHow well it does the job it exists for. | 30% | 9/10 |
| Value and real costWhat you actually pay, and whether you can find that out before you buy. | 25% | 6/10 |
| Ease of useSetup, day-to-day use and how long it takes to get something useful. Only scored when we have used it. | 20% | Not scored |
| Integrations and extensibilityWhat it connects to, and whether your data can come back out. | 15% | 7/10 |
| Support and documentationWhat is committed to in writing, and what you can read without a sales call. | 10% | 8/10 |
Rewritten to the rebuilt review format and rescored against the five review criteria rather than inheriting the AEO ranking score. The rating moves from 3.5 to 4.0. Nothing about the product changed: the review frame weights capability above cost transparency and does not credit the observable getting-started signals the ranking does, which suits a deep, well-supported product carrying a confusing price structure. The ranking is a separate assessment with its own criteria and is unaffected.
Integrations scored up from 4 to 7 and support from 5 to 8 after a category-wide audit. The CDN and hosting connections and the 24-hour response commitment had never been credited, and both are unmatched anywhere else on the ranking.
By Rajat Kapoor. SaaSInsight is funded by labelled placements and affiliate links. Some links may be paid — always labelled. Read our disclosure policy.
