Profound AI Search Visibility for ChatGPT and Perplexity

Profound AI search visibility for ChatGPT and Perplexity is the brief that arrives with a logo and one engine. The Starter card is $99 annual, ChatGPT only, 50 prompts. Perplexity is not on that card. We say that in the first meeting, because the gap is the whole project. Growth is $399 annual for three engines. Enterprise is unpublished, reported at $2,000 to $5,000 or more per month, with months of onboarding. The pricing ladder tells you what the product actually is at each tier, and the bottom rung is a single-engine mention tracker dressed up with an enterprise brand.

Read the ladder as a product description, not just a price list. Starter gives you 50 prompts on one engine. That is enough to answer "did ChatGPT name us this week" and nothing else. Growth adds two more engines and lifts the prompt ceiling, but you are still buying mentions, not the crawl and visit layer that turns a mention program into a traffic program. Enterprise is where the full coverage lives, and even there the onboarding is measured in months. When a client shows us a Profound logo on a slide and asks why the Perplexity column is empty, the answer is almost always the tier, not the vendor. The tier is the product.

The ChatGPT-plus-Perplexity log we can start this week is Promptwatch. Essential is $95 per month for prompts, citations, and visitors, but it has no listed crawler-log allowance. Professional is the first brand plan with crawler logs. Explore is free and ChatGPT only, which is the honest comparison to Profound Starter. We bring this up early so nobody on the client side confuses a free taste with a working program.

The reason we name Explore next to Starter is that both are single-engine, low-volume cards. Matching them keeps the conversation honest. If the client is comparing a free Promptwatch login to a paid Profound Starter login, they are comparing two mention trackers, and the difference is price, not capability. The capability gap opens at Professional, where crawler logs arrive, and at the agency tiers, where the response and log budgets get large enough to run a real program across multiple brands.

Procurement hears the word Profound and assumes both engines. The public Starter card is a ChatGPT mention product. Calling that "enterprise AI search" in a steering deck is how we get fired later, when someone opens the tool in a QBR and finds one column where the brief promised two. We would rather have the awkward conversation in week one than the angry one in month four.

The awkward conversation is cheap. It costs a meeting and a slide note. The angry conversation costs the account. Procurement buys the word, the analyst runs the tool, and the analyst is the one who has to explain the missing column to a marketing director who was promised "full AI visibility." Putting the SKU on the slide in week one moves that explanation upstream, where it can still change the contract instead of just the report.

How we handle "we already bought Profound"

Read the invoice. The SKU on the invoice is the only thing that tells you what was actually bought. Marketing slides are not the contract. The slide says "Profound" and the invoice says "Profound Starter, ChatGPT, 50 prompts, annual." Those are two different documents describing two different products, and the invoice is the one that runs.

If it is Starter, Promptwatch holds Perplexity, and usually Gemini, Claude, and Overviews as well. Slide titles will not imply four engines. We will write "Profound Starter: ChatGPT" on the slide if we have to, because a labeled limitation is better than an unlabeled promise. The client keeps the Profound login for the ChatGPT column, and we add the second column on Promptwatch. This split is not a slight against Profound. It is a way to use the SKU that was actually paid for instead of pretending it covers more than it does.

If it is Growth, we confirm Perplexity is one of the three engines on the contract. If it is not, we still add a log. The $399 annual bill is real money either way. We will start Essential if Perplexity is in the statement of work now, rather than waiting for a contract change that may take a quarter. Waiting is how a quarter of measurement gets lost. A quarter of empty Perplexity data is a quarter where the client pays for a program and gets a screenshot.

If it is Enterprise and the engines match, we work from Profound as the primary. We still want crawler logs and visitor analytics on top of it. If those are missing from the login we are given, Promptwatch Professional, Business, or an agency plan holds the crawler slice. Essential can hold the visitor side. We do not sit idle for one to three months of Profound setup while the program is supposed to be running. The prompt list can start this week, and the crawl layer can join it when the enterprise onboarding finishes. Running the prompt layer and the crawl layer in parallel instead of in sequence is how you keep a quarterly cadence when a vendor onboarding is slow.

G2 reviewers have seen citation counts from Profound that do not match manual ChatGPT checks. We treat Profound numbers as directional at any tier. That is not an insult to the vendor. It is how we write the SOW, so the client knows the number is a signal and not an audit. Directional is useful. Directional sold as exact is the problem. A directional number tells you whether you are trending up or down. An exact number tells you whether you hit a target. Confusing the two is how targets get missed and someone gets blamed for a rounding error.

Otterly includes both engines from $29 with lag. Peec can include both inside three models at $95. Neither is the log we run when the retainer needs citations, crawls, and visits on a weekday. They are mention checks. A mention check answers a different question than a citation program. A mention check says the brand appeared. A citation program says where, why, whether the bot fetched the page, and whether anyone followed the link. Those are different deliverables sold at different prices, and the client should know which one they bought.

Allow OAI-SearchBot if ChatGPT Search is in scope. GSC generative reports stay for Overviews. They are not Perplexity, and they are not a substitute for the Perplexity column. Google and Perplexity are separate systems with separate footnotes. A Perplexity footnote and a Google AI Overview link are two different events on two different surfaces, and reporting them as one "AI visibility" number hides which surface actually produced the click.

What we report

Two columns. ChatGPT and Perplexity. Named versus cited URL. We do not blend them into one "AI visibility" number, because blending hides which engine did what. ChatGPT has 820 million or more weekly active users. Perplexity has 22 million or more monthly. The smaller monthly audience still footnotes more often, which means a Perplexity citation can drive a real click where a ChatGPT name-drop does not. An empty column is worse than a small audience, because an empty column means you bought the wrong SKU for the brief.

The named-versus-cited split matters inside each column. A model naming your brand is one event. A model linking to your docs is a different event, with a different downstream action. The first can move perception. The second can move traffic. Reporting them as one number hides which one you actually have. A client who reads "47 AI mentions this month" does not know whether that is 47 name-drops or 47 links, and those two outcomes deserve different reactions from the content team.

Kick-off at $199 per month with crawler logs is the choice if we hold more than one brand. Professional at $245 per month is the single-brand choice when crawler evidence is required in the report. Business at $579 per month is the choice if they need five projects under one roof. We do not put a roster on Essential, because Essential is one project and one seat, and a roster on a one-seat plan is how accounts end up shared by password. The plan limits are not fine print. They are the thing that decides which tier fits the retainer.

FAQ

Should we wait for Growth to add Perplexity?

$399 annual is a real bill. We will start Essential if Perplexity is in the SOW now. Waiting for a contract amendment is not a measurement strategy. A contract amendment is a procurement event. Measurement is a weekly cadence. The two run on different clocks, and the measurement clock should not wait for the procurement clock.

Do we cancel Profound?

Not if Enterprise is live and the engines match. We cancel the habit of calling Starter a platform. Starter is a ChatGPT mention tracker, and labeling it correctly costs nothing. The label is the fix. The contract stays, the slide changes, and the client stops expecting four engines from a one-engine card.

Does Promptwatch replace ten-engine Profound?

No. It replaces Starter and the missing Perplexity column. At Enterprise coverage with matched engines, Profound stays primary and Promptwatch fills the layers Profound does not ship on the login we get. Filling layers is a different job from replacing a platform, and the SOW should name it that way so the client sees two tools with two roles instead of two overlapping bills.

What to do this week

1. Forward the Profound SKU to us so we can read the actual line item.
2. Run 15 prompts on both engines by hand and save the answers.
3. Load Promptwatch Essential if only the Perplexity prompt and citation column is empty. Use Professional, Business, or an agency plan when crawler logs are also required.
4. Report two columns, named versus cited, and keep them separate.
5. Email hello@1001seomedia.com if you want us to wire the second column this week.