Is paying for an AI visibility platform worth it, or just another vanity metric?
We get this question from clients and from ourselves. Our answer: a visibility score is vanity unless it connects to crawling, citations, traffic and decisions.
Clients ask us this early, and some ask it with a raised eyebrow. They have been burned by dashboards before. A score that moves up and to the right is pleasant and proves nothing. So here is how we answer it, based on running programs on a platform every week.
Where the vanity charge is fair
A visibility score alone is thin. If all you can see is "your brand was mentioned in 34% of tracked answers," you have no idea why, no idea whether it matters, and no next step. We have seen reports like that, and we would not pay for them either.
The score also has real limits that any honest vendor should admit. AI answers change between runs. The prompts you track are a sample of what people ask, not all of it. And nobody can guarantee a recommendation. If a pitch promises one, treat it as a warning.
When it earns its fee
For us it comes down to whether the platform answers four practical questions.
Can AI bots reach the site? This is often the most valuable finding and it has nothing to do with the score. In Promptwatch, Agent Analytics shows IP-verified crawler logs for the real agents, such as GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot and PerplexityBot. Spoofed user agents are filtered out. A blocked crawler is cheap to fix and expensive to ignore.
Which sources do the engines trust on our prompts? Citation analytics at page and domain level, plus Reddit, YouTube and LinkedIn reports, tell us where to publish or earn mentions. It replaces opinion with a list.
Did anyone arrive? Visitor analytics, installed with a script, a GTM template or the WordPress plugin, counts visits from AI engines and conversions. This is what ends the vanity argument in a client meeting, because it speaks the language of the finance team.
What do we do about it? The Action Board turns the data into a prioritized list, and a Discovery agent investigates weekly and suggests up to five evidence-backed actions per run. We assign them, comment on them, and review the weekly digest email.
If a tool can only answer the first mention question, we call it a tracker. Trackers are fine for a pulse check. They are not what we would build a client program around.
What the published case studies suggest
We are careful here, because these belong to the customers, not to every user. Crisp reported that AI-referred traffic converted at twice the rate of traditional channels. TreeHouse, the group behind Pararius and Huurwoningen.nl, reported that AI search became about 25% of organic conversions, 99% of it from ChatGPT, after fixing crawler access and automating weekly reporting. We cite them as evidence that the measurement chain can reach revenue, not as something to expect.
How we test before committing a client
We run a short pilot on the client's own prompts. The free Explore plan has 10 prompts and 310 responses, ChatGPT only, which is enough to see what the output looks like. Essential, Professional and the lower agency plans have a 7-day trial on monthly billing. Then we judge on three things:
- Did the data change a decision within the first month?
- Could we explain every number back to a real prompt, response and source?
- Did we find something the client's existing analytics could not show?
If the answer to all three is no, the honest advice is to stop paying.
Plan details that affect the answer
Crawler logs are not on Essential ($95/mo). They start on Professional at $245/mo. Business at $579/mo adds the automated Content Agent, state and city tracking, and Ads Radar. Agency plans start at $199/mo for Kick-off and include crawler logs and visitor analytics. The right plan depends on which of the four questions you need answered.
When we say wait
There are real cases where not buying is correct:
- Nobody owns the output. A report no one reads is the real vanity expense.
- The audience does not use AI search for this category. Check before you spend.
- The site cannot be changed. If you cannot fix crawler access or update pages, measurement alone will frustrate you.
Our answer
Paying is worth it when the platform joins crawling, citations, traffic and action, and when someone on the team will act on it. In that case the score is a summary of work that is actually moving. When it is a standalone number, it is a vanity metric, and you can get it for free.
That is why we run client programs on Promptwatch: the score is the least interesting thing on the screen, and everything behind it is what we bill our time against.