AI visibility tools pricing comparison: how we budget it for clients

Pricing is the first thing prospective clients ask about, and it is also where most of the confusion starts. Two tools can both advertise a low monthly price and deliver completely different programs. This is how we compare AI visibility tool pricing when we scope a client engagement, and what we have learned about the gaps between the headline number and the real one.

We keep competitor numbers out of this piece on purpose, apart from general patterns. Vendors change plans several times a year, and a stale price quoted in a blog post does more harm than no price. For specifics, check each vendor's live page.

The five things a price hides

When we put a quote next to another quote, we normalize them against five questions.

Engines. Is the headline price for all the models you care about, or for a few with the rest as add-ons? A plan that covers three engines is a different purchase from one that covers all of them.

Prompts and frequency. Prompt counts are easy to compare. Frequency is not. A daily run of 50 prompts and a weekly run of 50 prompts give you very different data, and daily runs often cost more or come on higher tiers.

Credits. Some vendors bill credits, where one credit may equal one AI response and heavier engines may use more. A credit plan can be fine, but you need to work out how fast your prompt set will use it. A multi-engine prompt set burns credits faster than the headline suggests.

Gating. Crawler logs, API access, exports, SSO, white-label reporting and city-level tracking are commonly reserved for higher tiers. Ask what you need to do the job, then price that tier.

What happens after measurement. A tool that only reports leaves the fix to you. The cost of the analyst time, writer time and engineering time is real, and it belongs in the comparison.

What we run on, and what it costs

We run client programs on Promptwatch, so here is its published pricing in USD, monthly billing. Annual billing is 12 months for the price of 10.

Explore, free: 10 prompts, ChatGPT only. We use it to show a prospect a first read, not to run a program.
Essential, $95: 50 prompts, 6,000 responses, all active models, site audit, content gap analysis, visitor analytics. No crawler logs.
Professional, $245: 150 prompts, 18,000 responses, crawler logs, 5 AEO articles, custom reports and the Data Studio connector.
Business, $579: 350 prompts, 42,000 responses, the automated Content Agent, state and city tracking, local GEO/SEO coordinate targeting, Shopping Insights, Ads Radar, audit log.
Agency plans: Kick-off $199, Growth $399, Scale $799. Unlimited projects, unlimited prompts, 10 seats, all models, crawler logs, Ads Radar and state and city tracking. They do not include AEO articles, the knowledge base, 360 Insights or the automated Content Agent. That matters when you scope content work.

A 7-day trial runs on monthly billing for Essential, Professional and the first two agency plans. Extra Agent Credits and extra AEO articles are available as add-ons, and yearly add-ons are 20% off.

How we match a plan to a client

A small business or solo operator often starts on Essential, or just Explore to see the baseline. The crawler logs are missing on Essential, so we use the weekly site audit for crawlability checks.

A mid-sized company with a content program is usually a Professional or Business conversation. Crawler logs start on Professional, which is where we can say why a page is or is not cited. Business adds the Content Agent and city-level tracking.

An agency or consultancy has a different problem: many clients, one bill. Agency plans give unlimited projects and prompts, which changes the economics compared to buying a brand plan per client. The question is volume of responses and crawler logs per month, which differ by tier.

An enterprise has custom needs: SSO, audit logs and managed content workflows. That is the Enterprise conversation.

Do not compare on price per prompt

Price per prompt is the number people reach for, and it misleads. Ten tracked prompts that map to real buyer questions are worth more than 500 generated prompts that do not. When we scope a program, we start from the question set, then decide how many prompts, which engines, how often, and whether crawler logs and attribution are needed. Only then do we price it.

Questions to ask any vendor

Which engines are in this plan, and which cost extra?
Is the prompt count per project or per account? How often do prompts run?
If it is credit-based, how many credits does my prompt set use in a month?
What is gated to higher tiers: API, exports, crawler logs, SSO, white-label?
Is the headline price monthly or the discounted annual rate?
What happens when I run out of credits or responses?
What do I do with the data, and does the product help me act?

We have seen quotes where the second and fourth answers changed the real cost by a large margin. The price on a homepage is rarely the price of the program.

A note on trials and contracts

Some vendors offer a short trial, some a free tier, some neither. If a vendor has no way to test the product with your own prompts before a contract, treat that as a cost. Run the same 20 prompts in each tool you are considering, and compare not just the data but how clearly each one tells you what to do next. That test costs little and tells you more than any price page.