You run thisWritten for ChatGPT, Claude or Gemini (with web access)
Work out what a vendor will actually charge me
Founder, Automation Squad
Pricing pages are designed to produce one impressive number and defer everything that determines your actual bill. Sections 3 and 4 are where the money hides — the minimum you can really spend, and the thing you assumed was included — and section 7 turns a vague unease into a specific list you can put in front of a salesperson.
The prompt
I'm trying to work out what [VENDOR / PRODUCT] will really cost us. Their pricing page is here: [URL — or paste it] Us: [TEAM SIZE, YOUR REALISTIC USAGE, ANY EXISTING CONTRACT OR PLAN] Go and read it, then answer as though you're on my side of the table. 1. THE HEADLINE NUMBER, and whether anyone actually pays it. Say plainly if the advertised price applies to a configuration nobody real uses. 2. WHAT'S PER-SEAT AND WHAT'S USAGE-BASED. Separate them. These behave completely differently as we grow, and vendors mix them deliberately. 3. THE FLOOR. Minimum seats, minimum term, minimum spend, annual-only tiers. The smallest cheque I can actually write, not the smallest number on the page. 4. WHAT'S NOT INCLUDED that I'd assume is. SSO, audit logs, support with a response time, API access, extra environments, data export. There's usually at least one, and it's usually the one you find out about in month three. 5. THE OVERAGE. What happens when I exceed a tier — does it throttle, auto-upgrade, or bill at a punitive rate? Quote the exact wording. 6. MY REALISTIC ANNUAL COST at the usage I described. Show the arithmetic. Give me a low, expected and high figure, and say what drives the high one. 7. WHAT I'D HAVE TO ASK SALES, because the page genuinely doesn't say. List the exact questions. Cite the page and note the date you read it. If pricing is entirely hidden behind "contact us", say so and tell me what third-party evidence exists, clearly labelled as third-party.
Copy it, paste it, run it. Nothing here is gated.
More for the same job
Written for ChatGPT, Claude or Gemini (with web access) as of August 2026. Models change — if this stops working the way it should, that is worth telling us about.
