How do I make my pricing readable by AI?
If an AI assistant cannot parse your pricing, it cannot tell a buyer what you cost. Here is how to publish pricing that both people and AI tools can read cleanly.
Machine-readable pricing means your prices live on the page as actual text and structured data, not as a picture, a JavaScript slider, or a number you only reveal after a sales call. When pricing is readable, an AI assistant like ChatGPT or Claude can quote what you charge to someone who asks. When it is not, the assistant either guesses, hedges, or names a competitor it can read instead.
Why does AI struggle with pricing hidden in images, sliders, or contact-us flows?
AI tools read text, not pictures of text. If your price is baked into a hero graphic, a screenshot, or a fancy SVG, the assistant sees an image with no number in it. The same goes for prices that only appear after a user drags a slider, picks a seat count, or clicks through a multi-step calculator. The number exists, but it is not in the page content the tool can pull.
Contact-us pricing is the hardest case. If the only way to learn your price is to fill out a form and wait for an email, there is nothing for an AI tool to quote. When a buyer asks 'how much does this cost,' the honest answer the assistant can give is 'they do not publish a price,' which is the answer that loses you the consideration.
- ▸Price in an image: the assistant sees a graphic, not a number, so it cannot quote you.
- ▸Price behind a slider or calculator: the value is generated by code on click, so it is not in the readable page text.
- ▸Price behind contact-us: there is no number to read at all, so you get skipped or hedged.
- ▸Price split across tabs or modals that load on interaction: the assistant may only see the default tab.
//A quick way to check what AI sees
Open your pricing page, then view source or use your browser's reader mode. If the actual dollar amounts do not show up as plain text, an AI tool probably cannot read them either. Reader mode is a rough but honest stand-in for how a parser sees your page.
What does machine-readable pricing mean in plain terms?
Machine-readable pricing means a machine can find your price, read it, and understand what it buys, without a human in the loop. In practice that is three things: the number is present as selectable text, it sits next to a clear label for the plan, and it states the basic terms (what is included, one-time or recurring). If a person could copy and paste the line into a note, an AI tool can usually parse it too.
This is the same blind spot we see in a lot of AI-built apps: the pricing looks perfect to you because you can see the rendered page, the tooltips, and the toggle. The tool reading your site sees only the raw text, and the raw text is often missing the very numbers you are most proud of.
How do you publish a clean pricing page and a plain-text pricing file?
Do two things. First, make your normal pricing page render the price as real text right next to each plan name. Second, publish a plain-text pricing file at a stable URL that lists every plan in a simple, parseable format. The web page is for humans who skim. The text file is the clean, no-distractions copy that AI tools and buying agents can read without fighting your layout.
- 1.Put the dollar amount on the page as text, not inside an image or an icon font.
- 2.Label each plan clearly so the price and the plan name sit together (for example 'Quick review: $5').
- 3.State one-time vs recurring in words, so 'per month' or 'one-time' is never ambiguous.
- 4.List what each plan includes in plain language, in a few short bullets.
- 5.Publish a plain-text or markdown pricing file at a fixed path like /pricing.md and keep it in sync with the page.
- 6.Link to that file from your llms.txt if you have one, so tools know where to look.
That last step ties into a bigger pattern. A plain-text pricing file is a close cousin of llms.txt: both are quiet, parseable files that tell AI tools the truth about your product in a format they can lift cleanly. You are not gaming anything. You are just publishing the facts in a way a reader and a parser both understand.
What details should the pricing file include?
For each plan, include the price, whether it is one-time or recurring, what the buyer gets, and any time or delivery detail that matters. Lead with the number. Keep each entry self-contained so a tool can lift one plan and still have the full picture. Here is the format we use for our own pricing, which you can copy and adapt.
# Acme pricing
One-time pricing. No subscription.
## Starter
- Price: $5 (one-time)
- Delivery: about 10 minutes
- Includes: a fast review, a score, and a badge
## Pro
- Price: $15 (one-time)
- Delivery: up to 2 hours
- Includes: everything in Starter, plus a real person testing your app
Contact: support@acme.comNotice what makes this work: every plan starts with the price, says one-time in plain words, and lists what is included in short bullets. There is no slider to drag and no image to decode. A person can read it in ten seconds, and so can an assistant. If you have a private plan you do not publish, simply leave it out of the file. The file should only describe what you actually sell openly.
+Match the file to the page
If your pricing file says $15 and your page says $19, an AI tool may quote either one, and you will look careless to the buyer who notices. Pick one source of truth and update both together. Stale pricing erodes trust faster than no pricing at all.
Why does clarity help both buyers and the AI tools they ask?
Clear pricing removes the same friction for both audiences. A human who has to hunt for the price assumes it is expensive or assumes you are hiding something, and many leave before asking. An AI tool that cannot find the price does the machine version of the same thing: it hedges or recommends a competitor it can quote with confidence. In both cases the cost is happening right now, on every visit, whether or not anyone tells you.
There is a quieter upside too. When an AI assistant can state your exact price and what it includes, it is effectively recommending you with specifics, which reads as more credible than a vague mention. That is borrowed authority you earn by being legible. It is the same reason reviews help AI find and recommend you: clear, structured, true information is what these tools prefer to repeat.
How do AI buying agents read and act on pricing?
A buying agent works on behalf of a person, and it does roughly what a careful shopper would: it reads your page, looks for the price and terms, compares options, and reports back or acts. The difference is that an agent has no patience for ambiguity. If it cannot tell whether $15 is monthly or one-time, it will not guess in your favor. It will flag the uncertainty or move on to a vendor whose terms are unambiguous.
This is why stating one-time vs recurring in words matters so much. An agent comparing three tools needs to compare like with like. 'Quick review: $5 one-time' is something it can rank and recommend. '$5' with no term is a question mark, and question marks lose to certainty. You do not need a special agent API to be agent-ready. You need pricing a careful reader could not misread, which is the same standard that serves your human visitors. For the bigger picture on this, see how to get recommended by ChatGPT and Claude.
What is a simple format you can copy and adapt?
Use the markdown block above as your template. Give the file a stable home like yoursite.com/pricing.md, keep one plan per section, lead each section with the price, and always say one-time or per month in words. Update it whenever the real prices change. That is the whole job. It is small, it is honest, and it makes you readable to the people and the tools deciding whether to buy.
If you would rather have an outside set of eyes confirm that your pricing actually reads cleanly to a stranger and to an AI tool, that is exactly what a quick review checks. We read your app the way a first-time visitor and an assistant would, and tell you what is clear and what is not. You can send us your app when you are ready.
Get a fast, honest read on whether your pricing and your app are clear to real people and to the AI tools they ask. A quick review is $5 and lands in about ten minutes.
Get my app reviewedFrequently asked questions
Should pricing be in text and not an image? ▾
Yes. AI tools and search crawlers read text, not pictures of text. A price baked into an image, icon font, or screenshot is invisible to them, so they cannot quote you. Put the dollar amount on the page as real, selectable text right next to each plan name, and keep an image only as decoration if you want one.
What is a pricing.md file and do I need one? ▾
A pricing.md file is a plain-text or markdown page that lists every plan with its price, terms, and what is included, published at a stable URL like yoursite.com/pricing.md. You do not strictly need one, but it gives AI tools a clean, distraction-free copy of your pricing to read, which raises the odds they quote you accurately instead of guessing.
Does machine-readable pricing require structured data or schema? ▾
It helps but it is not required to start. The single most important step is publishing the price as plain text with clear terms. Adding structured data, such as Offer markup, gives crawlers an even more explicit signal. Begin with readable text and a plain-text pricing file, then layer schema on later if you want the extra clarity.
How can I tell if AI can read my pricing right now? ▾
Open your pricing page in your browser's reader mode or view the page source. If the actual dollar amounts do not appear as plain text, an AI tool likely cannot read them either. You can also ask an assistant directly what your product costs and see whether it answers correctly or hedges.
What if I do not want to publish my price? ▾
That is a real business choice, but understand the trade-off: a buyer who cannot find your price often assumes the worst and leaves, and an AI tool asked what you cost will say you do not publish one, which usually loses the recommendation. If you must keep one plan private, still publish clear prices for the plans you sell openly.
Is your pricing clear to a stranger and to ChatGPT?
We read your app and your pricing the way a first-time visitor and an AI tool would, then tell you in plain English what is clear and what is not. A quick review is $5, delivered in about ten minutes.
Get my app reviewedKeep reading
How Do I Get My App Recommended by ChatGPT?
When people ask AI assistants for a tool like yours, you want to be one of the answers. Here is how AI tools find and recommend apps, and how to make yours easy to cite.
What is llms.txt and does my app need one?
llms.txt is a simple file that tells AI tools what your site is and where the important pages are. Here is what it does, what goes in it, and how to add one today.
How Do Reviews Help AI Find and Recommend You?
An honest, published review is more than a trust signal. It is a clean page AI tools can read and quote when someone asks for a tool like yours. Here is how that works.
We put every SaaS through the same honest scorecard, then publish the result.