Back to Blog
prop firm rules pageprop firm ai searchprop firm rules ai engines

How to Publish Prop Firm Rules That AI Engines Quote Correctly

How to build a prop firm rules page AI engines can quote: one page per program, rules in plain text, one wording, a dated changelog, schema and llms.txt.

Jonathan Jean-Philippe
Jonathan Jean-Philippe·Founder & GEO Specialist
10 min read
Published: November 21, 2026Last updated: November 21, 2026
A structured holographic rules page with rows, a table, a date chip and a version history ribbon, read by several AI engine orbs connected to it by beams of light

A prop firm sells a set of rules. When a trader asks an AI assistant whether your evaluation uses a trailing or a static drawdown, or how long the first payout takes, the answer is assembled from whatever the engine can find or remember. Your rules page is the one source you fully control, and this guide is about making it the easiest version to find, read and quote.

This is the how-to. The evidence behind it is in our 5-engine accuracy test of AI answers about prop firm rules, which measured how often assistants state a firm’s rules correctly. For the broader picture of AI visibility in this market, see AI visibility for prop firms and brokers.

What Did the Accuracy Test Show?

The short version

In our test of 30 rule questions about 6 prop firms, the two engines that searched the web answered 60 of 60 correctly, almost always citing the firm’s own site. The three engines that answered from model memory stated a correct value in 13 of 90 answers and a wrong one in 26. A clear, current, consistent official page is what the correct answers had in common.

On 21 November 2026 we asked ChatGPT, Perplexity, Gemini, Claude and Grok 30 factual questions about the rules of 6 well-documented prop firms, once each, and coded the answers against the firms’ official pages. Three findings shape this guide:

  • When an engine searched the web, it got the rule right. ChatGPT and Perplexity, both with web search, answered 60 of 60 questions correctly. ChatGPT cited the firm’s own site in 30 of 30 answers, Perplexity in 27 of 30. The official page was doing the work.
  • From model memory, stated values were mostly wrong. Gemini, Claude and Grok, queried without web search, declined or gave no usable value in 51 of 90 answers; when they stated a value, it was wrong in 26 of 39. The errors were the kind a rules page can prevent from spreading: an end-of-day drawdown called intraday, a static drawdown called trailing, a consistency rule invented or said to be removed, an activation fee that does not exist.
  • Third-party pages sit next to yours. Perplexity listed third-party pages, mostly comparison and review sites, in 30 of 30 answers, alongside the official page.

The test was small and the firms were chosen because their rules are easy to find, so it shows a pattern rather than a market-wide rate; the full method and its limits are in the accuracy test. The practical reading is simple: a firm cannot control what a model remembers, but it can make sure that any engine that looks finds one clear, current version, and that this version is the one that spreads.

Step 1: One Rules Page per Program

Most rule errors in AI answers are not invented from nothing. They are a real rule attached to the wrong product: the end-of-day drawdown of one plan given to another, the daily loss limit of a 50K applied to a 25K, a consistency rule that exists at payout described as an evaluation rule. When every program shares one long page, or when rules are spread across a dozen help articles, an engine has to stitch the pieces together, and that is where products get mixed.

Give each program its own page, with a stable URL and a title that names the program. On that page, state every parameter a trader asks about, for each account size:

  • Profit target, per phase, in dollars or percent.
  • Daily loss limit: the amount, what it is measured from (balance, equity, or the higher of the two), when it resets, and whether hitting it fails the account or only stops trading for the day. If it is optional, say so.
  • Maximum drawdown: the amount, the type (static, intraday trailing or end-of-day trailing), what the floor follows, and where it stops trailing.
  • Consistency rule: the percentage, how it is calculated, and the stage it applies to (evaluation, funded, at payout).
  • Profit split, including any tier or add-on.
  • Payout timing: when the first payout can be requested, what it requires, and how often after that.
  • Fees: evaluation price, activation fee, reset fee, monthly fee.
  • Changes at funding: if the drawdown model or any other rule changes once the trader is funded, say it on the same page.

Write each rule as a sentence that stands on its own, with the program name in it: “On the 50K [program name] evaluation, the maximum drawdown is [amount], end-of-day trailing, and it stops trailing at [balance].” A sentence like that can be lifted into an answer without losing its meaning. A table cell that only says “[amount]” under a header three screens up cannot.

Step 2: Every Rule in Plain Text

Rules published as an image, a PDF, a slider or a comparison table that only appears after JavaScript runs are, for many crawlers and fetchers, not there at all. Check the rules page the way a crawler sees it: view the page source (not the inspector) and search for each value. If the drawdown amount is not in the HTML your server sends, an engine that fetches the page may not see it.

  • Put every number in the HTML as text. Images can illustrate, never carry a rule.
  • Avoid hiding rules behind tabs or accordions that are only rendered after a click. If you use tabs, render all panels in the HTML and toggle visibility with CSS.
  • Use real HTML tables with header cells for per-size values.
  • Keep the rule and its condition together. “No daily loss limit” and “(available as an add-on)” belong in the same sentence.

The Citation Readiness score in a Rankeo audit looks at this kind of extractability on your pages, and the technical audit flags content that only appears after JavaScript runs.

Step 3: One Wording on Every Page

A prop firm typically states the same rule in four places: the pricing page, the program page, the FAQ and the help center. If one says “trailing drawdown”, another “end-of-day drawdown” and a third “max loss limit”, an engine sees three different facts. During the test we read the official pages of each firm to establish the ground truth, and on one of them the help article stated the lock level of the drawdown in the text and contradicted it in its own worked example. If a careful reader has to stop and ask support, an engine will not do better.

  • Pick one name per rule and use it everywhere (“Maximum Loss Limit, end-of-day trailing”), including in your marketing pages and emails.
  • Keep the values in one source in your CMS or codebase and render every page from it, so a change cannot reach the pricing page and miss the FAQ.
  • When a legacy rule still applies to older accounts, say so explicitly and date it: “Accounts bought before [date] keep the previous [value].” Legacy values without dates are how old numbers stay alive in answers.

Step 4: A Visible Date and a Changelog

Prop firm rules change often, and the old version does not disappear from the web. Two signals help a reader, and an engine, tell which version is current: a visible “last updated” date on the rules page, and a short changelog that lists each change with its effective date.

  • Show the date near the top of the page, in a format a person can read.
  • Change it only when a rule actually changes. A date that moves on every deploy stops meaning anything.
  • In the changelog, write the old and the new value: “From [date], the consistency rule at payout is [new value] (previously [old value]).” That sentence is what lets an engine, or a trader, recognize the old number as old.

Our guide to content freshness and AI citations covers when an update is worth signaling and how to do it without faking freshness.

Step 5: Structured Data That Matches the Page

Structured data does not make an engine quote you, and it does not replace the visible text. Its job here is to state, in a machine-readable form, the same facts the page already states, and to tie them to one entity. Use a single <script type="application/ld+json"> block with an @graph:

  • Organization, with legalName, url and sameAs pointing to your official profiles. Leave out aggregateRating about yourself: Google treats review markup about your own organization as self-serving and does not show it.
  • Product + Offer, one per program or account size, with the price and currency exactly as on the checkout page.
  • FAQPage, using the same questions and answers as the visible FAQ. Its value is a clean question-and-answer layer for extraction, not a special Google result.
  • WebPage with dateModified, set to the same date as the visible “last updated” line.

The rule that matters: every value in the markup must come from the same source as the visible page. Markup that says one price while the page says another is worse than no markup. Rankeo’s schema generator produces Organization, FAQPage and Product/Offer JSON-LD in one @graph; the JSON-LD glossary entry explains the format.

Step 6: Point AI Crawlers to the Rules With llms.txt

An llms.txt file at the root of your site is a short, plain-text map of your most important pages, written for AI crawlers. For a prop firm, list the rules page of each program first, then the payout policy, the fee page and the FAQ, each with a one-line description that names the program. It costs little, and it gives an engine a direct path to the official version. Our llms.txt guide covers the format; Rankeo’s llms.txt generator drafts one from your site.

Step 7: Work the Third-Party Sources

Your own pages are only part of what an engine reads. Comparison sites, review platforms, forums and video descriptions describe your rules too, and they are not updated when you change a rule. Off-site GEO explains why those sources weigh so heavily in AI answers in general.

In our accuracy test, the web-search answers that listed third-party pages leaned on prop firm comparison and review sites, next to the official page. The answers were correct on that date, but each of those pages is a copy of your rules that you do not edit. After a rule change, the copies are where an outdated version lives on.

  • Keep a list of the third-party pages that describe your programs, starting with the ones engines cite.
  • After each rule change, send the sites that list you the change, its effective date and the URL of the official rules page.
  • Correct the sources you control outside your site: your own video descriptions, pinned community posts, social bios.

How to Check Whether It Worked

Write down the five or six rules traders ask about most, phrase them the way a trader would, and ask them to the AI engines that matter to you. For each answer, record the engine, the date, the passage about your firm, the sources it cited, and a code: correct, outdated, wrong, vague or no answer. Repeat after each rule change, and again once the updated pages have had time to be recrawled. The AI citation tracking guide covers how to build and run that question set. Keep two scoreboards apart: whether an engine mentions you at all, and whether it describes you correctly (the mention versus citation gap explains why being named and being used as a source are different things).

Rankeo can run the questioning part: it queries 5 AI engines on demand (ChatGPT and Perplexity with live web search, Gemini, Claude and Grok from model memory) and shows, for each answer, whether your firm is mentioned, cited as a source or recommended, the passage, and the third-party sources behind it. The accuracy check itself stays with your team: Rankeo does not compare answers with your rules page, does not scan continuously and does not send alerts.

See what AI engines say about your prop firm

Start with a free audit of your domain, then compare the answers with your rules page.

Run a free audit →

This article is about publishing information, not about trading. Nothing in it is financial advice or a recommendation of any firm or program.

Frequently Asked Questions

Jonathan Jean-Philippe
Jonathan Jean-Philippe

Founder & GEO Specialist

Jonathan is the founder of Rankeo, a platform combining traditional SEO auditing with AI visibility tracking (GEO). He has personally audited 500+ websites for AI citation readiness and developed the Rankeo Authority Score — a composite metric that includes AI visibility alongside traditional SEO signals. His research on how ChatGPT, Perplexity, and Gemini cite websites has been used by SEO agencies across Europe.

  • ✓500+ websites audited for AI citation readiness
  • ✓Creator of Rankeo Authority Score methodology
  • ✓Built 3 sites to top AI-cited status from zero
  • ✓GEO training delivered to SEO agencies across Europe