Terms of Service

Last updated: 2026-09-01

These terms govern your use of Rankli, operated by Rankli (“we”, “us”). By creating an account or using the service you agree to them. If you are accepting on behalf of a company, you confirm you have the authority to bind it.

1. What the service is

Rankli measures how AI answer engines — ChatGPT, Perplexity, Gemini and Claude — respond to questions your buyers might ask about your category. It runs a set of prompts repeatedly against those engines, records the full answers and their citations, and computes visibility metrics with confidence intervals. On plans that include experiments, it holds out a control set of prompts, registers your intervention as a hypothesis before you make it, and reports a difference-in-differences readout afterward.

When no engine API keys are configured for a measurement round, the engines are simulated and every screen that shows the resulting numbers says so. Simulated numbers demonstrate the product; they say nothing about your actual AI visibility.

2. What the service does not promise

This section is the product, so it is written plainly:

• We do not guarantee any change in your AI visibility — not in citations, mentions, recommendations, traffic, or revenue. Rankli measures; it does not move the engines.
• Measurements are proxies. We query engine APIs, which bypass the browsing history and personalization that shape what any individual buyer actually sees. The same prompt can return materially different answers run to run; that is why every number we report carries an interval, and the interval is part of the measurement.
• The engines change without notice. A platform-side change can move every brand’s numbers in a month, independent of anything you or we did. The control arm exists to subtract that, but it cannot always succeed.
• “Inconclusive” is a valid result. When an effect cannot be separated from noise, the product says so and tells you what more data would be needed. An inconclusive readout is the service working as designed, not a defect.
• Per-plan “detectable effect” figures are indicative estimates. The real minimum detectable effect is computed per experiment from your own baseline variance and may be larger or smaller.

3. Accounts and access

Sign-in is by passwordless magic link sent to your email address; there is no password to set and none is stored. Anyone with access to your email inbox can access your account, so keeping that inbox secure is your responsibility. You must provide an email address you control and keep it current.

4. Plans, billing and cancellation

• Payments are processed by Stripe. Your card details go to Stripe directly and never touch our servers.
• The Trial is a $1 one-time purchase of a single measurement round. It does not renew and does not convert into a subscription without your action.
• Subscriptions bill monthly and renew automatically. You can cancel at any time from the billing page; cancellation takes effect at the end of the paid period, and we do not refund partial periods.
• Plan limits (prompts, projects, runs per prompt) are fixed at the time of purchase for the current period. If we change a plan’s definition, existing subscribers keep their purchased limits until renewal.
• We may change prices with at least 30 days’ notice; changes apply from your next renewal, never mid-period.

5. Acceptable use

• Use the service only for lawful purposes and only to measure brands and categories you have a legitimate interest in.
• Do not probe, overload, or attempt to bypass the service’s limits or security, and do not use it to violate the terms of the AI engines it measures.
• Do not strip the confidence intervals and caveats from our readouts and republish the point estimates as guarantees. Reselling Rankli reports as your own service is permitted only on the Agency plan, which exists for exactly that.

6. Your data and ours

You own the data you put in — your domain, brand names, competitor list, category description — and the measurements and reports generated for your account. We own the software, the estimators, and the calibration harness. We may use aggregated, de-identified measurement data to calibrate and improve the estimators; we will never publish or share anything that identifies your brand or your numbers without your consent. Details of what we store and where are in the privacy policy.

7. Third-party engines

Measurement necessarily sends prompts to third-party model providers — OpenAI, Perplexity, Google, and Anthropic. Those prompts contain your brand name, your competitors’ names, and category phrasing, and the providers handle them under their own terms. We cannot control how those providers use or retain API traffic.

8. Disclaimers and limitation of liability

The service is provided “as is” without warranties of any kind, express or implied, including merchantability, fitness for a particular purpose, and non-infringement. To the maximum extent permitted by law, our total liability for any claim arising from the service is limited to the amounts you paid us in the twelve months before the claim, and we are not liable for indirect, incidental, or consequential damages — including decisions made on the basis of measurements that were, as disclosed, uncertain.

9. Termination

You can stop using the service and delete your account at any time. We may suspend or terminate accounts that violate these terms, with notice where practical. On termination, we delete your account data as described in the privacy policy.

10. Changes and governing law

We may update these terms; material changes will be announced by email to the account address at least 14 days before they take effect. Continued use after that date is acceptance. Governing law and venue: to be determined with counsel before launch — this is one of the items the review at the top of this page is for.