Every tool on CompareMyAI receives a rating out of 10. This page explains exactly what goes into that number, which criteria shape it, and what a score does — and doesn't — tell you.
What a score is
A score is our editorial rating out of 10, judged against the same five criteria for every tool and grounded in published pricing, features, and compliance data. It is not a hands-on review, and it is not the result of lab testing — we gather what each vendor publishes on its pricing pages, documentation, security certifications, and terms of service, and rate tools consistently against that.
Two tools scoring 8.3 and 8.1 are genuinely close. A small score difference reflects a small difference in the underlying data — not a definitive verdict. Your best choice depends on your specific needs, which is why we show you the full picture alongside the score.
Scoring criteria
Five criteria shape the score. Together they reflect what matters most when comparing AI tools for real-world use — pricing, capability, security, data handling, and transparency.
Pricing and value
We compare published pricing tiers across tools in the same category. We look at what each tier actually gives you — usage limits, feature access, and whether a meaningful free tier exists — not just the headline price. (See the note on currency below for how we compare across regions.)
Features and capabilities
We record the capabilities we can confirm from a tool's official documentation — such as whether it offers API access, whether it's open source, and whether it's enterprise-ready.
Security and compliance
We record a tool's published compliance status: GDPR compliance, SOC 2 certification, and ISO 27001 certification. GDPR is a vendor-stated position; SOC 2 and ISO 27001 mean the vendor holds that certificate — which may only cover its business or enterprise plans, not every tier. Check the vendor's own trust page or privacy policy for the detail on your plan.
Data handling and training policies
We record whether a tool trains on user data, based on the vendor's published privacy policy and terms of service. This is a vendor-stated position; check the vendor's own trust page or privacy policy for the detail on your plan.
Transparency and trust signals
We record whether multi-factor authentication (MFA) is supported.
A note on currency
We show each tool's price in the currency we recorded it in — usually GBP, as a UK visitor would see it, since that's what most of these vendors display here. Some vendors publish in US dollars instead, so the catalogue mixes currencies. Where it does, we don't rank tools on price directly: treating £20 and $20 as the same would mislead (£20 is roughly $25). Any figure we convert to GBP for comparison is shown as a clearly labelled estimate (≈£), from a recent exchange rate — never presented as the vendor's own price. Prices also often include VAT in the UK and exclude sales tax in the US, so even a converted figure is an approximation, not a like-for-like total. We re-check pricing on a rolling basis; each tool's “last verified” date shows how fresh its figures are.
What scores don't do
- Reflect hands-on testing of the product
- Account for personal workflow preferences
- Capture every edge case or niche use
- Replace reading the full tool profile before deciding
Affiliate relationships have no effect on scores. A tool we earn a commission from is rated exactly the same way as one we earn nothing from. The criteria are fixed; the published data shapes the number.
When scores are updated
We review and update a tool's score when its underlying data changes — a pricing update, a new compliance certification, or a change to the tool's data handling policy. Every tool shows a “last verified” date. If the data is more than 30 days old, we flag it so you can check before making a decision.