Editorial Policy

1. What gets listed

We list genuine AI tools and AI-powered products across our categories. Every submission is reviewed by a person for category fit, working links and quality before publishing.

Every published listing carries a stated downside. At least one con is required before a listing can go live — enforced in the database, not by habit — because a listing with no downside is a press release, and telling a reader when a tool is wrong for them is the most useful thing we can do for them.

We reject anything we would rate below 2.0 out of 5 rather than listing it and marking it down.

2. How tools are ranked

Ordered lists use one score: 40% the editorial rating one of our editors gave the tool, 25% how complete the listing is, 20% how recently the product shipped or we re-verified it, and 15% how often readers who see the listing click through to the vendor. A listing claimed and confirmed by its vendor within the last twelve months is multiplied by 1.05.

Payment is not a term in that formula and there is no code path from what a vendor paid to where they appear. Paid placements — Spotlight, category-top and sponsored cards — are labelled every time they appear and sit outside the ranking rather than inside it. Paying speeds up your review; it does not move you.

The complete formula, the point values behind the completeness score, the freshness curve and the tie-break order are all published at /how-we-rank.

3. Independence

Affiliate links and paid placements never influence whether a tool is listed or how it is rated. Editorial decisions are made independently of commercial relationships.

4. Accuracy & updates

AI tools change quickly. We update or remove listings that become inaccurate, and we encourage owners and users to flag corrections.

5. Corrections

Spotted something wrong? Use the contact page or the report link on any listing and we'll review it.