Editorial standards
How we review AI tools
How SynaBot discovers, evaluates and verifies AI tools: sourcing, evaluation criteria, when hands-on testing happens, how pricing is checked, update frequency and affiliate disclosure.
How do we find the AI tools we list?
Tools reach the directory three ways: reader and user submissions, monitoring of launch channels and vendor changelogs, and gaps we spot while building assistants for SynaBot customers. Anything that cannot be signed up for publicly, or that duplicates an existing listing without adding something, does not get a page.
How is a tool evaluated?
Every listing starts as a documented profile: what the tool does, who it suits, what it costs, the model or technology behind it, and where it falls short. That profile is compiled from the vendor's own documentation, pricing pages and product tour — not from other directories or listicles. An editor then checks the claims are internally consistent and that the tool actually solves the job the category implies.
When does hands-on testing happen?
We do not claim to have personally tested every tool in a directory of this size, and we never let a template imply that we have. Hands-on testing is prioritized for tools people are likely to pay for, tools we recommend at the top of a category, and tools where vendor claims are hard to verify on paper. Where a tool has been used hands-on, the tool page says hands-on tested explicitly and the write-up contains our own observations. Where it has not, the page says the listing was compiled from the vendor's public information. Only genuinely tested tools carry a rating or review markup.
How is pricing checked?
Pricing is taken from the vendor's public pricing page and recorded with the date it was checked. Prices change often, so the figure on a SynaBot page is a snapshot, not a quote — always confirm on the vendor's site before buying. If you spot a stale price, use the report link on the tool page and it goes straight to the review queue.
How often are listings updated?
Listings are re-checked when a vendor ships a significant change, when a user reports a problem, and on a rolling schedule otherwise. Each tool page shows its last-verified date so you can judge how fresh the information is rather than having to trust it blindly.
Does SynaBot make money from these reviews?
SynaBot's revenue comes from its own subscriptions and consultancy, not from directory placements. We do not sell rankings, sponsored positions or "featured" slots. Where an affiliate relationship exists it is disclosed on the tool page and has no effect on whether a tool is listed, how it is described, or where it appears in a category.
How do I report a mistake?
Use the report or claim-listing link on any tool page, or email hello@synabot.ai. Corrections from vendors and users are treated the same way: we check the primary source, update the page and move the last-verified date.
How is the integrity score calculated?
The Integrity Analysis on a tool page is a probity check of the business behind the tool, kept deliberately separate from any judgment of product quality. Five categories are each scored from 0 to 10 against what can be verified publicly:
- Business Transparency — can the operating legal entity, its registration and its registered office be identified?
- Legal Documentation — are Terms of Service, a Privacy Policy and related documents published, complete and dated?
- Consumer Clarity — are pricing, cancellation, refunds and the product’s stated limits set out plainly?
- Corporate Governance — is there registry evidence, a stated jurisdiction and a named accountable entity?
- Trust & Confidence — are there working contact and complaint channels, and are AI limitations disclosed?
The overall figure is a simple, unweighted average of the five category scores, multiplied by ten: overall = round(((t + l + c + g + tr) / 5) × 10). No category carries more weight than another, and no score is adjusted for popularity, spend or affiliate status. A tool is only given an overall score when all five categories have been assessed. Bands: GREEN 75–100, AMBER 50–74, RED 0–49.
Findings record presence or absence of evidence — for example “no Terms of Service found at the vendor’s domain at time of review” — and never a judgment about the character or intentions of the people involved. Sources are limited to the operator’s own pages, corporate registry records and named review platforms; search-engine result pages are not accepted as citations. Every analysis is drafted with AI research assistance, then reviewed and approved by a human editor before it is published, and is automatically re-checked every 90 days. It is an opinion based on published information at the date shown, not a certification.