How we review.
Every published review receives editorial evaluation by Carlos Gil against the same five-part framework. Each page states the evidence behind its assessment; hands-on reviewed is reserved for documented product use.
Every page says how it was assessed.
Every published review receives editorial evaluation. Each carries one of three evidence statuses near the headline. Hands-on reviewed is reserved for documented product use; editorially reviewed and research preview describe the other kinds of evaluation.
- Editorially reviewed
- A substantive editorial evaluation using verified product information, official documentation, pricing, features and published user evidence, informed by Carlos Gil's 15+ years in marketing. Demos and screenshots may support the evaluation where available, but without enough documented product use we do not label it hands-on.
- Hands-on reviewed
- Carlos has used the product directly and genuine testing evidence is documented on the page. Testing notes, workflows, test dates and original screenshots appear only when the stored evidence supports them. This status is never inferred from a score, word count or commercial relationship.
- Research preview
- Initial editorial coverage awaiting deeper evaluation. Any score and verdict are provisional, and the page clearly identifies this status.
Five things we score.
- 01Features that matter
We judge the jobs creators actually hire the tool for, not the feature matrix on the pricing page.
- 02Ease of use
How quickly a creator reaches a usable output, and then a repeatable workflow. Tools that only make sense after a course get marked down.
- 03Pricing accuracy
We read the live pricing page and the plan limits, and check what the plan actually unlocks. Hidden seat minimums and export caps count against a score.
- 04Support and documentation
Published support channels, response commitments, documentation depth, and the pattern of complaints creators report publicly.
- 05Integrations
Exports, APIs, and the handoffs into the rest of a creator stack — editing, publishing, email, analytics.
The 1–10 scale, defined.
- 9.0 – 10
- Category leader on our framework, with no trade-off we consider disqualifying.
- 8.0 – 8.9
- Excellent with a known trade-off we name in the review.
- 7.0 – 7.9
- Solid for a specific creator; wrong for others.
- 5.0 – 6.9
- Works, but something is broken, overpriced, or badly supported.
- Below 5
- We would not recommend it, and we say why.
SCORES ARE RELATIVE TO THE OTHER TOOLS WE COVER IN THE SAME CATEGORY — NOT TO SOFTWARE IN GENERAL.
An editorial judgement, labelled as one.
Scores are editorial judgements, not lab measurements. They are relative to the other tools we cover in the same category, and a page labelled Editorially reviewed carries an editorial score rather than a tested one — the label sits directly above the number so there is no ambiguity.
We don't publish product screenshots we didn't capture ourselves, and we won't pass off vendor marketing images as evidence. When you see a claim on this site, the review status above it tells you exactly what kind of evidence supports it.
What we refuse.
Our editor discloses any paid relationship with a vendor and will not score a tool he has a commercial interest in beyond affiliate links, which are labelled everywhere they appear. Vendors never see a review before publication, and no vendor can edit, delay, or pull one.
How we make money.
Some links are affiliate links, labelled on every page that uses them. Affiliate and sponsor relationships never affect review status, score, ranking or verdict. Commission rates are never a scoring input, we keep negative reviews of tools we earn from, and nothing in a roundup is bought.
Reviews get revisited, not archived.
Reviews are revisited rather than archived — when a tool changes its pricing, ships a major release, or breaks something, the page and its last-updated date change with it.