Editorial policy

Scope

This policy covers every page on this domain. It describes how the ranking is built, who is accountable for it, and what happens when it is wrong.

Evidence standard

Every scored claim must trace to a source a reader can open. Sources are recorded in the source ledger with the date checked, because review scores and vendor claims move. Where a claim cannot be verified from public material, it is either excluded or explicitly marked as unverified rather than scored on assumption.

We distinguish between what public evidence establishes and what it does not. A Clutch aggregate establishes a review record; it does not establish SOC 2, ISO 27001, insurance coverage, or the seniority of any individual engineer. Those distinctions are stated inline rather than smoothed over.

Independence

Placement is determined by the published scoring model. No vendor can purchase inclusion, position, or removal.

This publication is created and published by Valerii Petrianyk in a personal capacity, independently of every vendor covered here. His employer, Linvelo, is a software development consultancy; it does not appear in the ranking, was never evaluated for it, and has no role in producing this site. The full disclosure is on the author page.

Because the author works in the same industry he writes about, the safeguard here is method rather than assertion. Every score traces to a dated public source in the source ledger, which means the ranking can be rechecked by anyone without trusting the person who wrote it. That is the point of publishing the model and the ledger at all.

Revisions

The ranking is reviewed at least every twelve months, and sooner when a vendor’s public evidence changes materially — an acquisition, a change in delivery model, or a significant shift in its review record. Revisions are substantive: a review that changes nothing but the date is not recorded as an update.

Each page carries a visible publication date and last-updated date, matching the machine-readable dates in its structured data.

Corrections

Factual errors are corrected as soon as they are confirmed. Corrections that change a score, a rank, or a material claim are noted with the date and what changed. Typographical fixes are made silently. Reports go to the contact page.

Use of AI

This site is produced with AI assistance. Language models are used to draft and structure page copy and to format the comparison tables. They do not determine placement: scores follow the published scoring model applied to the public sources recorded in the source ledger.

The author is accountable for everything published here, however a draft was produced. A claim that cannot be traced to a listed source does not belong on this site — report it through the contact form and it will be corrected under the process above.

Authorship

Every page carries a named author who is accountable for its claims, with a linked profile and a verifiable external professional presence. We do not publish under “staff”, “admin”, or an unnamed editorial team.