> how_we_check
Every claim says who made it.
A trust card is only useful if you know which parts were verified. So every line carries one of three labels, and a date.
✓ checked
We saw evidence ourselves: the code or manifest, the privacy policy, a public register.
~ maker says
The maker told us and we couldn’t verify it yet. Treat it as a claim; ask for proof if it matters.
○ unclaimed
We wrote this card from public information before the maker joined. Makers can claim and correct it.
The check, step by step
- 01Maker fills in the cardWith links to evidence: privacy policy, store listing, security page, certificate numbers.
- 02We check what we canPublic registers (ICO, Cyber Essentials, ISO bodies), policies, the code or manifest, and network traffic where we can test it.
- 03Each line is labelledChecked or maker says, with the date. Nothing is hidden because it looks bad.
- 04Re-checkedOn every new version we hear about, and at least every 90 days. Changes go to everyone watching.
- 05Anyone can reportSpot something out of date? Use “report something out of date” on the card. We look within 2 working days.
The rules we don’t bend
→
Money never changes a trust card or a ranking.
Featured launches pay for timing and newsletter placement only. They are always labelled “featured”.
→
One person, one vote, from a verified work email.
Anyone can sign in, but only work email addresses (not Gmail, Outlook.com and similar) count towards rankings. Makers can’t vote for their own tool.
→
Rankings weigh buyers, not followers.
A tool’s score is: useful votes + 2 × shortlists + 2 × watches, all from verified work emails. That’s the whole formula.
→
Free launches go out within four weeks.
No queue measured in years. If a week is full, you get the next one with space, never more than four weeks out.
→
Buyers decide what makers see.
Makers see a buyer’s role and sector only if the buyer allows it (on by default, change it in your account), and name and email only if the buyer ticks “open to contact”.