Builder trust is scored from 0 to 1000 and ranked by standing against the whole builder community, not against fixed milestones.
A score built on fixed thresholds (“100 feedback items for the top tier”, say) is either unreachable while the community is small or meaningless once it grows. Percentiles adjust themselves: the top tier always describes who is leading right now.
The boundary for each tier is recomputed monthly across every builder and stored, so everyone scored in the same month is measured against the same yardstick.
Missing data costs nothing. If a builder has connected no external source, or a product has not been reviewed yet, that component is dropped from the calculation rather than counted as 0 — they are scored on what actually exists, and never penalised for something that was never collected.
| Component | Weight | How it is measured |
|---|---|---|
| Shipping frequency | 20% | Months with at least one changelog over the last 6. It counts MONTHS, not entries — ten changelogs in one month still counts as one month. |
| Recency | 10% | Time since the last changelog, decaying exponentially with a 90-day half-life. |
| Shipping quality | 5% | Changelogs per product, reaching full marks at 5 per product. |
| External data | 15% | Sources actively syncing (GitHub, Google Analytics, Polar). This is third-party data and the hardest to influence. Builders who have connected none are NOT penalised. |
| Traction | 10% | The average standing of weighted claps and product views. |
| Product quality | 12% | The product's automated review score (0-100). Products not yet reviewed are NOT penalised. |
| Profile completeness | 8% | Avatar, gallery, video, category and a description of real length. |
| Community contribution | 10% | Comments written and claps given to other people's products. |
| Commitment | 10% | How often the product dashboard is used, and how many products are actively run. |
On top of this: shipping user feedback adds up to 30 points, while neglected or removed products subtract from the score.
| Tier | Standing | Minimum evidence |
|---|---|---|
| Unikorn Legend | Top 5% | 5+ changelogs OR 1+ external data source |
| Serious Builder | Top 15% | 2+ changelogs |
| Builder | Top 30% | 1+ changelogs OR 1+ external data source |
| Maker | Top 55% | — |
| Rookie | Everyone else | — |
A high score alone is not enough: the top three tiers also require evidence of real shipping. That is not something a launch-and-leave product can satisfy — only continued updates can.
Tiers are settled at the start of each month across the whole community, so one can rise or fall: holding a tier asks for the same shipping that earned it. Scores still update daily, so you can always see how far the next tier is. Every surface shows the tier held now; the highest you have ever reached is recorded in the tier history on your profile.