The series

Benchmarks, numbered like they matter.

Every Boundstone benchmark is a numbered public event: announced before data exists, frozen before results, disputed by vendors before publication. Cherry-picking dies in that order. Read the full methodology →

№ 001

US phone validation

1Lookup, Trestle, Twilio Lookup and NumVerify head-to-head — ~1,000-number owned dataset; validity, line type and carrier scored; abstain ≠ wrong. Boundstone is self-excluded from № 001 and is scored under the identical protocol from the first re-run.

Frozen · collection next
Announced · 2026-07-14 Frozen · 2026-07-25 — tag methodology-v1.0 Collect — all vendors, same day Dispute — vendors get 14 days Publish — win or lose
METHODOLOGY V1.0 · FROZEN 2026-07-25 · SHA-256
24fb254ed62b0aef493520e60d29652234405cb7e77a7dcedfa1f636ee34969d
Tag methodology-v1.0 in the public repository. Verify it yourself: shasum -a 256 METHODOLOGY.md — if the hash above ever stops matching, we cheated.
№ 002 — International HLR liveness

Per-country mobile liveness via HLR — queued until № 001 publishes. Numbered now so nobody can accuse us of inventing the series after a win.

№ 003 — IP intelligence

Proxy/VPN detection and geo accuracy — queued until № 001 publishes.

The ritual

  1. AnnounceThe run gets a number and a scope — in public, before any data exists.
  2. FreezeThe methodology repo is tagged and its SHA-256 posted. Scoring rules can't move after this.
  3. CollectEvery vendor gets the same dataset on the same day, fully logged.
  4. DisputeVendors see their raw disagreement rows and get 14 days before anything publishes.
  5. PublishResults ship win or lose. Abstain ≠ wrong. The chart space is already reserved.

Be there when it drops

The API is live — start free without waiting. This list gets benchmark № 001 the day it publishes. One email. We validate emails for a living — we're not going to spam yours.