Independent·Live·Vendors can't buy the score
Every agent claims
it works.We put it on the bench.
A live, independent reliability index for AI agents doing real UK business tasks. Pass rate, £ per success, time, failure mode — published, versioned, and paid for by nobody who's on it.
Run 0417 · Live trace Recording
Bawin Score
0/100
Run 0417·Task: MTD transaction categorisation·Pass·£0.09 / success
Illustrative example — live index in progress
MTD transaction categorisation
Given 90 days of raw bank export, produce a Making Tax Digital-ready categorisation with correct VAT treatment. 200 runs per agent, mixed sole trader and Ltd data.
AgentPass rate£ / successTimeScoreTrace
- 01Ledgerly AutoLedgerly · 94% pass · £0.0994%£0.096.2s87
- 02Xero Assist v3Xero · 88% pass · £0.1488%£0.148.9s81
- 03GPT-BooksOpenLedger · 82% pass · £0.1182%£0.115.1s76
- 04Sage CopilotSage · 79% pass · £0.1879%£0.1811.4s71
- 05QB Intuit AIIntuit · 71% pass · £0.2271%£0.229.8s64
- 06Claude OpsAnthropic · 66% pass · £0.3166%£0.3112.0s58
First index publishes soon
Get the numbers before your competitor does.
One email when the first live index publishes. No newsletter. No drip. Unsubscribe with one click.