Legal review

DoNotPay · IBM, Facebook

Independently reportedProfessional servicesUnited KingdomScaledNLP

Independently reported. An independent or peer-reviewed source reports this result, alongside the vendor.

Reported by bestpractice.ai (independent reporting). Checked against the source page on 2026-09-06. 1 of 3 figures below appear on that page word for word; the rest are marked.

The problem

Automating the appeal of parking tickets and providing legal advice.

What was deployed

Vendor
IBM, Facebook
Products
IBM Watson, Facebook Messenger
Technique
NLP
Build or buy
Bought and customised
Deployment
API
Data used
user inputs (choice of defence from a list of 12 options)
Scale
more than 86,000 people have used it

What changed

Each row is quoted from the source. Figures we could not find on the page in those words are marked — they are kept, not deleted, so you can judge them.

£2 millionValue of parking tickets appealed

Page title: 'DoNotPay has succesfully appealed £2 million in parking tickets...' (not restated in body text)

Not found on the source page in these words

40%Claim success rate

Nearly 40 per cent of them were successful, according to a poll of the site's users.

Not found on the source page in these words

more than 86,000Number of usersin its first four months

more than 86,000 people have used it to appeal against council parking fines.

Quoted word for word from the source

Difficulties and limits

Sources

Similar deployments

Quora · Ivo

2 days — NDA approval turnaround time

Vendor-reportedTechnology & software · Legal review · NLP · In production

Terry Irwin

15 hours — Hours recovered per week

Vendor-reportedProfessional services · Process automation · AI agents · In production · 2026

Impact · Microsoft

$1.72 million — Annual net ROI

Vendor-reportedProfessional services · Process automation · LLM · In production