Vehicle Damage Bot

Vehicle Damage BotGathering information about a vehicle's condition was previously a slow, manual, paper-heavy process. Information would be captured on a form using a "splat", images taken and merged manually later on in the process.

This often led to different language being used (and potentially ambiguous data), slow update times and the manual synchronisation of information between photos and written information.

This essential data capture process previouly took approximately 40 minutes to complete for each vehicle. Using a combination of Google Speech Recognition technology and a Conversational AI bot from, this was reduced to less than 1 minute with many additional benefits for our customer:

Easy to Use - The bot guides the user through a simple set of questions. The bot helps wherever possible with contextual options and guides the user through to completion

Multi-modal - Free-form text entry, speech and touch interfaces allow users to enter information even within noisy environments

Disambiguation - Not specific enough in your answer? The bot's AI would suggest answers based on your previous inputs

Domain language and normalisation - The bot would help with domain-specific terminology and disambiguate e.g. "passenger door" - "Did you mean nearside front door or nearside rear door?"

Real-time backend updates - Once all information was captured, backend systems were updated immediately. Slow, cumbersome, paper-heavy processes became paperless overnight