Skip to main content
All customers
Würth·Logistics and distribution·Europe·End-to-end AI delivery

Voice AI cut order entry to seconds and reduced order input errors by about 9% for field reps.

Order entry time
Seconds
Order input errors
~-9%
Driver safety
Restored
Würth illustration
At a glance

What we shipped

A voice assistant that listens to live sales calls, extracts structured order details, and writes them to the order management system.

Challenge

Field sales teams in industrial distribution, wholesale, and manufacturing spend much of their time traveling between client locations. While driving, reps often take phone orders.

Approach

Blueprint → AI Pilot → Production launch → Scale and operate.

We followed the Datablooz Delivery Model. See our process.

  1. Blueprint

    Mapped field sales call flows, product catalog integration, and SKU matching logic with stakeholders.

  2. AI Pilot

    Built speech recognition and NLU tuned to the product catalog, validated on real call audio.

  3. Production launch

    Shipped the voice assistant integrated into the order management system with voice confirmation flows.

  4. Scale and operate

    Extended to sales coaching, CRM updates, upsell recommendations, and voice inventory checks.

Outcomes

Business, technical, and governance outcomes.

  • Order entry compressed to seconds.
  • About 9% fewer order input errors.
  • Safer field operation for driving reps.
  • Reps focus fully on customer conversation.
Architecture and stack
  • Python
  • Whisper
  • LLMs
  • FastAPI
  • PostgreSQL
  • Telephony APIs
Governance

Voice confirmation loop per order with transcription logs and SKU-matched audit trails into the order system.

Working on something similar?

Schedule a call. We will tell you honestly whether AI is the right move.

Reference calls available under NDA after the second working session.