Orange LuxembourgZero discrepancy trade-ins with Vision AI
Orange Luxembourg replaced manual device grading across every retail store with Pandas Vision AI on standard iPads — and hit a zero discrepancy rate with downstream buyers.
The Challenge
Orange's retail teams were grading trade-in devices by hand. Two employees in different stores could grade the same device differently, and downstream buyers regularly disputed those grades — applying penalties that ate into program economics.
Beyond the financial cost, manual grading was slow. Each handover took up retail floor time and created friction at exactly the moment a customer was deciding whether to upgrade.
The Solution
Orange deployed Pandas Vision AI across every Luxembourg store. The system runs on a standard iPad — no specialized hardware — and assesses a device's condition from a camera scan in seconds.
Rollout took less than a week. Each store needed only an iPad and a short staff briefing; grading consistency was instant from day one.
The Outcome
The discrepancy rate with Orange's downstream buyer (CompaRecycle) dropped to zero. Every device Pandas graded was accepted at the assessed grade — no penalties, no disputes.
Because grading is now objective, Orange can confidently offer customers higher trade-in prices that reflect actual device condition. Retail staff are freed from a task that wasn't a good use of their time.
“The grading accuracy and automation we've introduced have been key in driving trade-in performance.”Mustapha Rahem, Head of B2C, Orange Luxembourg
