Everything you'd want to know before you install.
A detailed look at how AI Returns Predictor works, why we built it the way we did, and the thinking behind the features above.
Why anticipate returns?
Product returns weigh heavily on margins: round-trip shipping, restocking, unsellable items. Most shops only spot a risky return once the parcel has already left. AI Returns Predictor moves that detection upstream, to order validation, giving you time to act.
An explainable score, not a black box
The engine combines six well-known e-commerce signals: the customer's historical return rate, bracketing (the same product ordered in several sizes), basket value relative to the average basket, the presence of products in high-return categories, new-customer status and basket size. Each factor shows its point contribution, right on the order page.
A logistics alert at the right moment
As soon as an order crosses your high-risk threshold, a summary email is sent to the logistics address of your choice: order reference, score and contributing factors. Your team can verify the address, confirm the size or contact the customer before shipping.
AI as reinforcement, never an obstacle
Enable the Mistral layer to refine the heuristic score and get a one-sentence explanation. On any network error or API outage, the module silently falls back to the heuristic score: scoring never blocks order preparation.
There are no reviews yet.