Sweethearting Detection
Discount and void patterns suggest sweethearting. Correlates operator and lane.
Employee-discount, manual-override or skip-scan indicators concentrate on an operator well outside the peer distribution, and repeat across shifts.
Review case opened in LP analytics with the signal set and peer comparison attached; observation scheduled.
Every write is gated on an approver role you name. Nothing runs unattended.
- Combine the weak signals, since no single one is evidence: scan gaps, discount frequency, and basket value against lane norms.
- Test whether the pattern follows the operator across lanes and shifts, which separates behaviour from equipment.
- Check the mundane explanations: a faulty scanner, a produce code everyone gets wrong, an untrained new starter.
- Route to LP as a review with the statistical basis stated plainly, so a human judges it rather than a threshold.
- Close when the pattern resolves, whether through coaching, a repair, or a substantiated case.
Signal concentration against the peer distribution over 45 days, and recovered value where substantiated.
The case closes on this number, not on the action being taken. A playbook without a close condition is a dashboard.
The rest of Loss Prevention
Run Sweethearting Detection on your data.
Pick three playbooks from the catalog. We wire them against your system of record for the pilot.
御社のデータに何が隠れているか、確認する。
お客様のオペレーションについてお聞かせください。デモまたはPoCをご提案します。