Core Definition
POS (Point of Sale) data includes only records of transactions that were actually completed. Customers who visited but did not buy, products picked up and put back down, and losses from shelves that sat empty and unsold are never recorded in POS. This “invisible data” is the key to store improvement.
What POS cannot tell you
**Abandoned checkouts**: A customer who gives up on paying because the checkout line is too long leaves no trace in POS. Optimizing checkout operations requires knowing how often this happens and what causes it.
**Products picked up and put back**: The decision to skip a purchase after seeing the price or size is never recorded in POS. Yet this is the most direct signal for pricing adjustments and package improvements.
**Stockout losses**: A product may sit in inventory but never reach the shelf, or it may genuinely be out of stock and create a missed sale. POS cannot tell these two cases apart.
Filling POS gaps with AI data
CCTV AI analysis connects the “result” in POS data with the “process” happening inside the store. If sales were low on a given day, POS can only tell you that it happened. AI data can tell whether visitor numbers were low, whether visitors were high but conversion rate was low, or whether stockouts were frequent.
This context is what makes it possible to set the right direction for improvement. If visitor numbers were low, strengthen marketing; if conversion rate was low, change the display; if stockouts were frequent, review the ordering system.
How SAAI Uses It
saai insight connects POS sales data with CCTV AI analysis data in a single dashboard. Operators can check “sales relative to visitor count” in real time and trace the cause when conversion rate drops. The point is to find and address the causes of sales loss that POS alone could not show.
See it in action