Scattered inventory data turned into usable information
“Paper invoices become data without anyone keying them in.”
Problem
Much of the inventory and purchase record existed only as paper invoices, handwritten notes and PDFs, and online store data sat apart from the physical stores, so no one could see in real time what was actually available to sell. Items showing zero stock had their orders auto-cancelled, so inaccurate inventory data translated directly into lost sales.
- Unstructured purchase records — paper, handwriting and PDFs only
- Split channels — online and in-store inventory managed separately
- Experience-dependent — ordering and pricing judged from memory
What BeSir did
Manage online and offline inventory accurately as one, and improve ordering, pricing and store operations from sales and inventory data.
- Reads unstructured purchase records — paper, PDF and images — and digitizes them
- Designed and built an integrated database in the Australian region, synced in real time with the existing POS (KPOS) and inventory data
- Unifies ordering, inventory, pricing and sales into one business ontology, resolving online and offline stock into a single sellable quantity
- Synchronizes data across heterogeneous systems in real time, closing the data silos
- Encodes discount and pricing rules into the system, and provides sell-through analysis by store and by product
Results
Once inventory could be trusted, 17,340 products were listed online. Purchase automation is running across 4 stores and 596 suppliers, and is expanding store by store.
Verified confirmed in the operating environment · PoC confirmed within PoC scope · Target a target set on work in progress · Estimate calculated from the stated assumptions