Buying at the right moment, not out of habit

Sector Consumer product brand, own-manufactured range sold by subscription and retail, perishable stock and volatile raw material costs.

The situation

Buying ran on gut and fixed reorder points. Cash sat trapped in stock that moved slowly, hero lines still ran out, and product was written off when it passed its shelf life. Raw material prices swung, but purchasing never moved with them. All the data sat in NetSuite, and none of it turned into a buying decision.

What CAPIO did

The assessment put procurement and inventory at the top of the list, high value and high feasibility. We built a predictive supply chain agent that reads the business end to end. It pulls inventory, purchase orders, sales orders and bills of material from NetSuite, layers in subscription numbers and order volumes as the real demand signal, and adds supplier lead times, delivery windows, shelf life and live raw material price feeds. From that it forecasts demand, then recommends what to buy, how much and when, timing purchases to favourable price windows while respecting shelf life and storage limits. The agent proposes, a buyer approves, so a human stays on every purchase. Demand and price models orchestrated with n8n, written straight back into NetSuite, with a full log of every recommendation and who approved it.

The outcome

  • Operational cost down by twenty three percent, across stock holding, waste and buying price *representative*
  • Fewer stockouts on the lines that matter most, and far less written off to expiry *representative*
  • Working capital freed from slow-moving stock, and buying shifted to better price windows *representative*
  • Buyers spending their time on supplier relationships rather than spreadsheets *representative*
It does not just tell us what to buy. It tells us when, and it has been right often enough that we trust it.
Illustrative, operations director

Book a call

Case studies are anonymised. Representative outcomes are benchmarked to independent research and illustrate typical results rather than a specific client figure.