Prices are set once and left to drift away from demand.
→ Sets every price to the margin your demand will bear — by store zone and channel — and never starts a price war it can't win.
+3–6% gross margin
AUTONOMOUS OPTIMIZATION · THE RETAIL BACKBONE
Which price to set, which SKU to range, which order to place, which store to staff — decided continuously, prescriptively, on your own numbers. The planning spreadsheet goes out the window.
Prices are set once and left to drift away from demand.
→ Sets every price to the margin your demand will bear — by store zone and channel — and never starts a price war it can't win.
+3–6% gross margin
Stock is everywhere except where it sells.
→ Forecasts, reorders, and transfers SKU-by-node so the right stock sits in the right place at the lowest cash cost.
+15–30% inventory turns
Cash is trapped in slow-moving inventory across the network.
→ Re-balances inventory against working-capital targets and tunes safety stock per SKU-store instead of flat policies.
−8–20% working capital
Shelves go empty while service quietly bleeds away.
→ Trades service against cost continuously and moves stock before you ever go out of stock.
+5–15 pts service
RETAIL NETWORK — ALL STORES
Reprice 412 SKUs, 1 markdown opened
elasticity + competitor move, margin floor held
Place 38 POs, 6 store transfers
POS + seasonality, stops 9 local stockouts
Shift 84 hours to peak dayparts
forecast traffic, overtime cut, coverage intact
Seven closers. One coherent retail plan. Decided on your numbers.
Up and running in one day.
HOW IT WORKS
On a schedule when the day starts. And the moment something changes — a competitor cuts, the radar shifts, a top-seller runs low.
Sets every price and markdown to capture the most margin demand will bear, store zone by channel.
→ +3–6% margin
Decides what earns its place on the shelf and what makes room for something better.
→ +sales / sq ft
Builds one demand forecast, then reorders and transfers stock before a stockout happens.
→ +15–30% turns
Staffs each store to its traffic, holds the service bar, and kills the overtime.
→ −5–10% labor
Picks the fulfillment node per order at the lowest cost to serve — without starving the stores.
→ +omni margin
The decision
What Opti reads
What you get
Everyday price, markdown depth/timing, and promo mechanics per SKU — net of cannibalization and forward-buy — and the match-or-hold call within guardrails.
The decision
What Opti reads
What you get
List/delist per store and category, must-carry vs optional ranging by cluster, and shelf facings reallocated toward contribution-per-foot.
The decision
What Opti reads
What you get
One operational forecast, per-SKU-node PO create/expedite/cancel, store-to-store transfers, and safety stock tuned to hold service at minimum cash.
The decision
What Opti reads
What you get
Labor shifted to match traffic (overtime cut, coverage intact), flagged underperforming stores, and the lowest-cost fulfillment node per order — store, DC, or pickup.
You stay in control. Everything starts as a suggestion you approve. Turn on auto-pilot only when you trust the math.
You set the guardrails once — margin floors and ceilings, competitor match rules, working-capital targets, service SLAs, labor budgets and fairness rules — and every closer obeys them, every time. Decisions stay recommend-only until you flip them.
UP AND RUNNING IN ONE DAY
Open the sandbox — it's loaded with a realistic multi-store retail network dataset.
Watch the closers decide: pricing, merchandising, replenishment, allocation, labor, and omni fulfillment.
Approve a day's shelf plan and see the margin math and the guardrails behind every move.
+3–6%
gross margin
+15–30%
inventory turns
−8–20%
working capital
The sandbox runs on a sample multi-store retail network — no card, no commitment. Bring your own numbers when you're ready.