Techbanq
Retail & E-commerce

QuickShelf Quick-Commerce Demand Forecasting

A quick-commerce operator replaces a weekly forecast cycle with hourly per-dark-store demand prediction — cutting stockouts and waste in the same operating period.

Client
Quick-commerce operator, 64 dark stores across three cities
Category
Retail & E-commerce
Engagement
6-month engagement
Services delivered
Data & AI systems, AI strategy & implementation, Digital transformation
The Challenge

The challenge

The CFO had identified a structural commercial problem: the operator was simultaneously over-ordering to protect the 15-minute promise on perishables and writing off the resulting expired stock at month-end. The weekly forecast cycle that drove replenishment decisions was the root cause — it was structurally incapable of responding to the hour-by-hour demand variation that a fast-fulfilment perishables business operates on. Both numbers needed to move together, not traded against each other.

What we found

Discovery

Six weeks of dark-store observation identified the specific inputs the existing forecast was missing. Demand varied hour-by-hour in response to signals the forecast cycle could not incorporate: weather patterns, hyper-local events, and marketing pushes the growth team had launched that same morning. The forecast model itself was well-constructed; it was responding to the wrong inputs — historical patterns rather than the real-time signals that actually drove demand.

How we did it

Our approach

1

Forecast model redesign: built an hourly per-SKU per-store demand forecast model ingesting weather, local events, marketing actions, and the operator's promotional calendar

2

Replenishment programme redesign: replaced weekly replenishment runs with a continuous reorder programme respecting supplier minimum order quantities and lead times

3

Shared operating view: built a waste-and-stockout console at city operations level — the same data the CFO and the city operations manager see and act on

4

Continuous improvement programme: established a daily backtest quantifying forecast error per SKU and feeding the worst-performing categories into the next training cycle

What we built

The Outcome

The engagement delivered an hourly forecast model coupled with a continuous replenishment programme and a single waste-and-stockout view shared across operations and finance. The operator stopped managing stockouts and waste as competing objectives and started moving both together.

Measured Impact

The results

-44%
Reduction in stockouts on perishable categories
-29%
Reduction in monthly write-off of expired stock
+18%
Improvement in 15-minute promise reliability
64
Dark stores live on hourly forecasting
“We had been trading stockouts against waste for three years. The hourly forecast was the first thing that let us move both numbers at once.”
C
Chief Operating Officer
Quick-Commerce Operator
Technologies & PlatformsHourly per-SKU per-store demand forecast modelContinuous reorder loop with MOQ and lead-time constraintsOperations-and-finance shared waste-and-stockout console
Next case studyVitalCare Clinical Operations IntelligenceA multi-site hospital network builds a clinical operations layer that gives bed managers, theatre coordinators, and ED leads a live, shared view of patient flow — measurably cutting length of stay and ED boarding time.

Next step

Have a project like this?

We would love to hear what you are building. Let us explore how we can help.