The situation
One business, seven storefronts, a dedicated warehouse and roughly £2 million in annual sales — run end to end by eleven people. That ratio was not achieved by working harder. It was achieved by software. Signal North's founder co-founded the footwear e-commerce business in 2010 and spent the next fifteen years as its lead developer and system architect, growing a single website selling flip-flops into a multi-site operation on a proprietary operating system — one platform connecting purchasing, stock, cash flow, fulfilment, pricing, marketing and customer service, so that every commercial decision could see the whole business at once.
Why the existing approach was limiting
Generic back-office tools support individual tasks, not the connected decisions a trading business makes every day. Off-the-shelf repricers tracked a fixed list of competitors and defaulted to undercutting them, blind to stock position or restock dates — sometimes the right move is a higher price. Standard forecasting models collapsed on products selling one unit a week, where a run of 0, 1, 0, 0, 2 defeats anything built for steady volumes. And in the warehouse, orders were picked in the sequence they were placed, so every afternoon became a race against carrier deadlines: it took two full-time staff plus up to six seasonal hires to keep up, and nearly 6% of orders went out with a picking or packing error.
Signal North's role
Signal North's founder designed and built every system the business ran on — storefronts to warehouse floor — while carrying responsibility for the commercial consequences. The platform was never a project delivered and handed over; it evolved for fifteen years around real constraints, with the person writing the pricing logic also watching the margin it produced. That feedback loop — build, trade, measure, refine — is what made the tools fit the business exactly.
Key decisions and intervention
- reprice thousands of products daily from a live, market-wide search rather than a fixed competitor list, with rules weighing stock level, sales trend, incoming deliveries and season before any price moved;
- generate 24-month sales forecasts from two years of history, stable even on one-unit-a-week products, to steer purchase orders, stock-holding and cash flow — introduced into commercial use in 2023;
- re-sequence the warehouse around dispatch deadlines instead of order timestamps: batch picking sorted by location, optimised trolley routes and shelf-per-order packing;
- automate returns end to end: self-service requests with QR-code drop-off, automatic approval of straightforward cases, and barcode scan-in that surfaces the expected contents instantly for faster refunds and restocking;
- allocate warehouse bins by forecast, not guesswork — each product's twelve-month peak-volume profile encoded as a single sortable number, so goods-in gets a real-time answer to "which bin?" from one indexed query;
- rebuild size and colour selection around regional sizing formats and live stock visibility, on mobile and desktop;
- run £100k+ of annual advertising through direct Google Ads API integration, alongside email marketing and CRM workflows;
- favour self-administering rules over constant developer intervention, everywhere.
What changed
The warehouse went from two full-time staff plus six seasonal hires to three full-time employees year-round, and the error rate fell from nearly 6% to 0.02% — fewer than one order in 3,500. In the three months after the size-and-colour selector was rebuilt, add-to-basket actions rose 18% against the prior year's average. Repricing that would have been a full-time manual job ran automatically every day, holding margin when stock was scarce and clearing slow lines before they became dead stock. Purchasing was steered by forecasts rather than instinct — less cash tied up in stock, fewer stockouts through seasonal peaks.
None of these gains came from a single clever system. They came from the systems being connected. Forecasts steered purchasing; purchasing shaped where stock was stored; the stock position fed the pricing rules; pricing drove the sales the warehouse then fulfilled almost without error. Each part made the others better, and because the platform favoured self-administering rules over developer intervention, the business ran it — not a technical team. That is how eleven people traded at £2 million across seven websites: the operating system did the coordination that would otherwise have needed a layer of managers, spreadsheets and second-guessing.
This is the capability Signal North offers. The platform was built by someone who carried the commercial consequences of every design decision — where efficiency was measured in wages saved, forward thinking in stock that arrived on time, and growth in orders that went out the door correctly. Systems built from that position are intuitive because they follow how the business actually trades, and durable because they were designed to evolve with it. That is the difference between software that supports a business and software that runs one — and it is the standard Signal North builds to.