A bakery that has grown past a certain size stops being a cooking operation and starts being a logistics operation that happens to involve ovens. You have production schedules, ingredient stock levels, supplier orders, delivery runs, customer orders, and shift rosters — all moving at once, all interdependent, all in someone's head or on a shared whiteboard that gets wiped every morning.
That was where this client was when they came to us. The business had grown from a small local operation to supplying restaurants, retailers and catering companies across Cape Town. The whiteboard had become a liability. Deliveries were going out on time only because one person held the entire schedule in their head — and that person could not take a day off without things falling apart.
"We needed someone to look at how we actually work — not just install something off the shelf that we'd have to change everything to fit."
Understanding the Operation First
Before writing a line of code, we spent time on site. A bakery operation has rhythms that are not obvious from the outside. Production starts before dawn. Ingredient stock is used and replenished daily, not weekly. Delivery windows are tight — a restaurant needs its bread before service starts, not at noon. And the relationship between what is ordered by customers, what is scheduled for production, and what ingredients are on hand is tighter than in most businesses.
The existing system, such as it was, consisted of:
- A physical whiteboard for the daily production schedule
- A shared WhatsApp group for delivery coordination between the driver and the manager
- A spreadsheet updated periodically with customer orders
- A paper stock book for ingredients, checked manually each morning
- Phone calls to suppliers when something ran low
None of these were wrong individually. The problem was that they did not talk to each other. Stock levels did not drive production planning. Production planning did not automatically update delivery schedules. A change anywhere required manual updates in three other places — and those updates only happened if someone remembered to do them.
What We Built
The result was a custom web-based operations manager — accessible from any device on the bakery's network, and from the owner's phone remotely. It runs on their own hardware. No monthly SaaS fees, no data going to third-party servers.
Production Scheduling
The core of the system is a daily production board. Each morning, the production schedule is generated based on confirmed customer orders for the day. The baker sees exactly what needs to be produced, in what quantities, and in what order — grouped by oven batch to minimise turnaround time. Changes to orders before the cutoff time automatically update the production board.
Ingredient Stock Management
Stock levels are tracked against each product's ingredient bill. When production runs are marked complete, the system deducts the quantities used. The manager sees current stock levels with colour-coded alerts for anything approaching reorder threshold. Supplier orders are logged directly in the system, which tracks expected arrival dates.
Customer Orders and Delivery Scheduling
Customer orders come in by phone and WhatsApp. The manager logs them in the system, which slots them into the delivery schedule based on area and time window. The driver sees their route for the day on their phone — no more WhatsApp message threads to scroll through. When a delivery is completed, the customer gets a WhatsApp confirmation automatically.
Staff Shifts
A simple shift roster links staff to production or delivery roles for each day. If a baker is off sick, the system flags which production runs need reassigning before the morning schedule is finalised.
The Stack
Chosen for reliability and ease of maintenance — not technical novelty.
- Backend: Node.js/Express on a small local server at the bakery. No cloud dependency for day-to-day operation.
- Database: SQLite — fast, portable, zero administration. A daily automated backup copies the database to cloud storage.
- Frontend: Vanilla JavaScript, optimised for desktop (management) and mobile (drivers, floor staff). Fast load on any device.
- Notifications: WhatsApp Business API. Customer delivery confirmations, low-stock alerts and daily production summaries all route through the same integration.
Hosting on-site was deliberate. A connectivity outage should not stop production. On-site means the system is always available on the local network regardless of internet status. Cloud backup handles disaster recovery without requiring cloud dependency for normal operation.
What Changed
The whiteboard is still on the wall — old habits die hard, and we did not fight it. But it stopped being the system of record within two weeks of go-live. The operations manager told us the biggest change was not efficiency — it was that she could finally take a day off without the whole operation depending on her being reachable by phone.
For the owner, the shift was visibility. Previously, knowing what was happening required walking the floor or calling someone. Now a dashboard answers the key questions — what is being produced today, what stock is low, what deliveries are out — in under ten seconds from any device.
Is This Relevant to Your Business?
The bakery scenario is specific, but the pattern is not. Almost every production or operations business we have visited runs on a combination of whiteboards, spreadsheets and group chats that have been patched together over time. The information exists — it is just scattered, manual, and dependent on specific people remembering to update things.
A connected operations system does not have to be complex or expensive. It just has to be built around how your business actually works, not how a generic platform expects you to work. If you run a production, manufacturing, food service or logistics operation and recognise this situation, we would like to hear about it.