Which order runs on which machine tomorrow?
Ask Ramesh bhai.
When can we promise dispatch to the Pune customer? Ask Ramesh bhai.
Twenty-two years on the shop floor, and the entire production plan of a ₹9 crore fabrication unit lives inside one supervisor's head. He walks the floor carrying the schedule like a secret.
If you've read Who Is Your Rajan?, meet his brother. He works in your factory.
Last quarter, 34% of the unit's orders shipped late. Rejections were discovered at final dispatch — after the material had travelled through every single stage. The owner blamed the workers. The workers blamed the machines.
Nobody blamed the real culprit: there was no written plan to fail against.
Late delivery is not a labour problem. It's a planning problem.
When the plan lives in one head, three things happen quietly.
Every "urgent" order reshuffles the whole floor — silently, with no record of what got pushed. Delivery dates are promised in sales meetings by optimism, not by capacity. And quality is checked at the end, which is the most expensive possible place to find a defect: a part rejected at dispatch has already eaten six stages of labour, power, and time.
The factory isn't undisciplined. It's blind. You cannot hold people to a plan they have never seen.
How do you fix production planning in a small factory?
Not with a ₹15 lakh ERP. With a board and a rhythm:
- Make the plan visible. One board — physical whiteboard or one shared sheet. Machine-wise loading for the next 7 days. If the plan isn't visible, the factory doesn't have a plan; it has a person.
- Run a 15-minute daily production meeting. Standing. Same time every morning. Three questions only: What did we plan yesterday vs. what actually happened? What runs today? What's blocking us? Fifteen minutes a day replaces three hours of firefighting.
- Put quality gates at stages, not at dispatch. Check after stage 2, not after stage 6. A defect caught early costs a correction; a defect caught at dispatch costs the whole job.
- Promise dates from the board, not from memory. Sales commits what the loading chart shows. The day your delivery promises come from capacity instead of courage, your on-time percentage starts climbing.
- Track three numbers weekly. On-time delivery %. Rejection %. Plan-vs-actual %. Written on the same board, every week. What the floor sees, the floor improves.
Ninety days of this and the factory develops something it never had: a memory of its own — one that doesn't take leave, doesn't resign, and doesn't forget.
The real question
If your production supervisor took ten days of leave starting tomorrow, would your factory know what to run on Monday morning?
If the honest answer is no, your delivery dates are hostage to one person's health, mood, and loyalty.
If you want to see where your factory's plan actually lives — and how to move it out of one head and onto one board — we map exactly that in a complimentary Strategic Advisory Session with our Director of Strategy.
Book a Strategic Advisory Session →
Related reading: Who Is Your Rajan? · You Made SOPs. Nobody Follows Them
Founders also ask
What is production planning for a small factory?
A visible, written answer to three questions: which order runs on which machine for the next 7 days, who checks quality at which stage, and what gets promised to customers based on real capacity. For most MSMEs this needs one board, one daily 15-minute meeting, and three weekly numbers — not an ERP.
How can MSMEs improve on-time delivery?
Promise dates from a machine-loading chart instead of memory, freeze the weekly plan so "urgent" orders are consciously traded rather than silently inserted, and review plan-vs-actual every morning. Most late deliveries are created on the day the promise is made, not the day the truck leaves.
How do I reduce rejection in manufacturing?
Move inspection upstream. Add quality checks after early stages instead of a single check at dispatch, record which stage each defect came from, and review the top defect source weekly. Catching a fault at stage 2 costs a correction; catching it at dispatch costs the entire job plus the delivery date.
Do I need ERP software for production planning?
Not to start. Software automates a process; it cannot create one. Run a manual board, daily meeting, and weekly numbers for 90 days first. Once the discipline holds and volumes grow, software is a useful upgrade — bought to speed up a working system, not to rescue a missing one.