Spreadsheets are not the enemy
Almost every good machine shop in the UK started on spreadsheets, and plenty still run on them. They are flexible and cheap, and the person who built them understands the business better than any software vendor.
The trouble comes later. The spreadsheet that was a tool becomes a system, and a system held together by one person's memory is a risk.
Signs you have outgrown them
- One person understands the planning sheet, and everyone is nervous when they are on holiday.
- The same information is typed three times: the order, the job sheet, the invoice.
- There are several versions of the truth. The plan says one date, the whiteboard another, the customer was told a third.
- Questions take too long. "Where is this job?" means walking the floor. "Did that job make money?" means an afternoon.
- Traceability is a search, not a lookup.
- Month end is a rebuild, with stock and work in progress estimated by eye.
- You have stopped quoting for work because you cannot tell whether you have the capacity.
Two or three of these is normal. Most of them is a sign that the spreadsheets are now costing more than they save.
Why the move feels risky
Owners put it off for good reasons. They have heard of ERP projects that took two years and changed nothing. They cannot stop production to "implement" something. And the spreadsheets, for all their faults, do work.
So the aim is not a big switch-over. It is a series of small moves, each of which leaves you better off even if you stopped there.
One step at a time
1. Write down how the job actually flows. From enquiry to payment, as it really happens, including the workarounds. This is worth doing even if you never buy software, and it shows you which spreadsheet matters most.
2. Clean the data you already have. Customers, parts, routings, stock. Remove the duplicates and the parts nobody has made in ten years. Good systems import from spreadsheets; none can fix what is in them.
3. Start with the thing that hurts most. For most shops that is the plan. For some it is traceability, or knowing what jobs cost. Get one area working properly before adding the next.
4. Run alongside, briefly. Put live jobs through the new system for one customer or one cell while the spreadsheet carries on for the rest. Keep this phase short: two systems are double the typing.
5. Put the record where the work is. A tablet at the machine for booking parts and time does more for accuracy than any amount of office discipline.
6. Retire spreadsheets deliberately. When an area is trusted in the new system, archive its spreadsheet and say so. The old file left open "just in case" is how shops end up with two truths again.
What to look for in a system
- it can import what you have, and check it first
- it lets you start in one area without buying or configuring everything
- it works the way a UK machine shop works: drawing issues, batches and certs, subcontract processes, pounds and VAT codes
- it leaves alone what already works, such as your accounts package
- the people selling it will help with the change, not only the software
The short version
You do not move off spreadsheets in a weekend, and you should not try. Map the job, clean the data, fix the worst problem first, and retire each spreadsheet on purpose. A year later the whiteboard is still there. It just no longer runs the business.