Spreadsheets vs ERP in manufacturing: when Excel is no longer enough

When spreadsheets start to cost more than they save

A factory that began with a single workbook for orders and stock can end up with fifty linked files, macros that break every month, and a finance team that spends two days reconciling the month-end trial balance. The shift from useful tool to operational bottleneck is gradual, but the signals are concrete: negative stock that no one can explain, cost reports that arrive after the customer has already renegotiated the price, and purchase orders that are approved on paper but never entered into the system. These are not teething problems; they are the cost of staying on spreadsheets when the plant has grown beyond them.

Inventory: the first point of failure

In a small plant, one person can walk the stores, count the rolls and update a single sheet. When the plant adds a second shift, a dyehouse and a finishing line, the same sheet becomes a ledger that no one trusts. Negative stock appears because the cutting room updates its own file but the stores clerk is on leave. Batch numbers are overwritten because two users open the same workbook at once. The result is not just inaccurate stock; it is working capital that is either overstated (leading to unnecessary purchases) or understated (leading to stock-outs and rush orders).

An ERP system like Facteno maintains one stock ledger for every store, batch and rack. When the cutting room issues fabric, the system reduces the stores balance in real time and posts the transaction to the general ledger. There is no parallel file, no negative stock and no month-end reconciliation. The inventory module enforces the rule that every movement must be recorded before the next one can happen, so the ledger is always accurate.

However, spreadsheets still have a place for local decisions. A dyehouse supervisor can use a simple sheet to track daily production against target, because the data is only needed for the next shift. The moment that data is needed by finance for costing or by planning for the next week’s schedule, the sheet becomes a liability.

Costing: the gap between estimate and actual

Most factories start costing on a spreadsheet. They list material, wages, utilities and overheads, apply a markup and quote the customer. The problem arises when the actual costs are collected weeks later and the numbers no longer match. The dyehouse used more steam because the boiler was serviced late; the stitching line had two absentees; the packing department used extra cartons because the supplier sent the wrong size. Each variance is small, but together they turn a profitable order into a loss.

An illustration: a towel plant quotes a customer £1.20 per piece based on 1.5 kg of yarn at £0.60/kg, 0.2 hours of stitching at £12/hour, and £0.10 for packing. The actuals come in at 1.6 kg of yarn (£0.96), 0.25 hours of stitching (£3.00) and £0.15 for packing. The actual cost per piece is £1.31, a 9% variance that wipes out the planned margin. If the plant had used an ERP system with product costing built from actual material, wages and utilities, the variance would have been visible the day after the order shipped, not three weeks later.

Spreadsheets are still useful for ad-hoc analysis. A production manager can pull a week’s data into a sheet, pivot by machine and shift, and see which line is consistently over budget. The moment that analysis needs to be repeated every week and shared with finance, the sheet should be replaced with a report from the ERP system.

Compliance and audit: the hidden risk

Regulated plants—food, pharmaceutical, automotive—must maintain batch records, inspection reports and approval chains. A spreadsheet can record the data, but it cannot prove that the data was not altered after the event. Auditors look for three things: who entered the data, when they entered it, and whether anyone else changed it later. A spreadsheet fails on all three counts. The user name is the Windows login, the timestamp is the file’s last-saved date, and the history is lost when the file is closed.

An ERP system maintains an audit trail for every transaction. When a quality inspector records a defect, the system logs the user, the timestamp, the device and the exact values entered. If a supervisor later edits the record, the system logs the change and the reason. This trail is not optional; it is the difference between passing an audit and facing a recall.

Spreadsheets are still acceptable for internal reviews. A safety officer can use a sheet to track near-miss incidents, because the data is only used to identify trends. The moment that data is needed for an external audit or a customer complaint, the sheet must be replaced with a controlled document in the ERP system.

When a spreadsheet is still the right answer

Not every process needs to move to ERP. The rule is simple: if the data is only used by one person, for one purpose, and never needs to be shared or repeated, a spreadsheet is the right tool. Examples:

• The maintenance team uses a sheet to track weekly greasing rounds, because the data is only used to schedule the next round.
• The HR assistant uses a sheet to record leave requests, because the data is only used to update the payroll once a month.
• The sales team uses a sheet to track customer visits, because the data is only used to plan the next trip.
• The production manager uses a sheet to calculate shift efficiency, because the data is only used to brief the team the next morning.

The moment any of these sheets is emailed to another department, or used to generate a report, or relied on for a decision that affects the whole plant, it should be replaced with a process in the ERP system.

How to tell if your plant has outgrown spreadsheets

Use this checklist to assess whether the problems you see are caused by spreadsheets or by something else. If three or more items are true, the plant has outgrown Excel for core operations.

• The month-end stock count takes more than one day and still does not reconcile with the ledger.
• The cost report for the previous month is not ready until the 15th of the current month.
• Purchase orders are approved on paper but not entered into the system until the goods arrive.
• Negative stock appears at least once a month and no one can explain why.
• Finance spends more than two days reconciling the trial balance after month-end.
• Customer complaints about quality are traced back to data that was entered incorrectly or late.
• More than one person maintains a separate file for the same data (e.g., orders, stock, production).

If the problems are caused by spreadsheets, an ERP system will solve them. If the problems are caused by poor processes or lack of discipline, software will not help.

What to do next week

Start with the process that causes the most pain. For most plants, this is inventory or costing. Pick one product line or one store and map the current flow on paper. Note every file, every email, every manual entry and every approval. Then compare it to how the same process would work in an ERP system. Facteno provides a live demo with real data, so you can see the screens and reports without committing to a purchase.

Next, talk to the people who will use the system. Ask them what they spend the most time on, and what they wish they could see but cannot. If the answer is “reconciling spreadsheets,” the case for ERP is clear. If the answer is “entering the same data into three different systems,” the case is even stronger.

Finally, plan the transition. An ERP system is not installed overnight. Facteno uses a staged implementation plan that moves one process at a time, so production is never stopped. The first stage is usually inventory or sales, because these have the most immediate impact on working capital and customer service.

When to walk away

If the plant is not willing to change its processes, or if the team is not willing to learn new software, an ERP system will not work. In that case, the best next step is to tighten the existing spreadsheets: add data validation, protect the sheets, and enforce a rule that no one edits the master file. It will not solve the underlying problems, but it will reduce the errors until the plant is ready for a real system.