The tolerated pain
A delivery goes to the old address because the customer's change was noted in an email but not in the system. An invoice is issued at the wrong price and has to be credited and reissued. A job card is missing a part number, so the technician drives back to the branch. A booking is entered for the wrong date and the client arrives to find nobody expecting them. Each mistake is small, each is fixed, and each takes someone away from the customers they should be serving.
In many businesses, fixing mistakes has quietly become part of the job description. Staff are praised for sorting problems out quickly, and they are good at it. But the time they spend correcting avoidable errors is time they are not spending on new orders, enquiries or the customer standing in front of them.
This article is about the cost of correction work that a better process would have prevented. The question of how to reduce error without removing judgement is explored in how to reduce human error without removing human judgment. Here the focus is the money: what preventable rework costs, and why it usually costs more than the business thinks.
Why businesses put up with it
- Fixing feels productive. A problem solved is visible and satisfying. A problem prevented is invisible.
- Blame goes to people. Mistakes are treated as individual carelessness, so the response is a reminder rather than a process change.
- Corrections are not counted. Credit notes, re-deliveries and repeat visits are booked as normal operating costs, not as the price of errors.
- The cause is upstream. The person fixing the mistake is often not the person or system that caused it, so the source never learns.
Operational symptoms
- a steady flow of credit notes, reissued invoices or corrected orders;
- repeat site visits, re-deliveries or remakes;
- customer service staff spending more time on "sorting out" than on new requests;
- the same type of mistake recurring every month;
- checking steps added after each incident, slowing everything down without removing the cause;
- exceptions handled by hand so often they become the normal route, as described in when manual exceptions become the normal process.
Correction is a recognised waste
Lean management lists correction among the seven wastes, defined by the Lean Enterprise Institute as inspection, rework and scrap. All three appear in offices as well as factories. Inspection is the extra checking added because errors are expected. Rework is the corrected invoice or repeated visit. Scrap is the discounted order, written-off stock or lost customer.
The same tradition offers two practical tools. The first is poka-yoke, or error-proofing: simple, inexpensive mechanisms that stop a mistake being made or flag it immediately, so defects are caught at their source rather than passed on. Spell-check is the everyday example the Institute gives. The second is the 5 Whys, the habit of asking why repeatedly until the root cause appears, so the fix prevents recurrence instead of treating the symptom.
In business systems, error-proofing often means constraining what can be entered. Microsoft's guidance on applying data validation in Excel shows how a spreadsheet can restrict entries to whole numbers within limits, dates in a range or choices from a drop-down list, with an input message explaining what is expected and an error alert when something invalid is typed. Business applications offer the same ideas as required fields, picklists, lookups and validation rules.
How the pain becomes money
- Labour cost: time spent investigating, correcting and communicating about mistakes.
- Direct cost: re-deliveries, repeat visits, remakes, courier fees and written-off stock.
- Margin erosion: discounts and goodwill credits given to keep customers after an error.
- Opportunity cost: customer-facing staff who are fixing yesterday's problems are not serving today's customers or winning new ones.
- Revenue loss: some customers simply leave after repeated errors, and rarely say why.
- Cash-flow delay: disputed invoices are paid late, if at all.
An illustrative cost model
This example is illustrative; all figures are assumptions to show the method. Picture an Eastern Cape building supplies merchant with a counter, a sales desk and its own delivery vehicles.
| Input | Illustrative assumption |
|---|---|
| Preventable errors per month (wrong item, price, address, quantity) | 60 |
| Staff time to investigate and correct each error | 45 minutes |
| Loaded cost of staff time | R250 per hour |
| Errors needing a second delivery | 20 per month |
| Cost of each extra delivery (vehicle, fuel, driver) | R650 |
| Goodwill credits per month | R5,000 |
On these assumptions correction time costs about R11,250 a month, extra deliveries R13,000 and goodwill credits R5,000, a total of roughly R29,250 a month or R351,000 a year. That excludes lost customers and the sales that did not happen while staff were fixing problems. If error-proofing removed half the errors, the business would recover about R175,000 a year in direct costs and free around 270 hours of staff time to serve customers. Test the method on your own credit notes and delivery logs before relying on any figure.
When it becomes commercially urgent
- Credit notes or re-deliveries are a visible line in the monthly figures.
- Customer complaints mention repeated mistakes rather than one-off problems.
- Staff say they are too busy fixing things to follow up on sales opportunities.
- The business plans to automate a process that currently produces regular errors, which would only produce them faster; see when poor data quality becomes an automation failure.
What better looks like
Mistakes are counted, categorised and traced to where they started. The most frequent ones are designed out: drop-down lists replace free text, customer details come from one record instead of being retyped, prices are pulled from a maintained list, and orders cannot be saved without the fields that cause the most trouble. Checks happen at the point of entry, not after the delivery has left. When an error does occur, the person fixing it records the cause, and someone reviews the top causes each month.
Practical next actions
- Count corrections for one month. Pull credit notes, reissued invoices, re-deliveries and repeat visits into one list.
- Categorise by cause, such as wrong address, wrong price, wrong item or missing information.
- Ask "why" five times for the top three categories until you reach a cause the process can change.
- Error-proof at the point of entry with validation, picklists, required fields and single sources for prices and customer details.
- Remove redundant checks once the cause is fixed, so prevention replaces inspection.
- Review monthly and report correction costs alongside sales, so the trend is visible.
A structured review of where errors originate across systems is part of Mellorca's digital systems audit.
Bottom line
Every preventable mistake is paid for twice: once when it is made and again when it is fixed, often with a discount or an extra trip on top. The bigger cost is that the people best placed to serve customers spend their day repairing avoidable errors. Counting corrections for a month and designing out the top few causes usually returns more capacity than hiring another person to keep up.
Sources and further reading
- What are the 7 Wastes in Lean? - Lean Enterprise Institute (last verified 10 October 2026)
- Poka Yoke - A Resource Guide - Lean Enterprise Institute (last verified 10 October 2026)
- 5 Whys - What is it? - Lean Enterprise Institute (last verified 10 October 2026)
- Apply data validation to cells - Microsoft Support (last verified 10 October 2026)
Related Mellorca reading
- Answering the Same Question Again: The Payroll Cost of Undocumented Know-How
- Which Version Is Final? The Cost of Teams Working From Different Copies
- Recurring Tasks That Live in Someone's Memory: The Cost of Missed Work