Skip to content
LIRGOoperational intelligence

Blog · 2 November 2026

The costs nobody counts

Rework, chasing, waiting and scrap usually don’t get a line of their own in the accounts: where the idea comes from, which figures to ignore and how to start counting them yourself.

7kinds of waste into which Toyota divides mudaJoe Clifford, “Toyota Production System - what it all means”, Toyota UK Magazine, date on the page 29 August 2025
4large multinationals studied in 2006: only one used a formal method for costing qualityAndrea Schiffauerova, Vince Thomson, “Managing cost of quality: insight into industry practice”, The TQM Magazine, vol. 18, no. 5, 2006, pp. 542-550
32US plants compared: both transactions and volume counted, but transactions explained most of the variation in overheadsRajiv D. Banker, Gordon Potter, Roger G. Schroeder, “An empirical analysis of manufacturing overhead cost drivers”, Journal of Accounting and Economics, vol. 19, no. 1, 1995, pp. 115-137

An order comes in by email and someone retypes it into the system. A measurement is missing, so someone picks up the phone. Materials are late, so someone chases the supplier. A part comes out wrong and gets made again. At the end of the month your accounts show what you spent on wages and materials. They don’t show how many of those hours went on all this.

The hidden factory

None of this is new. Armand Feigenbaum, author of Total Quality Control, came back to it in a 1994 interview with IndustryWeek. He called it the hidden factory: the part of your organisation that exists to do bad work. Not because you want to, he said, but because the whole process drives you into it. After 75 years of cost accounting, he added, most American companies still didn’t know what things cost.

Two Boston University professors, Jeffrey Miller and Thomas Vollmann, used the same name in Harvard Business Review in 1985. Their subject was overheads. In the hidden factory, they wrote, the real driver of cost is transactions, not products. These are exchanges of materials and information that keep production moving without directly turning out anything physical. A chasing email is exactly that.

In 1995 three researchers, Rajiv Banker, Gordon Potter and Roger Schroeder, put the idea to the test by comparing 32 American plants. Where transactions and volume were higher, overheads were higher, but transactions explained most of the differences. They measured transactions through engineering changes, purchasing and planning staff, floor area per part and quality staff. They didn’t count chasing as such.

Poor quality and the seven wastes

Two other classic ideas point at the same spot. The cost of poor quality is whatever would disappear if the job were done right first time. A 2015 review of 57 studies explains why these costs stay out of sight: they usually aren’t quantified or recorded in the accounting systems.

Toyota divides waste, which it calls muda, into seven kinds: transport, inventory, motion, waiting, over-processing, over-production and defects. The list goes back to Taiichi Ohno. Try it on your own work. An order keyed in twice is over-processing. A part sitting idle until the drawing arrives is waiting. A part made again is a defect.

The figures doing the rounds

Feigenbaum gave a figure too: about 20% to 40% of the capacity of many American companies. He said it in an interview in 1994 and offered no data. It isn’t a measurement of your business.

You’ll find plenty of other percentages online, and to us many look like slogans. The American Society for Quality’s page on the cost of quality and an article on juran.com quote shares of sales, revenue or operations without citing a study. We haven’t found a serious number for small Italian firms, so we won’t give you one. The same 2015 review does give a figure: 16.91% to 26.90% of revenue, pooling the studies it collects. But it admits that the systems for measuring these costs aren’t standardised and that most of the studies can’t be compared. That isn’t a measurement of your business either.

If you’ve never counted them, you’re in good company. A 2006 study of four large multinationals found that all four worked on quality systematically, but only one used a formal method for costing it. That’s a handful of firms, but the same authors’ literature review says the same of most quality initiatives.

How to start, without consultants

Write down the steps. Take a real job and follow it from quote to payment. On one sheet, note each step and who does it. Ask the people who do it, and don’t work from memory.

Count for a month. Pick a few headings: retyped, chased, waited for, redone, scrapped. Keep a sheet where it happens, and each time mark a tally and the minutes. Make one thing clear to everyone: you’re counting the step, not the person.

Put a cost on each heading. For time: times a month, by minutes, divided by sixty, by the cost of an hour’s work. For scrap and rework, add the materials. This isn’t accounting, just a sense of scale. Rank the headings from most expensive down and look at the first. If the total turns out small, you’ve still learned something useful.

Our principle is this: what isn’t written down can’t be improved.

Don’t automate the waste

In the same interview, Feigenbaum said that after ten or fifteen years of automation and computers, the hidden factory had if anything grown. In too many cases, he explained, waste had simply been automated. We think the same goes for artificial intelligence. An agent that reads emails and retypes orders will also speed up a step that should never have existed. That’s why our work starts with the written steps and a baseline figure: the agent gets connected afterwards, and a person still makes the decision.

Sources

  1. Tim Stevens, “Dr. Armand Feigenbaum on the Cost of Quality and the Hidden Factory”, IndustryWeek, 4 July 1994. www.industryweek.com
  2. Jeffrey G. Miller, Thomas E. Vollmann, “The Hidden Factory”, Harvard Business Review, September 1985. hbr.org
  3. Rajiv D. Banker, Gordon Potter, Roger G. Schroeder, “An empirical analysis of manufacturing overhead cost drivers”, Journal of Accounting and Economics, vol. 19, no. 1, 1995, pp. 115-137, manuscript held by Cornell University. ecommons.cornell.edu
  4. S. Mahmood, N.I. Kureshi, “A literature review of the quantification of hidden cost of poor quality in the historical perspective”, Journal of Quality and Technology Management, vol. XI, no. 1, June 2015. pu.edu.pk
  5. American Society for Quality, “What is Cost of Quality (COQ)?”, page read on 2 November 2026. asq.org
  6. Maureen Goldman, “The Office: How Poor Quality May be Hurting Your Bottom Line”, Juran, 17 October 2017. www.juran.com
  7. Wikipedia, “Armand V. Feigenbaum”, for the book Total Quality Control, McGraw-Hill, 1961. en.wikipedia.org
  8. Joe Clifford, “Toyota Production System - what it all means”, Toyota UK Magazine, date on the page 29 August 2025. mag.toyota.co.uk
  9. Toyota Motor Corporation, “Toyota Production System”, official global site. global.toyota
  10. Jean Cunningham, “The Eight Wastes of Lean”, Lean Enterprise Institute, 18 January 2020. www.lean.org
  11. Andrea Schiffauerova, Vince Thomson, “Managing cost of quality: insight into industry practice”, The TQM Magazine, vol. 18, no. 5, 2006, pp. 542-550. www.emerald.com
  12. Andrea Schiffauerova, Vince Thomson, “A review of research on cost of quality models and best practices”, International Journal of Quality & Reliability Management, vol. 23, no. 6, 2006, pp. 647-669. www.emerald.com

It starts with an hour, on your real jobs.