How to reduce operational costs without cutting the work that matters

The usual advice for cutting operational costs is to renegotiate suppliers, freeze hiring and trim tools. Sometimes that is right. But it targets the price of inputs, and a lot of waste sits in how the work is done. A process that takes three touches where one would do costs the same however well you negotiated.

Step 1: price one process

Pick one repeated task and work out what it costs. The formula is simple:

minutes per case × cases per year × cost per minute of the person doing it

Here is an illustration with made-up round numbers. A team re-keys order details from emails into a system. It takes 20 minutes per order, with 25 orders a day over 230 working days. That is 115,000 minutes, or about 1,900 hours a year. At an all-in cost of USD 18 an hour, which includes salary, benefits and overhead, the task costs roughly USD 34,500 a year. Use your own figures, and treat the answer as an order of magnitude, not an accounting entry.

Now the useful part. If a form with required fields removes the need to re-key, you are not saving a share of a person. You are freeing around 1,900 hours for work that needs someone to think. Whether that becomes lower cost or more output is your decision, but you can now see the size of the prize.

Step 2: find the five common leaks

1. Waiting. Work stalled for approval or information. It costs little in labour but a lot in delayed revenue and unhappy customers. Look at handoffs.
2. Rework. Anything done twice. Every correction costs more than the original task, because it also involves finding the mistake and explaining it.
3. Duplicate entry. The same data typed into two or three places. It also creates inconsistencies, so there is a second cost in the errors it produces.
4. Over-checking. Approvals that are almost never refused. If a step approves 99 percent of what it sees, ask whether a limit would do the job.
5. Tool overlap. Several subscriptions covering the same function, or tools nobody has opened in three months.

Step 3: put the cost of errors on the same page

Errors are often missing from cost discussions because they are spread across teams. A wrongly keyed delivery address costs a failed delivery, a customer contact, a redelivery and possibly a refund. Count the last ten errors and estimate what each one cost to put right. The average becomes your cost per error, and the rework rate times that figure shows what you pay for quality problems.

Step 4: rank by payback

Candidate	Annual cost today	Effort to fix	Priority
Required fields on order form	High	Low	Do first
Remove approval below a set amount	Medium	Low	Do first
Automate reporting	Medium	Medium	Do second
Integrate two systems	High	High	Plan, then decide
Cancel unused tool	Low	Very low	Do now

What not to do

Do not cut the step that was catching errors. Some checks look like waste until they are gone. Ask what the check prevents before removing it.

Do not automate before simplifying. A ten-step process automated is still a ten-step process. Remove steps first. See fix the process before you automate it.

Do not claim savings that were not realised. If you free up hours, say where they went. Otherwise the saving exists on paper only.

A realistic expectation

Most teams find a handful of low-effort changes that return real time, and one or two larger projects worth planning. The numbers depend entirely on your starting point, and we do not give a percentage in advance. Our process improvement work starts with the pricing exercise above, and a worked example for e-commerce shows where it tends to land.

Related reading and services