Better Operations Start Small

Someone on your team copies the same information from one system into another. Then they copy it again into a spreadsheet, because the system does not show what the next person needs. A customer calls to ask where things stand. Two people look in two places and give two answers. Later that week, someone makes the same correction they made last Tuesday.

None of this feels like a project. Each instance takes a few minutes. You can live with a few minutes.

Until you count.

Say five people lose ten minutes a day to the same recurring snag. That is fifty minutes a day. Across a 250-day work year, that is more than 200 hours. You did not hire anyone to spend five weeks a year on a problem nobody put on the agenda.

The problem is not the ten minutes. The problem is the repetition.

Those ten minutes are usually a signal, not the whole job. The starting point is a contained process: customer intake, scheduling, onboarding, purchasing, invoicing, how information moves, or the handoff between two people. A small recurring task is how that process shows up in the week. You are not hunting for tiny chores to shave. You are picking one process that is narrow enough to understand and important enough to matter.

That is what "start small" means. It is not a suggestion to spend your time on trivial problems. A focused process can still affect customers, employees, time, cost, quality, reliability, or revenue. The scope is contained. The impact does not have to be.

Why the small problem lasts

Recurring work problems survive because each occurrence looks cheap. Ten minutes of re-keying. One extra status email. One "can you check this?" ping. Nobody wants to convene a meeting about ten minutes. That is how extra work becomes the job.

When a process does not give people what they need, they invent a workaround. Over time the workaround becomes normal. John Carrier at MIT Sloan calls this the 10,000 pennies problem. You will not find dollar bills on the floor. You will find pennies everywhere. In an office, those pennies often live in a spreadsheet, the place work goes when the main system does not do what the day requires.

The minutes add up faster than they seem. In a 2022 Harvard Business Review study, researchers watched 137 people in back-office roles (finance, HR, supply chain, recruiting) across three large companies. People switched applications nearly 1,200 times a day. Each switch cost a little over two seconds of reorientation. That added up to just under four hours a week, or about five working weeks a year. Those were large-company jobs. The pattern is familiar in a twenty-person shop running on email, an accounting system, and a side tracker: fetch from here, retype there, hope the two stay in sync.

Status-chasing, missing information, duplicate entry, and the spreadsheet that exists because the system does not: these are not small because they are unimportant. They are small because each instance is easy to shrug off.

Start where you can see the work

A contained process has edges. You can watch how the work actually moves. You can see where it stalls, who waits, what gets retyped, and what the customer experiences. You can change one thing and tell whether the week got better.

That is harder to do when the project is "fix how we operate."

In a 1992 Harvard Business Review article, Robert Schaffer and Harvey Thomson described an auto-parts plant that ran involvement meetings for six months. People offered hundreds of suggestions. Quality and delivery barely moved. Management then asked one line to cut its most common defect by 30 percent in two months. The line hit the target, cut that defect again, and within about four months the plant's scrap rate was inside the budget. Same plant. Different scope. The scope was specific—one defect on one line—and the result was visible in weeks, not years.

Harvard's Teresa Amabile and Steven Kramer studied nearly 12,000 diary entries from 238 people in seven companies. The event most associated with a good day at work was making progress on work that mattered. Small, finishable steps were nearly as motivating as rare breakthroughs.

You do not need a new playbook to use that. If people can finish a real change in a process they live with every day, the work gets easier to sustain. If the first finish line is "the entire system," a well-supported project can still wander.

You do not have to transform everything first

Some businesses need a large change. A merger, a new location, a system that truly cannot support the work. Those are real.

The mistake is treating a broad transformation as the automatic response to every operational problem. You do not have to remake the whole business before you are allowed to improve one process.

Large, multi-year programs often fall short of their original ambitions. In McKinsey's 2021 survey of more than 1,000 people who had been part of a transformation, fewer than one in three said the effort both improved performance and made the improvement last. That is not proof that big change is a bad idea. It is a reminder that increasing the scope also increases the complexity of the change, and makes it easier to spend a year without anyone's Tuesday getting easier.

Do not assume you need a program that touches every department before you start.

What happens after the first improvement

One finished process change is useful on its own. It also teaches you how work actually happens in your business. Then you do the next one.

A series of complete, visible results can add up. Karl Weick made that point years ago: one modest outcome looks limited; a series of them changes who will support the next step. Transformation, in that picture, is what accumulated improvements can produce. It does not have to be the required starting scope.

Imagine a service business where every new job starts in the inbox. Someone copies the request into a tracker. Someone else builds the schedule in a spreadsheet only they understand. The customer calls for a status update because nothing in the system answers the question. Fix how a job moves from request to schedule, and the copy-and-paste, the one-person schedule, and the status calls often move together. That is still one process. It is not a company-wide remake. And it is not a ten-minute chore.

Starting small is how you get a result you can see. It is not a promise that the first result will finish the job.

Look at your own week

You do not have to fix everything to make something meaningfully better.

Look at the work that happens every day. The copy-and-paste between systems. The spreadsheet that exists because the primary tool does not. The customer or employee who keeps asking for a status update. The missing information that creates another follow-up. The correction that gets made over and over. The work that sits between two people. The schedule that only one person can build.

Ask one question:

What is one recurring workflow or system issue making work harder than it needs to be?

Pick that. Understand how the work actually happens. Change something you can finish. See whether the week improved. Then do it again.

Better operations start small. One process. One improvement. Measurable results. Repeat.