Signs Your Business Is Wasting Time on Manual Work

Why is so much time wasted on manual work?

So much time is wasted on manual work because the waste arrives in pieces too small for anyone to flag, and it lands on people who were hired to do something else. The most expensive manual work is the work nobody thinks to question anymore. When a process has existed since before anyone thought to change it, the time it consumes stops registering as a cost and starts registering as "the job." A team can lose ten hours a week across a handful of tasks and never raise it, because no single task feels big enough to fix. The waste hides in the aggregate.

Four things keep it there. The waste is spread so thinly that it never shows up as a line anybody owns, and the people closest to it have the least time to step back and look at it. It gets counted as the work itself rather than as overhead, so nobody ever compares those hours against what else they could have bought. And the interruption around each task goes unlogged, so every one of them looks cheaper on paper than it turns out to be in practice.

The scale of it has been measured well outside any one business. The 2023 Anatomy of Work Global Index, conducted by GlobalWebIndex for Asana across 9,615 knowledge workers in the United States, the UK, Australia, France, Germany and Japan, found that people spend 58% of the day on what it calls work about work: coordinating, chasing, searching and updating, rather than the skilled job they were hired to do. The same respondents estimated that better processes would hand them back 4.9 hours a week each.

Part of the cost sits in attention rather than in hours, which is the part that never gets logged against the task that caused it. A Harvard Business Review study published in August 2022, which observed 137 users across three Fortune 500 companies for up to five weeks, found that people toggled between applications about 1,200 times a day and spent just under four hours a week reorienting themselves afterwards, close to 9% of their time at work. The keystrokes are the cheap part of swivel-chair work. The expensive part is getting back into whatever the person was doing before they were pulled out of it.

Naming the signs is the first step, because you can't recover time you haven't noticed you're losing.

What are the seven signs your business is wasting time on manual work?

The seven signs are variations on one thing: paid, skilled time going into work that follows rules nobody has ever bothered to write down.

1. Skilled people are doing copy-paste work. When someone you hired for judgment spends part of their week copying fields between files or systems, you're paying specialist rates for clerical work.

2. The same information gets typed into multiple places. If a piece of data is entered into one system, then re-entered into another, that swivel-chair work is pure waste and a source of errors.

3. Reports are assembled by hand. If someone rebuilds the same report every week or month by pulling data from scattered sources, that recurring assembly is a prime automation target.

4. Work stalls when one person is out. If a process only runs when a specific person does it manually, the knowledge is trapped in a human instead of encoded in a system, a risk as much as a waste.

5. People are hunting for information. Time spent searching through listings, records, files, or inboxes to find the right item is time a well-built system could return in seconds.

6. There's a task everyone dreads. The recurring task people procrastinate on is usually the one that's tedious, manual, and ripe for automation; the dread is a signal.

7. You're hiring to keep up with volume, not to grow. If new headcount is going toward keeping up with manual processing rather than new value, the process, not the staffing, is the constraint.

Two of those have been measured outside your business, which helps when the waste has to be argued for rather than simply noticed.

On the second sign, re-typing: a 2019 study in the Journal of the American Medical Informatics Association compared point-of-care glucose results that clinic staff keyed into the patient record by hand against the same results sent across automatically. Of 6,930 manual entries, 260 did not match, a rate of 3.7 percent, and 37 of those were wrong by more than 20 percent. That is trained staff, entering a single number, somewhere accuracy matters a great deal.

On the fifth sign, hunting for information: the McKinsey Global Institute's 2012 report The Social Economy found knowledge workers spend 19 percent of the workweek searching for and gathering information, before anybody does anything with it. Close to a day a week, per person, on retrieval alone.

Which of the seven signs should you fix first?

The sign to fix first is the one scoring highest on three things at once: how often the work recurs, how expensive the person doing it is, and how much of it software can reach without making a judgement call. Recurrence usually decides it, because recurrence is what turns a small irritation into an annual number.

The arithmetic that settles the order is simple enough to do on paper. A report that takes two hours every week is 104 hours a year. The quarterly task everybody dreads, at six hours a time, is 24. The dreaded one feels four times worse and costs a little over a fifth as much, which is why dread is a useful signal about where to look and a bad basis for ranking. Count the recurrences before you trust the feeling.

Then price each one off what an hour actually costs, not off the wage. In its Employer Costs for Employee Compensation release for March 2026, the U.S. Bureau of Labor Statistics put private-industry compensation at an average of $46.60 per hour worked, $32.60 of it wages and salaries and $14.01 of it benefits, so benefits were 30.1 percent of the total. At that average, the 4.9 hours a week the Asana respondents wanted back works out at roughly $11,900 a year per person, and a specialist costs a good deal more than the average. Use your own payroll for the real figure; the national number is only there for the ratio.

What if the wasted time is in site visits and field work, not at a desk?

The wasted time in a site visit is almost never the visit itself. It is the paperwork wrapped around it. A site visit is four separate pieces of work, and people tend to judge the whole thing by the one piece that has to happen: somebody has to be there and look at it. The other three are the ones quietly eating the week.

Salesforce's State of Service report, seventh edition, published in September 2025, surveyed 6,500 service professionals and put admin work at about 30 percent of a technician's working time. Mobile workers in the same research estimated that 18 percent of their hours, more than seven a week, went on admin they considered wasted: filling in forms, and hunting for information rather than doing the job they were sent to do.

The test for whether this applies to you is whether the same information gets recorded twice: once in the field and once again at a desk. If a technician writes six measurements on a form, photographs the form, and somebody types those six numbers into the job system the next morning, you are paying twice for one piece of information, and the second time is the one with the error rate on it.

Be careful with anyone offering to remove the visit rather than the paperwork. Remote assessment from photographs and video is real, and for some routine checks it works well, but it is a different decision with a different risk attached, and it is worth taking separately from the paperwork question. The paperwork saving does not need you to stop visiting anybody.

What is manual work actually costing you?

What manual work is actually costing you is not the hours themselves, it is what those hours would have produced elsewhere. A specialist spending fifteen hours a week on manual extraction is a specialist not doing the skilled work you hired them for. Multiply a few manual tasks across a team, and the real cost is the output you never get: the research not done, the customers not served, the growth not pursued, because skilled time is tied up in work a system should handle.

There is a second question that follows straight after the figure, and it changes what the figure is for. Recovered hours can come back as a smaller payroll or as more work through the same team, and those are different decisions with different arithmetic behind them. For most of the operations we see, the hours come back as capacity: the same people handling more volume, because the constraint was never how many of them there were. Which of the two applies to you is worth settling before anyone builds anything, and how much time and money AI automation actually saves sets out both cases side by side. The same page covers the question underneath that one, which is how much actual time AI cuts from one task against how little it cuts from a day spread across many.

To turn the signs above into a figure you can defend, work through how to reduce operational costs with AI, which sets out the five places operational cost sits and how to size each one at your own loaded hourly rate.

What it looks like when it's fixed

What it looks like when it is fixed is that the work stops existing rather than getting quicker. A 3D-scanning lab at Universidad Maimonides showed the pattern clearly: about fifteen hours a week lost to manual data extraction and transcription, skilled research staff copying fields by hand. Several of the signs above applied at once. We built a system that took the whole workflow over, and those fifteen hours went back to research.

Frequently asked questions

Related resources

• Where AI creates the most leverage in a business
• AI automation for HVAC, plumbing and roofing companies
• How to reduce operational costs with AI
• Case study: 15 hours a week, recovered (Universidad Maimonides)