How to Calculate the Real Cost of Manual Admin

Manual admin is not free just because it is absorbed into people’s working week.

To calculate the real cost, start with the people involved, the time spent, the frequency of the task and the loaded hourly cost. Then look at the extra cost created by rework, delay, management checking, weak evidence and dependency on local workarounds.

For CEOs, COOs and CFOs, this matters because repeated manual admin is often hidden inside normal operations. It rarely appears as a separate budget line, but it still consumes capacity, creates risk and slows decisions.

The aim is not to label every manual task as waste. Some manual work gives people control, confidence and connection to the detail. The question is whether the current process is using people’s judgement well, or whether skilled staff are quietly holding together a fragile workflow through repeated chasing, copying and checking.

Why manual admin is so easy to underestimate

Manual admin often looks harmless because it is spread across teams.

One person updates a spreadsheet. Another checks the numbers. A manager chases missing information. Someone else rebuilds the report before a meeting. A colleague keeps a second tracker because the first one does not quite show what they need. None of those tasks may look expensive on its own.

The cost appears when the pattern repeats every week, across multiple teams, sites or reporting cycles.

That is why the real question is not “how long does this task take once?” The better question is:

How much capacity, management attention and operational confidence does this manual process consume over a year?

Once you ask that question, manual admin starts to look less like background noise and more like an operating cost.

The simple formula

A useful starting point is this:

Manual admin cost = people involved × time per task × frequency × loaded hourly cost

For example, if three people each spend 45 minutes every week maintaining, checking and sending a manual report, the direct time cost is based on:

• the number of people involved
• the time each person spends on the task
• how often the task happens each year
• the loaded hourly cost of the roles involved

The important phrase is loaded hourly cost. Basic pay is only part of the cost. Employers should normally consider salary, employer National Insurance, pension contributions, levy costs where relevant, and any other employment costs that apply to the organisation.

For a broad UK reference point, the Office for National Statistics reported median hourly earnings excluding overtime for full-time employees at £19.67 in April 2025. Employer on-costs vary, but GOV.UK’s National Insurance guidance shows employer contributions are a material part of the cost above relevant thresholds.

That does not mean every calculation should use a national median. A CFO or operations leader should use actual role costs wherever possible. The point is that manual admin should be costed using the real cost of employing people, not just a rough salary guess.

A worked example

Imagine a weekly operational report that has to be rebuilt manually before every senior meeting.

• Four team members each spend 30 minutes updating source spreadsheets.
• One administrator spends 90 minutes consolidating the information.
• One manager spends 45 minutes checking, correcting and chasing gaps.
• The process usually repeats every week.

That creates 4.25 hours of direct manual work each time the report is produced.

If the process happens 52 times a year, that is 221 hours per year before any extra rework, delay, checking or meeting preparation is added.

If the illustrative loaded hourly cost is £30, the direct annual cost is:

221 hours × £30 = £6,630 per year

That is only one report. If similar processes exist across compliance, finance, HR, estates, operations and board reporting, the annual value of manual admin can become significant very quickly.

This example is not a savings claim. It is a way of making hidden workload visible. Whether that cost can be reduced, redirected or removed depends on the process, the people involved, the systems already in place and whether the organisation changes how the work is done.

Assumptions behind the example

The worked example is illustrative. In a real calculation, the assumptions should be replaced with actual role costs, task frequency and time estimates from the organisation.

• People involved: the example includes four team members, one administrator and one manager. In your own calculation, include everyone who updates, checks, chases, approves or rebuilds the information.
• Time per cycle: the example uses 4.25 hours per report cycle. Include preparation, data entry, checking, correction, chasing and final formatting.
• Frequency: the example assumes 52 report cycles per year. Use the actual number of times the task happens each year. Some processes are weekly, monthly, termly or ad hoc.
• Loaded hourly cost: the example uses £30 per hour for illustration. Use actual role costs where possible, including employer on-costs rather than salary alone.
• Direct annual cost: the example produces a visible labour cost of £6,630. Treat this as the visible labour layer only. Rework, delay, checking and assurance effort may add further cost.

The hidden cost stack

Direct time is only the first layer. The real cost of manual admin often sits in the layers around the task.

This wider view matters because some manual processes appear cheap until the organisation asks what they are doing to confidence, pace and resilience.

A spreadsheet maintained by one experienced person may look efficient. But if no one else understands the logic, the organisation has not only created manual work. It has created operational dependency.

Cashable savings and released capacity are not the same thing

This is where many automation business cases become too optimistic.

If a process improvement releases ten hours a week, that does not automatically mean the organisation has saved ten hours of salary cost. The value depends on what changes afterwards.

There are several different types of benefit:

• Cashable saving: a real budget reduction, avoided hire, reduced overtime, lower agency spend or cancelled software cost.
• Capacity release: time redirected to higher-value work, service improvement, customer support, assurance or management grip.
• Risk reduction: fewer missed actions, better evidence, clearer ownership or earlier escalation.
• Resilience gain: less dependency on heroic admin or one person’s workaround.
• Decision improvement: leaders get clearer information earlier and can act sooner.

All of these can be valuable. They are not all the same.

A credible business case should be clear about which type of value is expected. If the benefit is released capacity, say that. If the benefit is reduced risk, say that. If the benefit is a cashable saving, show the assumptions behind it.

Manual work is not always meaningless work

It is tempting to describe manual admin as pure waste. Sometimes it is. But in real organisations, manual processes often survive because they are doing more than moving information from one place to another.

A manual check can give someone confidence that the work is safe. A spreadsheet can feel tangible. A report rebuilt by hand may be how a manager stays connected to the detail. A local tracker can give a team a sense of ownership when the formal system does not answer the questions they need answered.

That does not mean manual admin should be protected for its own sake. It means process improvement has to respect the human role inside the work.

Even when a process is automated, people are still involved. They own the exceptions, interpret the outputs, maintain the trust and decide whether the new way of working becomes normal practice. The best automated processes do not make people feel replaced. They make people feel that repetitive friction has been removed so their judgement, attention and contribution matter more.

For a process to deliver, people need to want it to work. If automation removes a task without explaining what people still own, what stops, what changes and where human judgement remains essential, staff may quietly rebuild the old workaround.

The organisation then pays twice: once for the new process, and again for the hidden manual process that never really went away.

When the answer is not automation

A high manual admin cost does not automatically mean the task should be automated.

Sometimes the right answer is to remove the task completely. Sometimes the process needs clearer ownership. Sometimes the data definition needs to be agreed. Sometimes two reports should become one. Sometimes the existing system already holds the answer, but the organisation has not configured or governed it properly.

Automation is most useful when the work is repeated often, follows clear rules, uses reliable data and has known exceptions. If the process is unclear, politically unresolved or built on poor data, automation can make the problem faster without making it better.

Good candidates for automation

A manual process may be a good candidate for automation when:

• repeated manual steps happen frequently
• the rules are clear enough to describe
• the source data is reliable enough to use
• exceptions are known and can be routed
• the process has an accountable owner
• people understand how the new workflow helps their work

When to pause before automating

Automation should usually wait when:

• the process exists because no one has agreed the real requirement
• different teams use different definitions
• the data has to be manually corrected every time
• no one owns the exceptions
• staff are likely to keep the old spreadsheet anyway
• the organisation cannot explain what benefit it expects

How to build a better cost picture

A useful manual admin cost review does not need to become a large transformation exercise. Start with one repeated workflow that people already complain about or quietly work around.

1. Choose one recurring process: for example, a weekly report, compliance tracker, invoice check, HR return, estates log or board pack input.
2. Map who touches it: include everyone who updates, checks, chases, approves or rebuilds the information.
3. Estimate the time honestly: include preparation, correction, chasing and final checking.
4. Apply a loaded hourly cost: use actual role costs where possible, not a rough salary guess.
5. Identify rework and delay: note where poor data, missing ownership or unclear definitions create extra effort.
6. Separate value types: decide whether the likely benefit is cashable saving, released capacity, reduced risk, better assurance or improved resilience.
7. Ask what should change first: remove, redesign, standardise, improve using existing tools, or automate.

This approach gives leaders a more honest view. It avoids two common mistakes: pretending manual admin is free, and pretending automation automatically converts every released hour into a cash saving.

How Processive would approach it

Processive starts by making the work visible.

That means looking at the workflow, the people involved, the systems already in place, the handoffs, the hidden checks and the points where information has to be rebuilt or trusted manually.

From there, the question is not simply “can this be automated?” It is:

What should be removed, what should be redesigned, what can be improved using the existing stack, and what is stable, valuable and safe enough to automate?

That order matters. It protects organisations from buying tools for problems they have not properly understood. It also protects staff from changes that remove visible tasks without preserving ownership, judgement and contribution.

A better process should reduce avoidable friction while making the human role clearer and more valuable.

Manual admin is a signal

Repeated manual admin is rarely just an admin problem. It is a signal that something in the operating model needs attention.

It may be a data problem. It may be a system configuration problem. It may be a reporting problem. It may be an ownership problem. It may be that the organisation has outgrown a workaround that once made sense.

The useful move is to make the cost visible, understand why the work exists, and then decide what should happen next.

Sometimes that will mean automation. Sometimes it will mean a simpler process. Sometimes it will mean using Microsoft 365, Google Workspace, Power Platform or another existing tool more effectively. Sometimes it will mean stopping a report that no longer supports a real decision.

The value comes from making that judgement properly.