How Much Does Routine Work Really Cost Your Business?
A simple formula and a worked table show what repetitive admin, support and ops work really costs a small business per year, by team size.
Most small businesses underestimate what routine work costs, because it is spread across people and never shows up as a single line on a budget. A simple formula - routine hours per week, multiplied by the loaded hourly rate of whoever does them, multiplied by 52 weeks - usually turns into a number well into six figures once a team passes ten people. This guide walks through the formula, a worked example for three team sizes, and what a reasonable next step looks like once you have the number.
The Formula: Hours x Loaded Rate x 52
The formula is deliberately simple, so you can run it in five minutes instead of commissioning a time-and-motion study:
Annual cost of routine work = routine hours per week x loaded hourly rate x 52
Two terms need defining before the result means anything.
Routine hours per week
This is time spent on tasks that are repetitive, rule-based and would not need a skilled decision if someone wrote the rules down once: updating prices across a catalogue, copying data between a chat thread and a CRM, answering the same five support questions, chasing overdue invoices, building the same report every Monday morning. It excludes judgment calls, relationship work and anything that changes meaningfully from one instance to the next.
Loaded hourly rate
Not the number on the offer letter. The loaded rate includes payroll tax, benefits, software seats, a share of management time and a share of office or tooling overhead. Most finance teams land on 1.25x to 1.5x base salary. If a role pays $20 an hour, the loaded rate usually sits between $25 and $30. Skip this step and the formula undercounts the real cost by a quarter or more.
A Worked Example Across Three Team Sizes
The table below uses typical, publicly observed ranges for how much routine work accumulates as a team grows, not a specific client’s numbers. Your own figures will vary by industry and how much is already automated.
| Team size | Typical routine hours/week (team total) | Loaded hourly rate | Cost per week | Cost per year |
|---|---|---|---|---|
| 5 people | 40 hours | $30 | $1,200 | $62,400 |
| 15 people | 90 hours | $35 | $3,150 | $163,800 |
| 40 people | 200 hours | $40 | $8,000 | $416,000 |
The pattern is not linear. As a team grows, routine work tends to grow faster than headcount, because more people means more handoffs, more status updates and more of the same report requested by more stakeholders. A 40-person team here has 8x the headcount of the 5-person team but 6.7x the routine hours - better than linear, but still a very large number sitting quietly on nobody’s budget line.
Where the Hours Actually Hide
- Sales and client admin - copying leads between a chat app and a CRM, updating deal stages by hand, re-typing the same quote with new numbers.
- Support replies - answering the same handful of questions dozens of times a day, across however many channels customers use.
- Catalogue and pricing - updating prices, stock levels and product descriptions across marketplaces whenever a supplier changes a number.
- Finance - chasing invoices, reconciling payments, re-entering the same data from one system into another.
- Reporting - pulling the same numbers into the same spreadsheet on the same schedule, then formatting it for whoever reads it.
Any one of these looks small in isolation. Added up across a team and a year, they are usually the largest line item nobody has actually calculated.
What This Number Usually Misses
The formula above is a floor, not a ceiling. It leaves out at least three real costs:
- Error cost. Manual, repetitive work is where mistakes concentrate - a wrong price copied into a marketplace, a lead that never made it into the CRM, an invoice sent twice. Each one costs more than the minutes saved by doing it quickly.
- Turnover cost. Routine work is also the least satisfying work for most people, and it is disproportionately represented in the tasks a departing employee hands off badly, or a new hire takes weeks to learn.
- Contractor risk. Many businesses outsource exactly this kind of work to a freelancer or a rotating cast of assistants. When that person is unavailable during a busy week, the routine work either stalls or falls back onto someone more expensive.
Our Own Numbers, For Context
These are not projections - they are what routine work looks like once it is actually automated, from our own projects. We built a system that keeps prices in sync for a 25,000-product catalogue across marketplaces without a person touching each listing. A separate price monitor we run checks competitor prices continuously and caught 184 of 218 real undercut events with zero false alarms, work that would otherwise be a person checking prices by hand all day. On the recovery side, an automated process we built reconstructed 571 of 571 orders directly from a Telegram chat history, a task that would have taken a team days of manual reading and re-entry. None of these numbers are guarantees for your business, but they show the order of magnitude the formula above is pointing at.
Why the Number Is Usually Worse Than It Looks on Paper
Three patterns make the real cost higher than a single calculation suggests.
- Routine work expands to fill the time available. Without a hard limit, a report that takes 20 minutes grows a new section every few months until it takes an hour, because nobody owns the decision to stop adding to it.
- The same hours get paid for twice. A task done manually and inconsistently often gets redone or corrected later, which means the loaded hourly rate above is frequently an undercount of what that task actually costs once rework is included.
- It is the first thing cut when things get busy, which means the backlog piles up quietly until it becomes a customer-facing problem - a late invoice, a stale price, a support reply that never went out.
None of these show up in a single week of time-tracking, which is why the formula is a useful floor, not a precise audit. Treat the result as “at least this much,” and expect the real number to be somewhat higher once you account for rework and backlog.
What To Do With This Number
- Pick the largest or most error-prone bucket first, not the easiest one. The formula tells you where the money actually is.
- Get a fixed-price audit before committing to a build. A short, paid audit of one bucket gives you a real quote instead of a guess, and tells you whether automating it is worth doing at all.
- Decide what a human still needs to approve. Most routine work is not fully hands-off even after automation - someone should still sign off on anything touching money, refunds or customer-facing promises.
- Automate the highest-volume bucket first, then move down the list. Trying to fix everything simultaneously is how automation projects stall.
- Re-run the formula in six months. Headcount, catalogue size and support volume change, and so does the number.
A Quick Gut-Check Before You Run the Full Audit
If you want a rough signal before committing to a formal process, ask three people on the team, independently, to estimate what percentage of their week goes to repetitive, rule-based work. Most founders expect an answer around 10 to 15 percent. In practice, teams that have never measured it usually land closer to 25 to 35 percent once support replies, data re-entry and status reporting are all counted honestly. If your own estimate comes in anywhere near that range, the formula above is not an academic exercise - it is describing a real, recurring line item that is currently invisible on your budget.
How We Approach It
We run a dedicated routine takeover service for exactly this problem: we start with a paid audit from $400 that tells you, in writing, which routine tasks are worth automating and roughly what it will cost, then move into ongoing work from $1,200 a month once you have a real plan instead of a guess. For a single, well-defined repetitive process, our automation builds start from $500 per automation. If you want ideas before talking to anyone, our catalogue of 100+ ready automations is a reasonable place to see what has already been built for problems like yours.
Not sure where your own number lands? Tell us the task and we will run the formula with you, free, before you commit to anything.
FAQ
What counts as routine work for this formula?
Anything repetitive, rule-based and not requiring a skilled judgment call once the rules are written down: updating prices, copying data between a chat and a CRM, answering the same support questions, chasing invoices, building the same report every week. If a new hire could do it correctly from a one-page instruction sheet, it is routine.
Isn't this just hours times a wage - why call it a loaded hourly rate?
Because the wage alone understates the real cost. Payroll tax, benefits, software seats and a share of management time typically add 25 to 50 percent on top of base pay. Using the loaded rate instead of the wage is the difference between an estimate and a number you can actually act on.
How do I find my own routine-hours number without a time-tracking project?
Ask each person to list, for one week, every task that took more than ten minutes and felt repetitive. You do not need perfect data, a rough tally from a few people is usually enough to see whether the number is in the thousands or the hundreds of thousands per year.
Does automating routine tasks actually remove the cost, or just move it?
It moves most of it to a much smaller number: the cost of building and maintaining the automation, usually a fraction of the yearly labor cost it replaces. Some human time is still needed for exceptions and approvals, but it drops from hours per day to minutes.
What's a reasonable first step if the number scares me?
Do not try to fix everything at once. Pick the single bucket of routine hours that is largest or most error-prone, get a fixed-price audit on that one bucket, and decide from a real quote rather than a feeling.