What a six-minute task actually costs you in a year
Every business has a task that takes a few minutes and happens constantly. Filling in the same customer details. Copying an order from a message into a system. Writing a job card by hand.
Nobody fixes those, because 6 minutes does not feel like a problem. The reason it does not feel like a problem is that nobody has ever multiplied it out.
Here is how I do it, and you can run it on your own numbers in about 20 minutes.
Step 1: time it properly
Not an estimate. Stand there with a phone and time it, 3 or 4 times, on a normal day.
Owners guess low on this, consistently. The version in your head is the version where nothing goes wrong. The real one includes the customer who cannot find their phone number, the printer that jams and the interruption halfway through.
Then time it again on your busiest day, because that is the version that actually costs you.
Step 2: count how often it happens
Per day, then multiply out across the year including the busy season.
This is where the number starts to look different. A task that takes 6 minutes and happens 40 times a day is 4 hours. Every day.
Step 3: price the hour
Take the monthly salary of whoever does the task and divide it by the hours they actually work. A 6-day week at 8 hours a day is roughly 208 hours a month.
That gives you a salary-based hourly figure. The real cost is higher once you add everything that comes with employing someone, so treat the salary number as your floor rather than your answer.
Multiply your annual hours by that rate and you have the first half of the number.
Step 4: the half that nobody counts
Staff time is the easy part. It is also usually the smaller part.
The bigger question is what the slow version costs you in work you never got. If a task creates a queue, and there is somewhere else your customer could go, some of them go there. If a quote takes 2 days to produce, some of those customers have already bought elsewhere by the time it lands.
You cannot measure that directly, which is exactly why it gets left out of every business case I see.
Estimate it instead. How often does a queue form? How long do people wait before they give up? How close is the nearest alternative? Take your lowest honest guess, not your best case, and multiply it by what an average customer is worth to you.
Do not add the same hours twice
This is where these calculations usually fall apart, so it is worth being careful.
If you keep the same staff and the same hours, the time you save becomes capacity. You count the extra work it lets you handle. You do not also count the wage, because you are still paying it.
If you cut hours instead, you count the wage, and there is no extra capacity to count.
The one case where both are real is when the slow process is actively costing you customers. Then the recovered customers are genuinely new revenue and the staff time is a genuinely separate saving.
What to do with the number
Compare it against what fixing it would cost, once.
If a process is costing 40,000 a year and the fix is 15,000, that is not a hard decision. If it is costing 4,000 a year, leave it and go find a bigger one, because your attention is worth more than that.
The point of the exercise is not to justify a purchase. It is to find out which of your annoying tasks are actually expensive, because they are rarely the ones that annoy you most.
If you want to see what this looks like applied to a real build rather than in the abstract, the cost of doing nothing covers the same arithmetic from the other direction.
Want somebody to time your process properly and do this arithmetic with you?
Book a free 30-minute call. We watch the real work before we quote anything, which means the number you get is measured rather than guessed.
Questions about costing a manual process
Time the task properly, count how many times it happens in a year, and multiply. That gives you hours. Then divide a monthly salary by the hours that person actually works to get an hourly cost, and multiply again. That is the easy half of the number and usually the smaller half.
Both, but never the same hours twice. If you keep the same staffing, the saved time becomes capacity and you count the extra work it lets you take. If you cut hours instead, you count the wage. Adding both to the same hours double counts, unless the slow process is also driving customers away, which is a separate loss.
Take the monthly salary and divide it by the hours that person actually works in a month. A 6-day week at 8 hours is around 208 hours. That gives you a salary-based figure. The fully loaded cost, once you add visa, accommodation and everything else, is higher, so treat the salary figure as your conservative floor.
Compare the annual cost to what fixing it would cost, and be honest about the payback period. If a process costs you 40,000 a year and the fix costs 15,000 once, that is a straightforward decision. If it costs 4,000 a year, leave it alone and go find a bigger one.
You usually cannot measure them directly, which is why this half gets ignored. Estimate it from what you can see: how often a queue forms, how long people wait before leaving, how close your nearest alternative is. Use your lowest honest guess, and treat the result as a floor rather than a forecast.