- productivity
- microtasks
- ops
The hidden cost of 'I'll just do it myself'
Answering something 'in five minutes' feels free, but each micro-task breaks your focus and eats your morning. How to decide which ones to hand off.
A customer messages you on WhatsApp asking if you're open Saturday. You see it, reply in twenty seconds, carry on. A supplier needs a date confirmed: another minute. The invoice to re-send, the quote only you know how to put together, the reminder you fire off "quickly" before it slips your mind. None of those tasks actually takes five minutes, and together they're eating your morning. This piece is about why "I'll just do it myself" costs far more than it looks like, and how to decide which micro-tasks to hand off without losing control.
The five minutes that are never five minutes
The mistake is timing the task instead of the interruption. Replying "yes, we're open till 2pm" costs twenty seconds. Getting back to what you were doing —the big quote, the call with the bank, the order you were reconciling— costs a lot more.
A now-classic study by Gloria Mark at the University of California, Irvine measured that after an interruption, a person takes on average around 23 minutes to get back to full focus on the task they'd been working on (the study, PDF). You don't have to take the number to the second to recognise the pattern: the expensive part isn't the message, it's the cold start on everything else.
And there are two aggravating factors. First, these tasks never arrive at a good moment: they land exactly when you were deep in the one thing that genuinely needed your head. Second, each one leaves a small mental residue —"mustn't forget to send that"— that you keep carrying even after it's done. Ten micro-tasks a day aren't ten five-minute gaps; they're a whole morning in which you never quite get going.
Do the back-of-the-napkin maths
You don't need a dashboard to see it; a napkin will do. Say you handle fifteen of these micro-tasks a day: confirmations, "how much is it?", forwards, reminders. Say each one robs you of ten real-clock minutes, task plus refocus (sometimes less, often quite a bit more). That's two and a half hours a day.
Even if you go full optimist and halve it, you're still over an hour a day. A whole morning a week you don't spend on what only you can do: closing the big client, going over the numbers, deciding where the business is heading.
That's the real cost, and it isn't only time. The important work —the kind that grows a small business— is precisely the work that needs a long, uninterrupted stretch. And that's exactly the stretch micro-tasks never let you have. It's not that you work few hours; it's that you work them in fragments.
When "doing it yourself" is the right call
Before you automate anything, an uncomfortable bit of honesty: not every micro-task is worth building a system to remove. Automating well costs time upfront, and that time only pays back at volume.
- What happens three times a month. Not worth it. Writing the rules and babysitting the first week will cost you more than doing those three messages by hand for a year. Just answer them and move on.
- What's actually a relationship. A call to a long-standing customer, a "how did the move go?", a thank-you written by hand. That isn't a micro-task in the way; it's your business. Automating it spoils it.
- What costs trust rather than time when it goes wrong. Confirming a date that blocks half your team, applying a discount, accepting a late return. The issue there isn't efficiency, it's who answers for the decision.
Quick rule: if the task repeats and is always the same, get it off your plate. If it changes each time or builds a relationship, keep it.
Which micro-tasks are worth handing off
The good candidates all look alike: they're repetitive, they run on clear rules, and a mistake is fixed with an apology, not an invoice. Yours are probably on this list:
- The answers to the same five questions: opening hours, prices, availability, where you are, "do you have this?".
- Confirmations and reminders for appointments, deliveries or payments.
- Predictable forwards: sending the invoice, passing on the address, the "I'll let you know the moment it ships".
- The follow-ups you forget: that quote you sent five days ago and never heard back on.
- Sorting and prioritising email before you open it, so you see what actually matters first.
Notice that handing off isn't the same as disappearing. The idea isn't for a bot to answer on its own and for you to find out once it's gone sideways. It's the two-speed logic: the agent prepares and proposes, you approve whatever needs your judgement. You approve, they execute.
One warning for when you start: don't pick the micro-task that's easiest to automate, pick the one that interrupts you most. Getting rid of the one that pings you twenty times a day mid-everything is worth far more than automating the one that only shows up on Mondays.
Start by measuring one week
Don't decide from memory which ones slow you down, because memory lies: you remember the task that annoyed you, not the one you repeat without noticing. For five days, keep a dumb tally on your phone: one mark every time you stop to do something "five-minute". Nothing else.
By the end of the week you'll have a list ranked by frequency, and the top of it is exactly where to start. You don't need to automate everything at once: clearing the first two already buys back that long stretch you don't have today.
That logic —you set the rules and approve, the agent handles the repetitive part— is exactly how we build the operations agent at Yaqbot: it prepares quotes, chases the follow-ups and sends the reminders, and leaves the decisions to you. We don't sell technology; what you get back are the hours currently going into five-minute jobs that are never five minutes.
