Strip away the buzzwords and most small businesses are not asking about artificial intelligence at all. They are asking about the same handful of things they have always been asking about. Getting a quote out faster. Getting paid without three phone calls. Getting home before nine most nights.
The UK Admin Drain Report 2026, a survey of 167 UK small business owner operators published in March 2026 by the automation consultancy HeyBRB and reported in Electrical Times, put a number on that. Across the whole sample, owners reported losing an average of eight hours a week to repetitive admin. The report totals that as 384 hours a year, which assumes a 48 week working year rather than 52, the equivalent of ten working weeks gone to the paperwork around the job rather than the job itself. Among the trades businesses in that sample, 77 percent said they do that admin in the evenings, after the working day is meant to have finished. Nearly half said they do some of it at weekends.
Two things are worth saying plainly about those numbers. The survey was run by an automation consultancy that sells the remedy, with 167 respondents recruited online, not a government dataset. And the trades figures come from a sub-group of roughly twenty two people, which the report's own author says is too small for confident stand-alone analysis. So treat the exact figures as indicative rather than gospel. The shape of it will still be familiar to almost anyone reading this.
The same survey found that invoicing and chasing payment was the task people most wanted off their plate by a wide margin, followed by writing quotes, which for more than a third of all respondents took over twenty minutes each to put together properly.
None of that is a technology problem waiting for artificial intelligence to arrive. It has always been a time problem. What has changed is that some of it can now genuinely be shortened, not by handing the business over to a machine, but by taking the repetitive first draft off a person's plate and leaving the judgement with them.
What is genuinely usable today
A first pass at a quote. Describe a job in a voice note or a few typed lines and get a structured draft back in the standard format you already use, with your prices, not invented ones. You still check it, adjust anything that needs a human eye and send it. What changes is that the twenty minute blank page becomes a two minute edit.
Chasing payment without it feeling personal every time. A short, consistent follow up message, timed and worded sensibly, drafted for you rather than composed from scratch each time you have to do the awkward job of asking someone for money they already owe you.
Turning a messy call note into something usable. A conversation with a customer, a supplier or a site visit, captured properly and turned into a clean note against the job, rather than half remembered three days later when it actually needs to be acted on.
None of these remove the person from the process. They remove the blank page, which is usually where the twenty minutes actually goes.
What is not there yet and should not be pretended otherwise
Pricing a genuinely unusual job, the one that does not match anything you have quoted before, still needs a person who understands the trade. Handling a customer who is upset about something still needs a person who can read the room. Nothing on the market today does either of those well and anyone telling you otherwise is selling you something rather than helping you.
That is not a limitation to apologise for. It is the actual point. The tools worth having are the ones that clear the repetitive layer so the part that genuinely needs you gets more of your attention, not less. A composite picture, drawn from the kind of week described in the survey above rather than any one named business: a one van electrician in Newport spending less of Sunday evening on invoices is not a smaller business afterwards. It is the same business with Sunday evening back.
A second example, away from a van
The same shape shows up away from trades entirely. Take a second composite, an independent hair salon in Cardiff. A salon like that typically spends a chunk of every week on message replies, confirming appointments, answering the same handful of questions about availability and pricing, then chasing the inevitable late cancellation. None of that needs a person's full attention every time and all of it currently gets it, because there has never been an easy way to hand the repetitive half to something else while keeping a real person for the actual booking and the actual relationship.
A tool that drafts consistent, on brand replies to the common questions, leaving genuine or complicated enquiries to a person, is not a technology story. It is the same finding as the trades example above, translated into a different trade. The paperwork changes shape by sector. The pattern underneath it does not.
Sorting the noise from what needs you
One more genuinely usable thing worth naming: a first pass at sorting incoming enquiries by what they actually need. A message asking for opening hours does not need the same attention as a message from someone unhappy about a job and right now both usually land in the same inbox, read in the order they arrive rather than the order they matter. A simple triage layer, reading the message and flagging which ones need a person urgently, does not remove anyone from the conversation. It stops the urgent one sitting behind four routine ones purely because of when it happened to arrive.
None of this is exotic. It is the boring, unglamorous middle of the working day, which is exactly why it has been ignored for so long in favour of flashier applications of the same technology.
Why small and specific beats a big platform
There is a temptation, once any of this starts to look useful, to buy a single large system that promises to handle everything at once, quoting, scheduling, invoicing and customer messages, in one subscription. For a business your size that is usually the wrong order to do things in. A single small tool, aimed at the one task genuinely costing you the most hours, proves itself or does not within a month. A large platform takes months to set up properly, asks everyone to change how they work all at once and is much harder to walk away from if it does not suit you.
Start with the task, not the platform. If invoicing and chasing payment is genuinely where your ten working weeks are going, fix that first, on its own, before touching anything else. That order, one task at a time, built around how you already work, is the automation side of what we do. Most of the value in the survey data above sits in two or three tasks, not twelve. Fixing the two or three that matter beats installing something that touches all twelve badly.
What stays exactly the same
None of the above changes who your customers actually deal with. The person who answers the phone is still the person who answers the phone. The person who turns up to do the job is still the person who turns up. What moves is the invisible layer around those moments, the drafting, the chasing, the sorting, none of which the customer ever sees or cares about, all of which currently eats the evening that should belong to you. That is a genuinely different kind of change from the one most AI coverage implies and it is the one actually worth having. It is also the change most likely to still matter in five years, long after this particular wave of buzzwords has been replaced by the next one.
The evening test
There is a simple way to tell whether any of this is worth pursuing in your own business before spending anything at all. Think back to last Sunday evening, or last weekday evening if Sunday is not when your admin happens. What were you actually doing and was any of it something a first draft, generated in seconds and then checked by you, could have shortened. If the honest answer is yes, that task is where to start. If the honest answer is that last evening was genuinely spent on judgement calls, pricing something unusual, thinking through a decision, then that particular evening was not the problem and the search for a bottleneck needs to look elsewhere.
The real first step
The number worth finding is your own version of the ten working weeks above. Not a national average, your actual week, mapped honestly. The bottleneck audit is built for exactly that and it costs nothing to run.
Once you know where your own hours actually go, deciding whether any of this is worth adopting becomes a much shorter conversation. Get in touch for a free consultation and we will have that conversation properly, starting from your week, not from a sales pitch about ours.