This week brought government money for AI skills, hard numbers on time saved, and more evidence that small firms are still slower to adopt than the big players.
Government puts £200m behind AI adoption
The UK government announced a £200 million support package to boost AI adoption, built around partnerships with tech companies and trade unions to train workers in AI skills.
For a small firm, the useful part is not the headline figure. It is that training and adoption support now exist that you do not have to fund yourself. Upskilling staff has been one of the quiet costs of bringing AI into a business, and this package is aimed squarely at it.
Before paying for private AI training or consultancy, check what the scheme covers. The details will determine how useful it is, but the direction is clear: adoption support is now policy.
5.2 hours a week, measured
TechRadar reports that AI tools save UK SME workers an average of 5.2 hours per week. The same report notes that a portion of businesses struggle to get any benefit at all.
That split is the real story. The firms saving five hours a week are not running anything exotic. They automate repetitive work — data entry, drafting, scheduling, chasing — with tools that have an obvious return. The firms that struggle tend to start with something too broad and abandon it.
The lesson: pick one repetitive task, automate it, measure the time saved, then move to the next one.
Google makes its pitch to small firms
Google says AI can put small businesses on a more even footing with larger competitors, and points to customer service automation and data analysis as the use cases that pay off first.
Google is selling something here, so read it with that in mind. But the underlying point holds: the free and low-cost tiers of mainstream AI tools now cover a lot of ground. A sole trader has access to the same models as a FTSE company.
Start with the accessible tools for marketing and operations. Move to bespoke automation once you have outgrown them, not before.
The adoption gap is widening
A LinkedIn report found UK SMEs are adopting AI more slowly than larger firms. The reasons given are familiar: limited time, limited budget, and no in-house expertise.
The report's framing matters. SMEs do not need enterprise-scale systems, and trying to buy one is part of why adoption stalls. A small firm gets further by fixing one specific pain point — quoting, invoicing, lead response — than by attempting a broad transformation.
Early adopters are pulling ahead. The gap is not about spend. It is about starting.
What SMEs are asking
Two themes stood out in business owners' questions this week.
Response speed
Business owners are asking why it takes hours to reply to a new lead, how to handle overflow calls without the business feeling impersonal, and which WhatsApp tools come with proper CRM and pipeline features.
The common thread is response time. A lead that waits hours goes elsewhere. This is one of the strongest automation cases there is: an automated first reply within seconds, a human follow-up when it counts. The technology for this is mature and cheap. Most firms just have not set it up.
How much AI do you actually need
The other cluster: when is an AI agent overkill versus a simple automation, when does hiring an AI consultant make sense versus buying into hype, and which operations tools do real work rather than wrapping a chatbot.
The honest answer is that most business tasks need simple automation, not an agent. If the task follows fixed rules — move this file, send this email, update this record — a plain automation is cheaper, faster, and easier to debug. Agents earn their keep when the task requires judgment on messy input. And for anything with real consequences, like accounting, owners are right to insist on human verification. One owner asked specifically for AI accounting with a human check built in. That instinct is correct. Automate the data entry, keep a person on the sign-off.
Kontinuity builds AI automation for small firms. If you want to know which of these actually fits your business, book a call.