AI Adoption Is a Management Problem, Not a Tech One
Quick answer
HBR and McKinsey research shows AI adoption fails not because the technology doesn't work, but because organisations don't redesign workflows, build trust or train staff. 42% of firms scrapped AI initiatives in 2025. Here's what UK SMEs should do instead of buying more tools.
HBR and McKinsey research shows AI adoption fails not because the technology doesn't work, but because organisations don't redesign workflows, build trust or train staff. 42% of firms scrapped AI initiatives in 2025. Here's what UK SMEs should do instead of buying more tools.
I see this all the time with UK SMEs. The director reads about AI on LinkedIn, gets the team a ChatGPT licence or buys some fancy automation tool, and then... nothing changes. Six months later the licence is still being paid for, two people tried it once, and the spreadsheet is still king.
It's not that the AI doesn't work. It does. The problem is that buying a tool and adopting a new way of working are completely different things.
What the research actually says
Harvard Business Review published a piece in July called "The Hidden Realities of AI Adoption", and it nails the problem. They found that most organisations have moved past whether to adopt AI and are now grappling with how. And the "how" is where it all falls apart.
HBR's conclusion is blunt: AI adoption is often framed as a technology challenge, but its success depends just as much on how organisations redesign work, build trust, develop talent, and protect the quality of human judgement.
That's a management problem dressed up as a technology problem.
The numbers back this up. McKinsey's latest State of AI survey found 88% of organisations now use AI in at least one function — but only 39% can attribute any EBIT impact to it. And S&P Global Market Intelligence found that 42% of companies scrapped most of their AI initiatives in 2025, up from 17% the year before. That's not a few pilots not working out. That's a mass cull.
Why it keeps happening
Here's the pattern I've seen dozens of times. A company buys the tool, does a demo, maybe runs a pilot that looks brilliant in a controlled environment. Then they try to roll it out across the business and it hits a wall.
The wall isn't the technology. The wall is:
- Nobody redesigned the workflow. The AI was bolted onto an existing process that was already broken. So it just speeded up the wrong things.
- Nobody trusted it. Staff tried it, got a weird answer once, and went back to doing it manually. Trust is earned, not bought.
- Nobody owned it. There was no clear person responsible for making AI work in that part of the business. So it became nobody's job.
MIT Sloan and BCG's joint research found that organisations combining organisational learning with AI learning were 60–80% more effective. In plain English: the companies that win at AI are the ones that treat it as a change management project, not a software purchase.
What to actually do about it
If you're running an SME and thinking about AI, here are three things I'd do before spending another penny on tools:
- Pick one painful, repeatable workflow. Not "everything." One thing. Maybe it's manual data entry from emails into your CRM. Maybe it's generating the weekly production report. Pick something where you can measure the before and after.
- Redesign the process first. Map out the steps. Work out where the AI fits, where the human checks happen, and what "good enough" looks like. Don't just drop AI on top of a mess.
- Name an owner. One person. Not a committee, not "the team." Someone who is responsible for making this work and has the authority to change things when they don't.
That's it. No grand AI strategy document. No five-year roadmap. Just one process, fixed properly, with someone accountable for it.
Buying AI tools is easy. Making them work is management.
The companies getting value from AI aren't the ones with the biggest budgets or the fanciest tools. They're the ones that treated adoption as an operational challenge — something to be managed, measured and improved. Same as any other business change.
If your AI initiative is stuck, it's probably not the technology that needs fixing. It's the management around it.
References & Further Reading
- Harvard Business Review, The Hidden Realities of AI Adoption (July 2026)
- McKinsey, The State of AI — Latest Survey (2026)
- MIT Sloan Management Review & BCG, AI in Business: From Experimentation to Value
- S&P Global Market Intelligence, 2026 Enterprise AI Initiative Attrition Report
Stuck with AI adoption?
I help UK SMEs figure out where AI actually fits, redesign the workflows around it, and get real value instead of another unused licence. Not a strategy document — practical, hands-on stuff.
Let's have a chatMatty Hatton is the founder of Digital Adaption, an ERP and data consultancy based on the Wirral. He has spent 15 years delivering ERP transformations for manufacturers, including leading the data migration on a £4.5m consolidation of four legacy systems onto a single Infor LN cloud instance for a 220-user group. He holds an MSc in Digital Transformation and IT Strategy from Manchester Metropolitan University and is Microsoft PL-200 certified.
Start with a 30-minute data risk call