Subscribe to our newsletter to receive business tips, learn about new funding programmes, join upcoming events, take e-learning courses, and more.
Take the first step to successfully starting and growing your business.
Join for freePosted: Wed 12th Aug 2026
Most AI rollouts don't fail on the technology. They fail on the people front.
Tools get bought, training gets booked, and then adoption quietly stalls: anxious staff, quiet resistance, and an expensive tool nobody really uses.
In this practical 30-minute session, Mike Wakeham draws on 20 years of SME leadership and an MSc in Psychology to show why AI adoption really succeeds or fails, and what leaders can do about it.
Whether you're just starting with AI or already rolling it out, you'll leave with a clearer, calmer way to bring it into your business, one that builds confidence rather than fear.
Why AI rollouts lose people, and the human reasons adoption stalls even when the technology is right
How to spot the early warning signs of resistance, anxiety and quiet disengagement
How to bring your team along, building the confidence that turns reluctance into willing use
Michael has spent more than 20 years working inside owner-managed SMEs in recruitment, professional services, healthcare and franchising at the point where businesses have something that works and are figuring out how to build on it.
He sits on the advisory board at Arden University and mentors businesses through the Help to Grow: Management Course. He also works with founders and owner-managers through his practice, Brynley Knight.
Watch more expert webinars
Access a growing collection of expert-led webinars covering marketing, sales, finance, growth and more – ready whenever you are.
Lightly edited for clarity.
Beth: Hello everyone, and welcome to today's Lunch and Learn. My name is Beth, and I'll be your host today.
For those of you attending a Lunch and Learn for the first time, Enterprise Nation is a vibrant community platform for start-ups and small businesses. I'm very pleased to introduce Michael Wakeham, who is a business growth adviser. In this session, Michael will discuss why AI adoption really succeeds or fails, and what leaders can do about this.
If you have any questions throughout the webinar, please post them in the chat or in the Q&A, and we'll do our best to answer them at the end of the session. As always, this session is being recorded, and we will send the recording with follow-up resources later today.
So over to you, Michael.
Michael: Thanks, Beth, for the introduction. Welcome, everybody, to today's Lunch and Learn – thanks for joining.
I think with the title, "How to adopt AI without losing your people," that could probably be taken a couple of different ways. What we're talking about today is the psychological side of losing your people, rather than literally losing your people – so I just wanted to confirm what we're covering.
In today's session, there are six steps that we'll cover, and we'll go through each one in turn. But first, just a bit of an intro about who I am.
I've worked across operations and commercial roles for about 20 years, for various SME businesses, and been through many different change management programmes – whether HR-led or IT-led. So I completely understand the complexities of change management, especially around behaviours.
I'm also conscious that we'll likely have a variety of people on this Lunch and Learn at different stages of the AI adoption process. Some of you probably haven't started yet – maybe curious, maybe thinking about it, or worrying you're being left behind – so very much at the start.
Others will be further through the process, and may have started to embed AI use within their business, or even be looking to scale it. So there will likely be different requirements from the six steps we're covering today.
I'm going to keep it general, so hopefully, if you're just starting out, all six steps will be useful. But if you've progressed further on the journey, there might be one or two steps particularly worth expanding on, with a few takeaways of real interest.
AI adoption, in my experience, is very much a journey. It's something we can probably relate to in certain ways, but it's a very different approach compared with a typical technology rollout.
I suppose just to highlight that – these stats on the screen are from a very recent study by Stanford Digital Economy Lab. They looked at around 51 AI deployments across 41 different companies, focusing on the ones that were successful rollouts, to understand what made them work.
The two stats that stood out for me, which I've put on the screen, are that the choice of model wasn't necessarily the most important factor. The tool being used could be interchangeable – it wasn't the driving factor in whether the rollout was successful.
What they found was that the biggest impact on success was the behavioural side of things – the process, the change management, and the people. Specifically, how people adapted to new tasks and to using AI.
I wanted to share that because many of you have probably read reports and research from tech companies, AI companies, consultancy firms and government agencies on what helps with AI adoption. The vast majority point to the fact that it's a behavioural change as much as a technology change – that's really key.
On the screen, you can see a couple of questions. One simple idea to take away, if you're just starting out, is that rather than asking "what tool should I use?" or "what LLM should I buy licences for?", always think about what problem you're solving first – that's the real starting point with AI adoption.
I've been in many meetings where people think, "I have to use AI because everyone else is using it", but do you actually need it? And if you do, what are you using it for, and what problem is it solving?
Then, as you progress through the journey, what will your team need to do differently by using AI? Does it change work processes, organisational structures, or what they do day to day? Those questions are good takeaways for a simple review within your own business.
Before we get into the six steps, I want to look at what "losing people" actually means from a psychological or human perspective. This might be a reminder to keep an eye out for these signs in your own team, or it could be something you've already spotted, either in yourself or in your team.
The first sign is disengagement. You might find the work still gets done, but the extra involvement some of your team normally shows reduces slightly.
This ties into a related point about people going quiet – sometimes they feel overwhelmed, or unsure whether they should be using AI more or less, and so retreat slightly from what they'd normally do. If you spot that in your team, be aware it could be one of the reasons.
The second point is techno stress, a concept that actually predates AI by about 20 years. It covers five key areas: overload, invasion, complexity, insecurity and uncertainty – a topic detailed enough for its own session.
Generally, a normal technology rollout might trigger one, two or maybe three of these areas, whereas AI has the potential to trigger all of them. You may have heard of related terms like digital burnout or digital overload – it's a similar concept.
The final point is "what am I for?" – which hits at the heart of self-identity. From an employee's perspective, if AI is now doing tasks they've done for five, 10 years or more, what do they actually bring to the table, and what value are they offering, both to the company and in their role? That's really key.
The benefit a small business has compared to a larger one is that you're a lot closer to your people – you might be in the same room, with maybe five, 10 or 20 of you. So you can spot these signs quickly and handle them quickly too.
But if not handled, the impact is probably bigger, because one or two disengaged people in a business of 250 can likely be absorbed. Whereas if you're a team of 10 and two are disengaged, that's a large proportion of your workforce, so a couple of tips there that are worth keeping in mind.
So, moving on to the six steps. We have a slide for each, which we'll run through in turn – they're split into three sections.
As I said, you may be at different stages of your own AI journey, both individually and as a business. So not all of these may be relevant, but you might pick up a couple of ideas that prove useful.
Before you start the journey, there are two key areas: ask – find out what your people know and what they're thinking – and, from your perspective as a business owner or leader, tell them what you're thinking, what you don't know, and what you're looking at AI for.
When you're ready to start, the two steps are to show – go first yourself, and be visibly learning in front of everybody, so they see both the positives and the negatives of AI, and start, which means picking the right thing to point AI at. That links back to the ask step, which we'll cover shortly.
Finally, if you've started embedding AI in your business – whether that's LLMs, agents, or whatever it might be – the next stage is about anchor and share. An anchor means telling people what they're valuable for now, clearly setting out their new role within the business. Share means making it safe to talk about what they're trying, learning and doing.
So those are the six steps, and, as I said, we'll go through each one now.
So, the first step is to ask. This is really important because it gives you the information and guidance for what to do in each of the next five steps.
Before looking at tools or technology, it's about finding out where you are as a business and as an individual – what you know about AI at the moment, and what you're already using it for.
You might ask your team whether they use it personally, and how – is it just an advanced Google search, or are they developing their own automations and agents for their personal life? You could also ask whether they've used it at work, and for what.
This gives you a real picture of where their experience lies. I've been in businesses where the most unlikely person turns out to be the one who's tried the most and can offer guidance to others.
The next helpful question is to get your team thinking about which part of their role, or their week, they'd happily never do again – the administration, or mundane and repetitive tasks they'd gladly hand over to AI. This helps build use cases you can explore in the next section.
The final question to ask is what's worrying them about AI. This gives them a chance to express whether they're worried about falling behind, which might point to a training need, or about their identity – if AI can do their role quicker and better, what are they there for?
It could also be overload: their role might be busy already, and they feel they don't have time to learn AI on top of it. So those are three key questions, and how you approach them will depend on your team, who you'll know better than I do.
If you have a small, open team, you can ask these in a meeting where everyone feels safe contributing honestly, especially about what worries them. If you think it's better done anonymously, you could use a form, collect ideas, or even have people write on Post-it notes around the office – whatever works for your team.
The key thing is making sure you get clear ideas and advice from that exercise.
You could probably do this in the same meeting, but the next step is about you, as the business owner or leader, talking about your own ideas, thoughts and processes. Sometimes a company might feel that saying nothing is fine, but silence isn't neutral – it still gives an indication, and people will form their own minds about where things are headed.
So it's key that, as a business owner, you explain to your team why you're looking at AI. Not just "everyone's using it, so we must too", but what the actual business problem is.
It could be as simple as being a small business that can't compete with larger ones on headcount, so AI helps you do that more effectively, or saves time on administration so your people can focus on client-facing work. It's fine, as the business owner, to say you don't know yet what your business will look like in a year or two – you don't know how this will play out.
However, as with anything in business, you need to progress. You need to look at how AI could benefit your business, just as you might once have implemented a new invoicing system, or invested in machinery that saved time – it's part of the natural evolution of a business.
I think it's revealing to your team that, if you're open and honest about what you see and feel, it gives them the opportunity to do the same. So, although this is two steps, in a small business, you could cover it in one or two meetings, with a genuinely open dialogue with your team.
The next two stages come once you're starting to use AI and running some trials or use cases. The key benefit of you, as the business owner or leader, going first and trialling it with your team, is that you're actively modelling AI use.
There's been quite a lot of research on this – Gallup and also Microsoft have found that when people see their managers, or in this case, the business owner, actively using AI and learning on the job, it gives them more engagement.
They can see that you might be struggling, or that you've had a poor response back from a tool. Rather than concluding "AI is rubbish", you show that you need to iterate, look at the prompts you're using, and check the information involved.
They can see you learning alongside them, which gives them the psychological safety to do the same.
The other point is delegate to an AI enthusiast, which has two sides to it. You might have someone who's genuinely good at AI and also good at training and working with people, which is valuable.
But there's a danger in giving it solely to someone who's really good at AI: it can remove the psychological safety of being able to try and fail, because others may think, "I'm behind, I don't know what I'm talking about."
They might feel they need to put in extra work in the evenings, which creates worry and a kind of techno stress creeping into their mindset. So be careful about handing it to the one person who's brilliant at AI, where all others see is the highlight reel of success, not the trial and error behind it – that's really important.
So, the next stage is to start. As I said, you might be at different stages with AI – you may have progressed to using agents, or built your own AI system with all your client data that you can search against to help with proposals, or you might be at the very start, just using LLMs to ask questions or help draft emails.
If you're at that first stage, just start – but start with something low stakes, so if it goes wrong, it doesn't affect anything externally with clients.
It's also easier to judge the output when it's low stakes, so you can quickly spot when something isn't quite right, and build your judgement from there. It's really about picking one or two cases to start with, ideally from what your team said earlier, they'd happily hand over.
To give a basic example, I worked with a company last year that was just starting out. Their first use case was around incoming files, which used to take someone about 20 minutes to sort into the right folders, save in the right areas, and duplicate related documents.
They created automations within their existing Microsoft Copilot platform that did all of that automatically. It was a very easy thing to do, and it went from taking 20 minutes to being automatic – they still checked it, but everything was internal, so if something did go into the wrong folder, it would have been spotted and sorted quickly. Very low stakes.
What that does is breed confidence and gives you a platform to build on.
The next section is about once you've embedded AI, and it's the next phase of development within your people. This step is really key, because anchor means that once things have changed – if AI is now doing certain tasks and has taken on some duties or responsibilities from your team – you anchor that person's role and value to the business as it now looks.
This is a big topic, touching on job description redesign and organisational process redesign, but it's probably the area that gets stuck the most. That's because you've changed the tool and the process, but not the person.
So, say the person who used to do all your client proposals or bids is now largely supported by AI – what is that person doing now? It could be that they're the ones having the client meetings, adding judgement and context to the proposals, and having more time to follow up or work with the client.
Another example: I was recently on an HR workshop discussing AI's effect on employee wellbeing, where one company had put all their client files and knowledge into an internal AI system they could search against. This made their client work and meeting research a lot quicker and more intuitive.
It had no effect on their headcount – what it meant was that the people using it had more time to spend with clients, and better conversations, adding a lot more value to the business overall. So this anchoring point is really key to adjusting how your people's self-identity and role are now anchored within the business.
The final step, which is really important, is to share – share the wins, the losses, and what's happening now in your business. You may have an AI policy outlining what systems and data can be used, but it's also about setting a standard.
As you go along the journey, you might use AI to draft certain documents, like end-of-month reports. It's important to set the standard for what that means, so that if a team member sends you what's known as "AI slop", rather than just reacting negatively, you talk about it.
What is your standard? If you're going to create a report using AI, that's absolutely fine – but you need to check it, add your judgement, use your experience, and be able to defend or discuss what you've sent before it reaches me. It's not just about saving time by putting a request into an LLM and letting it produce something.
So, setting standards is important. Also include the failures as well as the successes – if someone's tried something that hasn't worked, that's a great learning experience, as it might save someone else from trying the same thing.
I think the key is making this a regular thing – a weekly or monthly meeting where you share AI successes, use cases that have worked well, and anything others can learn from. One point I haven't touched on yet, because it isn't strictly a technology or governance topic, is that when your team are experimenting and learning, you need the right structure in place to let them do so safely.
That includes your data protection and the systems you're using, especially around business or client-confidential information – some clients may now have policies that restrict AI use involving their data. Just make sure you have those guardrails in place while your team are learning and developing.
So, those are the six steps. Hopefully, that's given you some insight into the behavioural change that needs to happen – this has been a top-level overview, and there's a lot more depth within it.
But the key thing, in my experience, is that you're not protecting people from AI, or guarding against its use. What you're trying to do is make them more valuable, both to themselves and to your business, by utilising AI and working with it – that's really key.
So I think we've got two or three minutes. My email address is on there as well, so I'm happy to take any queries – if we don't get to your question, please email me, and I'll get back to you as soon as possible. But if there are any questions, we've got a couple of minutes to fire through them.
Beth: Yeah, that's right. Thank you, Michael. That was really, really informative.
So, what are the early warning signs that a team is becoming resistant or anxious about AI, even if nobody is openly saying so?
Michael: Yeah, that's a really important question, because sometimes even if you ask them directly what they're worried about, you might just get "we're all fine, everything's good." But what you might find is that they're not using it as much, or using it without telling you.
You might notice they're less engaged – when you start talking about AI, they may shut off a little, or not be as open as usual. There could be a few reasons for that: one is that they might not feel confident, worrying that speaking up makes them look less informed or behind, so it feels safer to say nothing than to ask.
It could also be that they're worried about AI progressing, and their part in that, might eventually cost them their job, because they've helped train it in some way. So it's an important step for you, as the business owner or leader, to say what you're seeing and experiencing, and share your thought process, to make sure things stay on track.
This isn't unique to AI, either – in other change management programmes, people often say everything's fine because they're worried about what might happen if things do move forward.
Beth: Sure, okay. So, kind of related to that – how can a leader distinguish between genuine concerns about AI and simple reluctance to change?
Michael: Yeah, another good question. Anyone who's been through change management knows that changing people's behaviours is one of the most difficult things, because if someone's been doing a role for many years, getting them to see why or how they should change is hard – on the other side of that change could be something they're not sure of.
Uncertainty is one of the biggest barriers to people wanting to change, which is why, as I mentioned earlier, silence isn't neutral. If you're not addressing or talking about it, people form their own view.
So if they're not engaging, it could be that they don't want it because they've heard that if they help train the AI, they'll lose their job. Talking through what they might be thinking, even if they're not raising it themselves, and addressing it that way, really does help from a leadership perspective.
Beth: Yeah, absolutely, that makes sense. Thank you. Oh, brilliant, okay – so that brings us right on time.
For those of you who had questions we didn't get to, please do connect with Michael. His email is up on the screen, and I've also put his Enterprise Nation profile and LinkedIn profile in the chat, so please do feel free to reach out.
So, thank you, everyone, for joining today's webinar – thank you, Michael. People are saying in the chat that this was really informative, a great session.
And as I mentioned earlier, we'll be sending the recording and some resources in the follow-up email today. So thank you again, and thank you, Michael.
Michael: Thank you. See you soon. Take care. Bye, everyone.
Unlock more on-demand sessions designed to help you sharpen your skills, grow your business and stay one step ahead. Find more Lunch and Learn webinars
