How to Upskill Your Team in AI (Without the Overwhelm)

Most AI training for teams doesn't work. And it's not because your team isn't smart enough — it's because the training is generic, tool-obsessed, and has zero follow-through.

In this episode, I break down the 4-step framework I use with my consulting clients to actually upskill teams in AI — without the overwhelm, without the tool fatigue, and without everyone nodding politely then going back to business as usual.

You'll hear:

• Why most team AI training fails (and the 3 reasons nobody talks about)
• How to pick your "AI Champion" — the one person to start with
• The 3 buckets of team AI training (and why most businesses start in the wrong one)
• How to create an AI policy that fits on one page
• What to actually measure (hint: it's not just "hours saved")
• A real client story of turning 6 hours/week into 20 minutes
• Plus the AI Confession Booth and your weekly Not Another To-Do

Show Notes / Links

• Book a consulting session → wrightmode.com/consulting
• AI Dream Team Workshop → wrightmode.com/aidreamteam
• Wright Mode Membership → wrightmode.com/membership
• Follow on Instagram → @wright_mode

• Want to work together? Check out my offerings at Wright Mode
• Connect with me on Instagram → @wright_mode

Transcript

Speaker 2 0:23

Welcome back to not another AI girl. I hope you're having a good one. over the weekend I was away for touch footy state championships and I am a little sore, a little tired. My voice sounds like I have been screaming. All weekend, which is exactly what has happened. so hopefully my voice holds out for this episode anyways. Today we're talking about something that I get asked constantly, and it's one of those topics where I see businesses waste so much time and money because they approach it completely wrong. and so what we're talking about is how to upskill your team in ai, not in the like, let's all sit through a webinar. Nod politely and try not to fall asleep way, in a way that sticks and saves real time and doesn't send everyone into a spiral of overwhelm or, feels like a waste of time. So let me paint you a picture. You are a founder. You know that AI is important. You've probably been using chat GBT yourself. Maybe you've even built a few things with it. And now your team is looking around and you're thinking. We need to be across this, like as an organisation or as a team. So you do what feels logical. So you book a group training and everyone logs into a Zoom. Most AI training for teams is, and I'm gonna say this with love, genuinely terrible. And today I'm gonna give you the four step framework I used with my consulting clients and my corporate training clients. This includes one question I ask before we do any of the training, and let's get into it. I'm also gonna walk you through a real client example, where their VA was spending. Hours a week on something that now takes about 20 minutes and I'll break down exactly what we did. And there's also the mistakes that I see founders make that actually makes their team more resistant to ai. It is something that sounds like a really good idea on paper, but it backfires almost every time. So before I give you the framework, I wanna quickly unpack why most team ai, like why most AI training for teams doesn't work. so the reason one is that it's tool first. It's not problem fault first. So. When someone books in for a training and the whole session, he is like, here's how to use chat GPT, or Here's how to write better prompts. That's fine. Like as far as it goes, but the team walks away thinking like, cool, I can do that without really thinking about any of their actual workflows or problems in their business. So it's like teaching someone how to use a hammer without showing them. what actually needs to be built. So they've got the skill, but they just don't understand where they need to apply it or where they want to apply it. And so reason two is that it's one size fits all. So not everyone on your team needs to learn the same thing or Yeah, needs to learn how to use AI in the same way. So your ops person and your content person and your client manager. All have completely different AI opportunities, but generic training kind of treats everyone the same and also like the same learner, and then everyone's frustrated because the training was either too basic or too advanced for what they actually want to do, and they're struggling to see how it relates to their role. So reason three, there's no implementation window. So, you know, they might watch hours of content go back to their desks and within a week it's like it never happened. And this is the same problem with courses that have like multiple modules. By the time you've watched everything, you haven't implemented anything. And I firmly believe that learning without building is just entertainment. So. If that's what not to do, let's get into what actually works. So this is the four step framework that I use with every single team that I train. So number one, and this is like one of my favourites. So pick an AI champion. And this surprises most people. So don't, you don't need to train everyone at once. And I know that can kind of feel counterintuitive. You know, you are like, I want the whole team across this. I want us to be AI first as, a business, an organisation, and I get that, but hear me out because this is that mistake that I mentioned. So the one that actually makes teams more resistant to ai, it's sending everyone to the same training at the same time. Because what happens is if you put like eight or 10 people in a room together. You might have two people that are super keen and they're actually the ones who've already been playing with chart GPT, kind of pushing the envelope on what they're getting it to do. And then you've got like a, I don't know, say six people out of that 10. Who are like, cautiously curious, maybe a bit nervous, you know, they're like, is this gonna take my job? and then you've got two who are actively sceptical. So they're, they're sitting in that room with their arms crossed, thinking this is just another, you know. Shitty training that I'm forced to sit in on. I hope the lunche, it's good. So when you train them all together, the keen ones get held back, by that pace. Those who are like curious but cautious, can often feel overwhelmed because the keen ones are asking all these advanced questions and they're like, whoa. what are they talking about? And then the sceptical ones feel validated because the training was too generic to solve their actual problems. They're like, C AI is overhyped anyways, so nobody wins in this situation. So instead, it's often useful to start with one person, and that could be your AI champion. And so how do you identify them? You are not looking for technical skill here. You're looking for curiosity. So it is the person who's already like using chat GPTA lot, who's maybe looking at how to automate things. It's the one who messages, you or the Slack channel saying, Hey, I found this tool. Can I try it? They don't need to be, you know, your most senior person. They just need to be genuinely interested in this stuff. And then, so what you wanna do is train that person deep so you get them. Like skilled. you give them a real win, something tangible that saves them time or makes their work better. And then you let that win, do the convincing for the rest of the team. so when I'm training. Bigger teams. this is kind of how I like to do it. instead of training the whole team, we might start with that ops person, and they, because they understand the workflows more than, I can help them upskill with AI and automations, but they understand, like those pain points where the bottlenecks are. So when you combine. That internal knowledge with this AI training, they can go and apply it to their actual workflows. and, you know, they can get some quick wins on the board there and then be like to the rest of the team, Hey, here's what I've like. Started to do with AI and automations. and that's how you get that buy-in for the people who have previously been resistant. and so that's the AI champion model. So it is kind of like peer led adoption. and that always is gonna be, you know, like top down like management kind of mandates every single time. And here's that question, the one that I ask every founder before we do any team training, and most of them get wrong. So they always ask me like, what, what AI tool should I, should the team, or should I be learning? But the right question is like, what process are you trying to improve? I want you to start with problem first or and strategy first. Always the tool comes second. that's just a non-negotiable, okay? Now once you've got your champion and you know the problems that you wanna solve, what do you actually wanna train them on first? Because the temptation is to just be like, here's everything. That is just gonna lead people to feel overwhelmed. So step two, what to train first and what to ignore. So the temptation, when you start training someone in AI is to go right? You need to learn chat, GPT and Claude and Mid Journey and make.com or N eight N and Notion AI in this tool and that tool. And that's how you create like tool fatigue and overwhelm and zero adoption because more tools isn't the answer. So the right tools used properly for the right problem. That's the answer and that can be quite small. something like if you are working in already in Notion already and you use something like Notion ai, then like that can cover most of what you need it to do. You also have the option to choose between Claude Chat, GPT, and Gemini within Notion ai. When I'm looking at the tools, here's how I think about it. I use what's called three buckets of like the team AI training. So bucket one daily AI use. So this is, you know how to use a chat based AI properly. So that's like chat GPT, Claude Gemini in your browser or on your desktop app. Chatting away with it. So pick one, it doesn't really matter yet. This is like about setting up custom instructions, setting tone and voice, using it as a thinking partner instead of just like a vending machine. knowing how to have proper back and forth conversations with ai, and trying to, you know, extract the best. Output out

9:54

of it.

Speaker 2 9:54

And for most teams, that's where you should start. And for most people, even the ones who are thinking like, I already use AI heaps, you are barely scratching the surface here. Like the amount of times people are like, oh yeah, I'm like so deep in chat GBT or clawed, and then I'll do a training session, and they're like, oh my God, I had no idea that any of this existed or that you could do this. Just within Chat or Claude. and then there's bucket two, so that's looking at our workflow automation. So that's when we start to connect tools. So building simple automate automations, like using something like N eight N or make.com or even a PR if that is, you know, the tool perhaps that people have already subscribed to. and this helps reduce. Repetitive tasks that eat up hours every week, and there's so much opportunity for small business, particularly to use these tools to cut down on that admin time. So that could be, you know, things like when a new client inquiry comes in and it automatically creates a project folder. or if a new client signs on and it's like sending them their welcome email with an onboarding pack, a Google Drive folder is created for their project. Add them to the CRM, notify the team on Slack, that kind of thing. that's where that AI champion comes in. that's where real time savings show up. but it only works if bucket one is solid first. So getting that kind of foundational use of ai, and because most people across your organisation are gonna need to at least have that base level. And then some people will need to have bucket two of like workflow automations, building that stuff. And bucket three is custom AI build. So this is like the advanced stuff. This is, you know, using, vibe coding tools to build internal tools. It could be creating custom GPTs, interactive dashboards if you're doing like client work that you can present these really nice, dashboards to clients. yeah, AI powered client experiences and this is where you can start to get genuinely creative, and start doing things that your competitors aren't. and, but what I will say here is you don't start in bucket three, so you build up to this. And most teams only need bucket one plus some people in bucket two. And you know, I have seen AI trainers try to sell all three at once because it looks more impressive. It's like, oh, here's everything that you can do. but what happens is the team gets overwhelmed. they don't really grasp AI and automation's properly. And then, so, you know, the ROI on that is very small because the actual organisational uptake is small. All right. So imagine you've picked your champion, you know which buckets you wanna focus on. here's something that I see a lot of people skip. And I reckon it's the bit that separates the teams who actually adopt AI from the ones who just get the training and then, things just go back to the way they work. So step three, we need to set up AI guardrails. So you need an AI use policy and before you tune out, because that sounds really corporate and boring, it doesn't have to be, it doesn't have to be a 47 page document that nobody reads, okay? A one page is fine. The point here is to be clear and so that everyone is on the same page, because here's what happens without it either your team doesn't use AI at all because they're scared of getting it wrong. And into trouble, or they don't know what's allowed, so they play it safe and they just don't touch it. Or the opposite. And this is called like Shadow AI use. And they use it for everything with zero oversight. there's client data going into public AI tools. They don't even understand the privacy policy. It's AI generated emails going out without anyone checking them first. first draughts with AI becoming final draughts because nobody reviewed them or. People are creating more work for those reviewers because they're like, I'll just throw this into chat, GPT suite, that's done. But then that second line of reviewing takes so much longer because, you know, like AI content can often be very long and very verbose. So it's creating more work for that level. So neither is great. So here's my simple framework for a team AI policy. It's just four things. So you need to define number one, what's okay to use AI for? Is it first draughts? Is it brainstorming? Summarising meetings? Is it data analysis, research, internal comms. So thinking about like the stuff where AI genuinely helps and the stakes are low, the risks are low here. Okay. Then defining number two, which is what needs human review. So is that anything client facing? Anything financial, anything legal, anything personal. So AI can help create it, but a human checks it before it goes out anywhere. And three, what is off limits? So sensitive client data into public AI tools. Confidential personal and business information. Personal details. So most chat based AI tools are not secure by default. So your team needs to understand what not to put in there so that they don't fall into, you know, either not using it all'cause they have no idea. Or using it way too much and, you know, breaching privacy because they have no idea how any of this stuff works. And number four is how to disclose when and how do you tell clients that AI was involved? So this is becoming more important as. Get savvyer about AI use. You know, people are assuming that you are using ai, whether you are or you aren't. And so having a simple, clear position on this builds trust. And if someone, if a client asks you, you can be like, yeah, look, here's our policy. Here's what we use it for. Here's what we don't use it for. So that's it. That's just the four things. One page give, it gives your team like clear boundaries and an outline of AI use in your organisation, and they'll feel safe. So I had a client which had a small service based business, and their VA or OBM was spending about six hours a week putting together like. Client reports, information, that sort of thing. So pulling data from different places, formatting it, sending it out every week, same process, just depending on clients, et cetera. So that was a huge bottleneck for them. and that was the process that they identified or we identified that they wanted to improve. So. I trained their, like their VA as their AI champion. and so that was, you know, getting her skilled in bucket one, like chat GPT. They were already using it, but they didn't know how to use it to its full capabilities. And then we focused on bucket two, which is that workflow automation piece. So, I taught her how to build an automation. to help with that client data, like processing things, using N eight N to pull that data format, that that report, using a template and sending a draught for review. So the VA's job went from like, I don't know, about six hours a week, down to 20 minutes just reviewing this before sending it off to, her manager for review. And also what happens here is that reports become more consistent because, you know, things like the formatting are automated, file saving, like all of that is, the same every single time. But here's the important bit to, you know, setting up guardrails. So in that whole process. The IS reports still had a human in the loop to review before it went to the client. and, you know, data source sources were internal, not going through a public AI tool. and. So that's kind of the whole picture. So focusing on an AI champion training from the right buckets, always starting at bucket one, and then if you need to build to bucket two, and then maybe looking at three if you have someone who can take that and run with it and guardrails. That then creates real measurable results. And that brings me to step four. How do you know if any of this is actually working? Because if you can't measure it, it's just vibes and vibes don't pay the bills or keep the lights on. So the problem with measuring AI IROI is that founders expect instant transformation and they also forget where they started from. They invest in training and they're like, okay, cool. So when do we see like the 10 x in productivity? And as much as I wish it did, it doesn't work like that. So the real wins from AI are cumulative. So they build over time and a lot of them. Can feel like they're invisible at first, like reduced decision fatigue. better consistency or your team spending their energy on high value work instead of like that low value admin that drains their soul. So here's what I actually measure when I'm working with teams. So, number one, time saved on specific tasks. So not just like vague productivity, so specific measurable hours. And even in my consults I have at the bottom, like, here's, you know, how to measure your ROI. Here's where we've kind of identified how many hours you're spending or the output, and here's where we wanna get it to. So this becomes specific and measurable. because then that's real and it's tangible and you know, you can even put a dollar figure on that if you want to. So it, the key here is it needs to be specific. Don't try to measure overall productivity. Measure it like task by task before and after. So number two is quality and consistency improvement. So are the outputs more consistent? Are there fewer human errors, better first draughts? Are there things that used to require three rounds of revisions now only needing one? This one's harder to quantify, but easier to feel, I guess. Number three, team confidence. So this one's a bit harder to pin down. And it's also the leading indicator, like are people actually using ai? Are they bringing ideas to the table? Are they saying things like, Hey, we could automate this, or, I tried this in Claude and it worked really well. if your team is voluntarily experimenting, that's a really, really good sign because that means that they feel confident enough to take that training and run with it. Not because they're just like following instructions, but because now they're kind of looking at things differently and looking for those AI opportunities. And number four is new capability. So can your team now do things they literally couldn't do before? Can they build like a super specific tool that solves a specific problem in your business? Can they create a client dashboard for using ai? Or automated a workflow potentially. You know, you now have space to offer a new service. and I always say, don't measure AI by what it replaces. Measure it by what it enables. So the biggest wins I've seen aren't actually like, oh, we've saved three hours. It's like, we can now offer this thing to our clients that we never could before because we just didn't have the time or the space to do that. So that's where the real ROI lives. That was a meat episode. That's the framework. but before I do the recap, I just wanna get a bit real with you. So, here's why I think this matters so much. And it's not just in a, like save a few hours kind of way. So the businesses that figure. Our team AI adoption in 2026 are gonna have a genuine competitive advantage, not because AI is magic, not because they've got access to, you know, some super amazing AI tool that's just dropped. But because most businesses won't do the work to implement it properly. So everyone is dabbling now like it is 2026. everyone's got a subscription to ai, but almost nobody is doing the actual work of like picking the right person, training them strategically, setting up guardrails and measuring the results.'cause that shiny AI tool that you saw on LinkedIn is very hard to resist compared to like. That. So the gap between having AI and actually using ai, that is where the opportunity is. and your competitors are not going to out I you with better tools. They're going to out ai you with better trained people using those exact same tools. So if you are listening to this and. As a founder thinking, okay, this feels big. It is, but you don't need to overhaul everything overnight. Okay? You just focus on one person who has shown an interest in this stuff. Focus on one problem and then get that first win and then you build from there. Like that is the whole kind of strategy. The four steps. Step one, pick your AI champion. Don't train everyone at once. find the curious person on your team and like. Give them the opportunity to go deep with this staff and then get them a win and then let that wind spread. Step two, train in the right bucket. So daily AI use first. Yes. Then workflow automation, and then custom builds in that order. Don't skip ahead. Step three. Set. Simple guardrails, one page. That's all it needs to be. What's okay? What needs review? What's off limits? How to disclose. Give your team like a clear picture of what is and isn't. Okay, so they feel safe experimenting. And step four, measure time, confidence and capability. Not just vibes, specific tasks, specific hours. And watch for the leading indicator. Is your team voluntarily bringing you ideas, like coming to you being like, Hey. You know, I've created this, or I wanna create this with ai. So that's the framework that I use when I work with teams, whether it's through the AI Dream Team workshop or through consulting or corporate training. So if you are a founder listening to this thinking, I need this for my team, but I genuinely have no idea where to start. That's literally what I do. So we can do it as a team workshop where I come in and I would train your AI champion or a small team, or we can start with a consulting session where. We map out what makes sense for your business and then look at, you know, identifying that AI champion and what the training needs to focus on. I will drop the link to my consulting and website in the show notes, and we can have a chat about what makes sense for you and if you are the person on the team. Who's been tapped as the AI person, or you kind of self-appointed yourself as the AI person. That's amazing. so yes, take that opportunity and run with it or ask for that opportunity. Present some ideas to, your manager and be like, Hey. Or I wanna do this training, something like that. take this opportunity and run with it because you will become very, very valuable within your organisation. all right, onto the AI confession booth. When I first started helping teams with ai, I definitely made that mistake I told you not to make, which was like, you don't train the whole team at once. one big session. So that's, I guess how I've learned what works and what doesn't. So you need to make sure that you kind of identify an AI champion. You can train everyone from Bucket one, like because we want people using AI daily now to Elevate and make them more efficient in their work. but you know, when we start to look at bucket two and bucket three, not everyone needs that training. Not everyone wants that training. and that is how you're gonna kind of get people to tune out and be like, I don't want, I don't wanna touch ai. and that's, you know, where that adoption rate kind of falls off. so that's how I started looking at this through the lens of like, let's identify an AI champion, get that one person a win first, and then, you know, that can like trickle out, And then, you know, it's not me just coming in, being like, you know, everyone's using AI now. It's like you have someone inside being like, here's what we can do, here's what I've done. That's made my role easier because that enthusiasm is contagious. Nice and simple. I want you to, if you're a founder with a small team listening to this, identify your AI champion. Or if you're a solo founder, maybe it's your VA or OBM, even if you just have them for, 10 hours a month. Or a contractor you work with regularly. and ask yourself, who's the most curious, who's already playing with ai? who gets frustrated by repetitive tasks and he is always looking for that better way. That's your person. and once you know who that person is, send them this episode so they can have a listen to it and get some ideas to. because the best thing as a founder that you can do too is not try and learn it all yourself. So it is. Trying to find that right person on your team and give them the support to run with it. You just need to be excited and kind of, Allow for this AI adoption, but then get someone who will be that AI champion, that AI operator for you. All right? That's the episode. If you're a founder who knows AI Matters, but doesn't know how to get your team across the line, you're not behind, you just need a better plan then. All right, everyone, watch this AI webinar, and if you wanna help building that plan, I'm right here. reach out, slide into my dms, and we can do this properly. So your team actually adopts ai. Oh, all right. That was a big episode, my voice held up. I'm stoked and I'll see you on the next episode. I.