Introduction

My name is Penny Martelenghi. I'm the Customer Success Lead at MindStone and what that means is I work with a lot of businesses that are figuring out what on earth are they going to do with AI.

They know they need to do something but they're not sure quite what to do and so quite often we have conversations with them about their strategy and what does that look like.

And I thought what I would do today is really bring it home as to some of the things that we're seeing in businesses that are really storming in their AI transformation journey and share with you some things we'll see as to why that might be but then make it practical so mind stone our ethos is to always try to be practical so there's going to be some practical

take -home actions and thought -provoking things that you can do as a result of this talk as well

Why Many AI Transformations Stall

so if that's interesting say to you let's let's get into it okay everyone knows they need to do aero transformation everyone's aware it's happening but actually very few businesses are truly doing it and what we tend to see is the initiative might be started there's this

excitement and then it's doors it's doors we've given a tool nothing's happening we've given our team access to come to my level to generalize to forward code the latest models and you know there's there's activity but there's not change as a business we're not seeing a change

The Three Components: Toolset, Skillset, and Permission

and that's because in AI transformation we've got three components you've got one set two set and skin set and quite often businesses are starting with a two set they're going there you go guys have fun but actually it's not about that and what happens is the human non -operation side

of transformation it really gets left behind it's not a technical transformation the tech is great we love the tech but actually AI adoption and transformation is about people it's about the human cyber transformation and business has been doing transformation in the way

that they want their organisations for many years and this is a bigger transformation project.

Mindset Shift: AI Isn’t a Better Search Engine

The mindset shift is so important. It's not about plugging in a new tool and using it like a better Google search.

Hands up, who's done it? I've got, you know, Chat2VT on my phone. Instead of opening up Google Google, I'm just going to ask you a question. I've done it in the past.

It's not a search engine. AI is not a search engine. If you're using it like a search engine, then you need to stop them, of course. Google is great for a search engine. Can you handle using Google? That's absolutely fine. But you need to think about what is it you're using this tool for.

AI is powerful. It can help you achieve different ways of working and productivity. It is not with Maze for Doom. And what do we mean by different ways of working?

A Simple Starting Point: Capture and Use Meeting Knowledge

Well, one of the most common use cases I think people use will be capturing meetings. So everyone should be, one of the practical things you should be doing is recording all your calls.

All those transcripts, meeting follow -ups, we don't need to do that manually anymore.

In a call now when I'm with a customer and I'm talking to them, I don't have to worry about taking notes. That mental load is gone what does that mean it means with my customer i'm having a much better quality conversation

because i am being a hundred percent present with that this time i'm not thinking well that was really a point i need to say no to that because i've got an agent that's transcribing that afterwards what does that mean i can interrogate that and have a conversation with my transcript

actually i think these are the three most important points to get others but actually put on a persona and think differently about that imagine you were imagine you are this kind of persona help me help me interrogate this what should i be recommending this is what i think what should i be in the conversation we interrogate that that's a basic use case but

really thinking about work pressure is not about just those easy use cases everyone should be doing that that's something that people can do it's very simple but actually you should be thinking

Work Backwards from Outcomes (Not Tools)

what are the outcomes I want to achieve? And then asking AI, hi, can you help me achieve this outcome? What could I do? How could I get there?

We are limited by our own experiences of the way that we've worked to date, which is a really hard concept to get your head around, particularly in a short -term in a talk.

But actually what we need to be doing is thinking about what is it we want to achieve and then asking AI, help me work through what the different ways I can achieve that that without limits.

Because we are really now without limits using AI. There's so much more that we can do.

Example: Rethinking Sales Development with AI Research

So to give you an example. Too many examples, but I'm going to pick one.

Imagine you are running a sales function. And traditionally, you'd hire an SDR to come on and do the calls for you. You've got an outcome, you've got a sales target you want to achieve.

You've got some input metrics about best practices and processes you've time, you can create your project with that knowledge, have a conversation and say, to achieve this outcome, do some deep research, that's the takeaway again, do some deep research and help me think differently about this problem. What are the different ways I could achieve that outcome?

And it could be that because of your sector and the research that it does for you and I, and there's some great research tools that you can use, PubLexity, Gemini, it will come up with some different ways and different campaign tactics that you might not have thought about before because traditionally we've either SDR or done that up and calling. One example.

Creating Space to Learn: Time, Friction, and Leadership Permission

The biggest thing though I see is not that people aren't willing to try and have a go at doing something differently or trying to fit new space. The BAU work always, always, always is there and it's that friction.

When you need to learn something new, change the way you work, you have to invest time. I know I've been there.

Every morning now I open my laptop and i'm going to show you the tool that i use it's called rebel we've created our own operating system out here which is still a milestone our company culture as you might expect

working in our business we've been permission to lean in and use that guy as much as we need to to make the time and to fail and to succeed and to fail again and fail again and fail again and then succeed when the senior leadership team and organization and managers explicitly gives gives permission and says, I would like you to time box, ignore the BA in work, I'd like

you to time box and spend your time testing and failing and learning and exploring and experimenting, that is when you'll start to see activity.

Permission to take the time to learn is so important.

Make Experimentation Visible from the Top

So there's lots of different ways you can do that, but ultimately creating that visibility of your executive leadership team doing that is one of the most important ways if you're having a company meeting and the ceo is talking about the fact that

yesterday morning during working hours they tried an experiment and it failed but they couldn't idea they think they're going to do something that's going to make a big difference to the way that they operate to the way that they do their pitches perhaps that is a core signal sign that the business is ready to take that step so condition is really really important and actually

actually quite often seen as the most important part of the reason why companies don't progress and their AI programs stall.

So it's this combination. It is skill set, understanding if you do things differently. And to get better, you need some training. Tool sets.

There's a lot out there.

But permission, it's the combination of those three things together that really makes that momentum in an organisation in order to progress and do new things with AI.

A Practical Demo: Using an Internal AI Agent (Rebel)

So, there's some thoughts. I'm now going to show you how I use AI, and I didn't want to run on the internet because it's a bit long today, so here's one I made earlier, quite literally, but welcome to Rebel.

This is Mindstone's AI agent of the project as and it has a database, quite like any LLM. There is a box at the bottom that you can type in.

I had a conversation this morning, this is what I would say if I could guarantee the internet today, and I said, I'm at the meetup, I'm going to be talking about AI transformation and the challenges that teams have. That would suggest three actionable things that attendees can do when they walk away. And I would like you to present it back to me in an HTML.

The way that Rebel works, it's got a memory, so there's a company -wide memory, so it's got a personal memory about me, things that I do. I've connected it to different tools that I use, so MCPs, I like to avoid jargon, but different tools connected behind the scenes, and it's got some inbuilt skills that we developed that enabled us to present information back.

So I had this conversation and said, please can you create this output, and this was all voice to text as well. I didn't type, it was just a conversation.

Why Voice Input and Contextual Memory Matter

Voice input is very, very important. And so what did we all do? He created this for me.

This was the first point. So we'll look at this together, and then I'm going to tell you what I said to it next.

So with that single prompt, with the context that it has about the business, he knew what I was going to talk about. This is what he came up with.

Three Actionable Steps for Human-Centered AI Adoption

So three practical next steps to be human -centered AI transformation.

The next move is not to buy new tools. It's make the human implications visible create one real behavioral shift and then a rhythm that builds adoption so it came up with three ideas map the work really lean in think about the tools

pick a team work with them understand the tasks that they're doing so understand their workflow first before you do anything and you really need to lean in and understand and highlight when ai changes decision making ownership or professional identity not just productivity and that's really important part.

Everyone will go through a bit of an identity crisis when they start to use AI. What will my role look like in four months? Five months? Will my role exist next year?

Will I be needed? What will I do? It happens very naturally. But actually what it gives you as a chance to have a conversation

and think about where is it that actually I would like my team to spend their time Customer success. Where would I like the team to spend their time? Having more

customer conversations. Definitely. Customer first. Customer focus.

Making that time.

So I work with another colleague that's based in the US. We have 42 skills that we use in order to do our role. Skills that we both share using my book.

As a result of those 42 skills, we spend more time with our customers, talking to our customers and don't do some of the manual work that's taking days, hours. So that's one thing.

The second thing, it then said run one live experiment. So choose a real business problem, get people together, talk about it, think it through, four to five minute session, end up by capturing three things, what worked, what didn't work.

Why this matters, you're going to be demonstrated that you can work collaboratively still. Just because AI exists doesn't mean people can't exist together so you're working through a problem collaboratively together

and thinking about those moments of creating trust confidence understanding what work for them might look like so i like that one i think it's really important just because you're using ai individually doesn't mean you can't be using ai together and a great way to do that is to do brainstorming you can put a whiteboard you go old school and post it notes around the topic

take a picture and use that as an input into an ai project with your team that's a really great way

doing things the third thing was create a weekly human adoption ritual 30 days to grow new habits but actually it's coming back to something again and again and again that's important so making that time could be 15 minutes it could be you already have a time that you come together as a business where you change the focus of that session and it's going to be this is our ai

learning time together or half an hour of this session is going to be talking about ai use cases where we failed and where we succeeded. So, that will produce that.

Iterate the Output: Give Feedback and Simplify

And we thought, do you know what, that was okay. It was okay, but it wasn't great. So we gave it some feedback and said, hey, I like that you created, it's got too many words.

Can you simplify it? I like really simple one words that make it more visual. Because actually, I want people to be able to take a screenshot and the past one,

they couldn't take a screenshot off. So, here we go. This is round two.

The Takeaway Framework: Map, Test, Repeat

And this is what I want you to walk away with today. If you don't want your AI program to stall, Think about the human first.

Map out the work your team is doing. You don't need to go anywhere near the tech first. Map out the work your team is doing.

Test something with your team. Work collaboratively. Notice what helps. Notice what gets in the way.

And then repeat.

It's about people. It's about putting people first. It's about having a go. Testing and trialling.

But continually coming back to it and having a go.

Conclusion: Reinvest Saved Time into Higher-Value Work

If AI removes part of the work, what do you want people to do more of?

and actually we should all be thinking about it's not about reducing or improving productivity it's about how do we make my business more profitable how can we all be more profitable how can we all be leveling up stepping up and

growing in our roles thank you very much