249. AI ROI: 7 Insights from the Economist Business Innovation Summit
Apr 09, 2025
Some companies waste millions on AI initiatives while others use it to gain serious market advantage.
In this episode, you'll learn the seven insights Sophia Matveeva, our CEO, got from The Economist's Business Innovation Summit and what they mean for your AI strategy.
This episode is for you if you:
- Are a leader considering AI investments but want to avoid chasing vanity metrics
- Need to justify AI projects to your CFO or board with realistic ROI expectations (!!!)
- Want to address "shadow IT" - when employees use unauthorised AI tools with company data
- Are responsible for upskilling your team in the age of AI
Timestamps
00:00 Introduction to AI and Market Dynamics
03:11 Key Insights from the Economist Business Innovation Summit
05:58 The Importance of Speed in AI Implementation
08:51 Understanding Shadow IT and Its Implications
12:12 The Role of Boards in AI Adoption
14:58 Training and Upskilling for the AI Age
16:24 The Impact of DeepSeek on Global Innovation
For more career & tech lessons, subscribe to Tech for Non-Techies on:
FREE COURSE:
5 Tech Concepts Every Business Leader Needs To Know
Growth Through Innovation
If your organisation wants to drive revenue through innovation, book a call with us here.
Our workshops and innovation strategies have helped Constellation Brands, the Royal Bank of Canada and Oxford University.
Transcript
Sophia Matveeva (00:00.234)
Our product, our ideas can reach the market faster. That means we can beat the competition. That essentially means we can grab more market share. Speed to market is something that AI can help us with.
Sophia Matveeva (00:16.366)
Welcome to the Tech for our Techies podcast. I'm your host, tech entrepreneur, executive coach and Chicago booth MBA, Sophia Matheva. My aim here is to help you have a great career in the digital age. In a time when even your coffee shop has an app, you simply have to speak tech. On this podcast, I share core technology concepts, help you relate them to business outcomes, and most importantly, share practical advice
on what you can do to become a digital leader today. If you want to a great career in the digital age, this podcast is for you. Hello smart people. How are you today? So a couple of weeks ago, I went to the Economist Business Innovation Summit in London, and the subtitle to that was AI ROI. So obviously I had to go find out what are people saying and so
Basically, I'm going to give you my summary of the trends, but also basically of what wasn't said. And there were, you know, fancy people, powerful people from some of the world's biggest companies. So there were people from LinkedIn, Deloitte, Rickett, Experian, Lenovo, Citi, WPP. So basically the big ones. So here is one of the things that people talked about. They said, stop chasing vanity metrics for AI, which is always useful advice. So basically.
That's when innovation projects are AI powered or when startups are AI powered. Like that's basically really silly. So focus on solving business problems faster and cheaper. So basically if you can't connect something to a problem that you're solving, it's not a thing. Okay. So obvious, it's blindingly obvious, but you know, we are living in this age of AI hype. So actually bringing things down to the blindingly obvious is really useful.
Okay, here's another piece of useful advice from the summit. So basically the AI transformations that work in large corporations, they start small and they get bigger. So what doesn't work is if somebody signs a check for like, I don't know, $3 million per year to open AI. Obviously open AI are very happy, but that basically is that too much. It's too big. Don't start with that.
Sophia Matveeva (02:41.615)
start with investing maybe 100k and basically see it as a series of experiments. know, invest 100k in a targeted pilot or a specific team and they're going to use AI to solve a specific set of problems and see how that goes. So basically roll out team by team, pilot by pilot, don't do one big massive spend and then basically just hope for the best.
That point should make the CFOs watching this happy. point number three, speed is the new core competency. So the thing that we've noticed in our programs on no code tools and on AI for product innovation is that basically not only can you make prototypes and you can make new products cheaper because you don't have to involve expensive experts from the start, it's also much faster.
So we found in our programs that, okay, we can teach people with ideas to use AI tools, to create a prototype, to test it, to get it to market, and then they can figure out is this something worth investing in or not. And we actually heard that LinkedIn is using the same methodologies, which kind of made me feel really good. Anyway, so what we heard from the AI summit, the economist AI summit rather, was that LinkedIn's
Product is also changing, product management is changing with these AI tools and they're measuring speed as the new core competency. So basically that means that if there's somebody who has an idea for a new product or a new major feature within a product, they can then use AI tools to create a prototype. They can get that prototype tested with the target users and actually what they're doing at LinkedIn apparently is then they can actually use AI tools
to code that whole thing. obviously cost savings are one measure, but another really important measure is speed. Speed beats size, right? So AI gives you speed. The faster you can get a product to market or a new feature to market, the faster you can get feedback. And also the faster you can start making money. So really, if you don't remember anything.
Sophia Matveeva (05:02.521)
from this episode, remember that AI gives you speed, speed is the new competency because people, time is money. I'm not the first person to have said that. Okay, here's another thing that I learned, which I thought was really interesting. Beware of shadow IT. So Deloitte actually did some reporting on this. So basically what happens is that there are some people in organizations that are using chat.
GPT and AI tools like ChatGPT just for their own use. you know, I have ChatGPT on my phone. I use it for work, but I also just use it for like hotel recommendations when I'm traveling. I use it for recipes. I use it for basically even advice on what to do with my cat. So there are people like that in lots and lots of organizations. And then there are also people who've never touched the stuff. So.
What happens is that at work in an organization, if there is no program for AI implementation, if there are no AI tools, also if there's only just a directive saying, don't use that here, don't use our internal documents with this stuff, basically people will ignore that. And the people who are using AI in their personal lives anyway, are then going to use chat GPT and tools like that for work.
And this is called shadow IT. So it's not official AI implementation. It's what some people are doing by themselves. So in some ways it's a good thing because those employees are more innovative. They're more productive, but also as an organization, you're losing control and you know, your data might be being fed into AI tools and then being used in all sorts of ways that you don't have any control over. So shadow IT is growing really, really fast.
So the message is give your teams approved secure AI tools and workflows. Yes, not everybody will use it, but some people will. So the people that are going to use it anyway, make sure that they're using the stuff that you want them to use, not what they're downloading of the internet. assume don't, another thing is don't assume that because people aren't necessarily talking about using AI, that they're not because
Sophia Matveeva (07:26.925)
Some people are using UI and they actually don't want you to know about it. So shadow IT is a really, really big thing. Okay. Another point that people mentioned at the economist summit, which I thought was really fun, was that they talked about how boards love AI, but don't understand it. So basically there was somebody who spoke quite honestly about the fact that most boards are not digitally savvy.
And actually there is a drive to get digitally savvy board members onto big company boards. And because boards are not digitally savvy, they are, and also they're kind of being bombarded by hype and by PR departments from Silicon Valley tech companies. And so they're kind of falling for this AI hype and they're like, my God, we need to implement AI and then we're going to be really, really rich and successful.
pushing for AI implementation. They don't necessarily really know what they're pushing. And what people at the summit said is that board members often have unrealistic expectations for what AI can do. And they have unrealistic expectations for ROI. And now given my experience, you know, working with governments and working with enterprises, when somebody knows only a tiny bit, but not like...
can't really go beyond the surface, they can very confidently ask you stupid questions and they can very confidently basically pressure you into doing things that just make absolutely no sense, like promising some kind of ROI metrics, which are kind of never going to be delivered. So number one, just understand that boards are going to fall for this AI hype. They are going to pressure. They don't necessarily understand. So what do we do? Well,
We educate the board, know, send them to TechFodOn Techies. We can help you with that. bring AI literacy into the boardroom because basically if you don't, you might end up having to have initiatives that are kind of harebrained. This is not about teaching coding. This is about teaching strategic oversight to the people who are overseeing the company. Help them to understand what use cases exist, what the real risks are.
Sophia Matveeva (09:48.461)
what questions they should be asking and what's realistic to expect in what timeframe. My advice is don't leave the CTO to just handle this because often, you know, it might just, the board might not appreciate that from an internal person because they might think, the CTO is just being difficult and just pushing back. This is why I actually find having an external expert with great brand.
background and expertise telling them this, that tends to go down better. Point number six, and we've got two more, so six and seven. So training now is a strategic priority and kind of not just this HR perk. And I noticed this because literally all the leaders that were at the summit
And they were talking about how their organizations are adapting to the AI age. They were all talking about what they're doing for training. They were talking about how they are, you know, increasing their training budgets and what they actually want to get out of that training. And this was not an LND thing. Like these people who were at the summit, they were, you know, strategic leaders, leaders of business units, people with P &L responsibility. So.
training for organizations, training for teams, and especially training for budget holders and people in power in the age of AI. This is something that the smartest companies are doing. so my, and obviously, obviously I'm not objective here. I mean, I run an education company called Tech for Nantechies, but honestly, I was really quite happy when I was at the summit because I was like, okay, well, what I have been saying, other people are now saying, and I have not paid them to do so.
Okay. So my advice, offer targeted upskilling across functions and make sure that the upskilling is aligned with the business goal. So this is why in general, I would advise that whatever, whatever education company you work with, obviously I'd love to work with you, but you know, whoever you choose, I don't think you should just get a random off the shelf pre-recorded training that they've done for lots and lots of people because you need to make sure that
Sophia Matveeva (12:12.003)
people can translate what they are doing, know, the business goals that they're working on with the technology that you're teaching them. Because often just these off the shelf solutions, when people can't make that link, they kind of don't understand why they need the training and how it really relates to what they're doing. So do not assume that a general AI webinar once a quarter is enough. That's basic. Like if you're going to do anything, don't do that. Don't either do nothing.
or do it properly. Go big or go home. Which is the opposite of what I was saying for the amount of money you should be spending on an air transformation. Go big or go home, I mean, take your training seriously. And by the way, good training does not have to cost lots and lots of money. Okay, final thing. You know what I noticed? I noticed that nobody was really talking about DeepSeek and I found that really weird.
because for example, I have friends who work in big tech in China and DeepSeek has completely changed the tech and venture capital scene in China. And you know, there's so much investment, there's, you know, like the Chinese tech scene has really been changed by this innovation by DeepSeek. And so there I am at this AI summit and everybody's talking about,
the different ways they're using AI and what's working and what's not. And there was barely a mention of deep sea. So what is going on? Is it that basically, you know, Europeans are in denial about what's happening in China? Is that Europeans are kind of in denial about the fact that, okay, Chinese tech is really now a thing? I don't know what, I don't know the reasons, but I thought it was really, really interesting.
Also, from my point of view, what I'm seeing on the innovation side of the spectrum is that actually smaller companies innovate as they're really excited about DeepSeek because it does allow you to build things for much cheaper. Anyway, so, you know, when you go to these big events, it's interesting to see what's and what's not said. Okay. So let's wrap up. What did we learn?
Sophia Matveeva (14:29.515)
AI, AI ROI is meaningless unless it is tied to business outcomes. mean, yes, obviously. Speed and experimentation. So speed to market. Speed is the new competency. AI gives you speed. So when you are trying to get an AI project approved, when you're trying to shake the CFO down for money, then you could not only talk about money, you could also talk about, okay, our product, our ideas.
can reach the market faster, that means we can beat the competition. That essentially means we can grab more market share. Speed to market is something that AI can help us with. If you are not training your people properly, then essentially some people are going to adapt AI by themselves. A lot of people won't. My advice is have targeted trainings specifically for, you know, kind of go business unit by business unit.
Do not do these general AI is a good thing webinars. Beware of shadow IT because even if you don't know it's happening, it is definitely happening. And every single big organization has sensitive data. So it is possible that your sensitive data is already out there in some sort of LLM. It's very, very likely in fact, so the way you combat it is not by telling people, don't do it, but by actually giving them tools.
that they can use. and final points in our wrap up, boards need guidance because they are getting very excited by buzzwords. So they need guidance and they need education. And finally, remember DeepSeek. DeepSeek is a huge innovation. So just the fact that, you know, maybe some leaders in Western companies are not talking about it does not mean that it is not an absolutely world changing event. Okay, so.
If you have enjoyed this lesson, then please give this video a thumbs up on YouTube or give the show a rating and review on Apple and on Spotify. Thank you so much. I've loved spending this time with you and I'll be back with you with some more fabulous education content next week. Ciao.
Sign up to our mailing list!
Be the first to hear about offers, classes and events
