What does AI do for an MD?
One of my team asked “so, what does AI do for you, as an MD?”
Business leaders have been falling over themselves to show how much they love AI and its potential. Bill Gates said it was the future, in a much-quoted piece last month. Sir Martin Sorrell wrote in an article – Why the Mad Men would be crazy to ignore AI (thetimes.co.uk) – that it’s the future of the advertising industry. And seemingly everyone thinks ChatGPT is amazing.
I would caution some of this praise. ChatGPT is rapidly going through the Hype Cycle, as people discover that it’s very good at providing generic advice or copy. But that’s not actually what people want – they want advice that is informed by reality, and with genuine human sentiment. By way of example, Martin Sorrell in his article says “One opportunity is to create AI-driven brand experiences — we are calling it “Brand AI”. For example, during the pandemic we used deepfake technology to create a virtual AI version of the hip-hop star The Weeknd on Spotify for his fans to interact with.”
Given all the ethical challenges with deepfake technologies – using someone’s image to say words or espouse opinions that they didn’t say or believe – this seems a strange claim for a benefit of a technology for an industry – brand identity – where authenticity is crucial. Secondly, on this point, the needs of consumers are very different now to what they were during the pandemic, which started three years ago; at that point, people weren’t allowed out of the house, so creating a “synthetic” immersive experience was the closest anyone could reasonably get to a real, authentic experience with The Weeknd. Now, consumers would expect a level of authenticity with such an experience, which deepfake technology will by definition struggle to achieve. This is not to question the potential for some of these technologies, but I think ethics, and authenticity, especially when it comes to areas like creative content, needs to be considered in a lot more.
However, I think we are guilty at the moment of conflating the sudden evolution of Generative AI and the growth in computational statistics in general with ChatGPT. The world of Generative AI will be very different, even by the summer, let alone this year. We are possibly witnessing the fastest maturing market ever, so whilst everyone is talking about ChatGPT at the moment, give it a few months, and different tools will emerge and the landscape will be different, and more mature. I also think we, as users, will become more sophisticated with the type of questions that we will be able to ask. Give it 6 months, for instance, and presentations will pretty much start writing themselves.
However, the question is still valid; how does AI benefit me and my role? If I didn’t think it was valuable, what am I doing speaking to clients about data and its opportunities! And in any event, I’m very positive about the opportunities over the next few years.
For me, AI is most valuable when it is driving greater efficiencies; enabling teams to do things more quickly, and so increasing productivity. This can happen at a general, or macro, level, or at a more practical, day-to-day one. So, I’ll give a couple of examples.
Firstly, we are always working on new propositions, either based on how we can solve clients’ problems with our solutions, or where we are developing or packaging our existing knowledge for clients. So, how do we build these out, and how do we find what the market is?
We have been recently working on an R&D project, which is pioneering in its use of computer vision. However, the challenge with pioneering solutions is that, almost by definition, they don’t have a clear market yet. We have established that there is a valuable methodology in what we have been working on, and its effectiveness in predicting its area has significantly improved in the relatively short period we have been working on it.
However, who will buy such a service? And what issue or pain point is the solution solving for them? And is that pain or need great enough for them to justify the commercials?
Typically, identifying such a market and the research needed for it would take months, and I’d need to build out a business case to take to them in the first place. These are significant barriers to entry for entrepreneurs or new service start-ups. However, with tools like VenturusAI, I can create a business plan, and outline business case. This is credible, reasonably comprehensive and thorough, and includes known frameworks, like SWOT Analyses and Porter’s 5 Forces. It also gives me some additional ideas about potential customers of the service which I hadn’t considered, so it is doing more than just writing up what I could reasonably achieve on my own.
A further search in ChatGPT (or Bard) gives me a list of the top 50 potential clients in that category. From that point, this allows me to identify the names of the Data Directors of those 50 clients.
So, in the space of a few searches, I not only have a good outline proposition which I can look to test, but I can focus on a list of those who are most likely to care about the issue, which I can then refine and validate myself. Weeks of work, condensed into an afternoon, and most of that is waiting for the AI tools to compose the answers. This is how I see AI being most valuable; in areas where there are known frameworks or clear sets of questions, like market research, that can be answered in a consistent way.
To give another example, as a more boutique consultancy, we may struggle to compete with the levels of research that a global management consultancy can provide, with access to expensive research reports or subscriptions. The same tools mentioned above now enable us to go into meetings better prepared, with a clearer idea about what the challenges that may be facing a client may be, and how we can help them. This increases our credibility, but reduces the effort around preparation, which may previously have been another barrier to entry. This has already had a significant impact on what I do on a daily basis.
At a more practical, day-to-day level on how we are helping clients, a clear example of the value is AI where it does help is with forecasting. As businesses shift towards understanding longer-term value metrics, as opposed to focusing on short-term revenue goals, being able to forecast becomes more and more important for business leaders. Because the end goal of a CLV strategy might be stretching well into the future, how do you make sure that your activities are on the right track? You can’t wait 3 years to find out that your marketing programme in the next six months didn’t work. So, you need to find a way to identify what are the right behaviour, and outcomes, along the way. That’s where forecasting can come in. By using your available data about successful customers, and how they behave over a longer-time frame, you can use predictive techniques to identify what those shorter-term metrics should look like, and what the parameters should be for successful longer-term outcomes.
Having a longer-term view of the potential of the business is in many ways what senior leaders and the board are there for. So, this is a huge benefit for what AI can do for your business right now, using your existing data. And that’s why we are providing a forecasting service for our clients, for any project, whether big or small. Because being able to forecast and understand what your outcomes should be will enable you and your teams to achieve those outcomes.