How to create business value with AI
AI is a top priority for CEO’s - they understand that in just a few quarters the technology has the potential to accelerate their company ahead of competitors, or see them fall behind more nimble companies that race past them.
How can CEOs create business value from this new technology, and ensure that they win this race, securing their company’s future, and their own jobs?
AI in itself is just a concept, a platform - it is what you do with it that will transform your business.
Here are four images to help you think through how AI can create value in your company.
AI and the steam engine
On an insightful 20VC podcast episode this week Ethan Mollick, co-director of the Generative AI Lab at Wharton used the invention of the steam engine as an analogy for AI. (link to the full episode at the bottom of this post).
In itself, the steam engine was interesting, but not very useful. It was able to generate lateral power - in and out, in and out - but not a lot could be done with that.
It took intelligent craftsmen that were close to the actual manufacturing processes in the factories to create ways of harnessing that lateral power and turning it into circular power for a lathe or a bottle shaping tool.
The steam engine was only useful if you could connect it to factory tools
So with AI, consider the same approach - just investing in the latest LLM doesn’t provide business value.
It needs to be harnessed and connected to your core processes.
As an aside, 12 years ago I was working on a project with a large UK utilities company. They had purchased 500 seats of Salesforce’s Force.com platform to ‘play around’.
They had no real business use case, and I used the analogy of a pile of lego bricks with them.
They had bought a nice toy but didn’t really have anything useful to build.
AI Lego Bricks
Connect to core business processes to avoid playing with AI.
They spent a year looking for different ways of using the platform, but because it was never linked to a strategic business process the project never went further than building an employee pulse survey that could have been handled with an HR tool they already had.
Be careful you don’t end up playing with your own set of AI lego bricks.
Align AI to your P&L statement
Senior executives are investing in AI for three main reasons:
Grow or protect revenues: by launching new or improved value propositions
Improve margins: by reducing costs across the business
Reduce or mitigate risk: particularly financial and cyber security risks
Most of the AI use cases in development today are focused on the second bullet point - improving margins by removing cost from the business:
Customer service chatbots, AI sales development reps, automated HR and finance processes - picking up work that previously would have been done by an expensive human.
As you explore the highest value opportunities for AI, take a look at your P&L statement and consider where you can have the most impact.
Current AI projects are focused on the red columns in the image - reducing the Cost of Goods Sold, reducing the cost of R&D, reducing the Sales, General and Administration costs.
But you can only cut these so far - there is a defined floor - zero.
In contrast, the green column on the chart is linked to growing or protecting revenues through new or improved value propositions.
Here there is no ceiling. The potential for you to launch new products that disrupt your company and the category you sit in has unlimited upside.
New products for new customers
As you then think through how you can create new products and services, another dimension to look at is who you create them for and how do you monetise them.
In this framework we have four quadrants:
Bottom left: we add AI features to our existing products and use that to increase or maintain the value of what we sell against our competitors. (i.e. image background remover in Canva)
Top left: we create new AI modules or products that we sell as an add on to those existing customers that want it (i.e. full self driving from Tesla)
Bottom right: we use AI to enable us to sell our existing product in new markets (i.e. we can now sell in new countries because we can support different languages or timezones without hiring new people)
Top right: we use AI to sell new value propositions to entirely new sets of customers (i.e. we use our data to sell aggregated predictions and analysis through a data marketplace like Snowflake)
The further up and to the right you go, the more impact you can have on the green column on your P&L image.
Build your AI management system
We looked at the steam engine analogy at the start of this article - and as a leadership team you need to build the capabilities and incentives for your business units to figure out how to harness the power of the AI tools you introduce.
As Ethan covers in his interview - in many companies there is a fear from the employee level of highlighting their use of AI. Will I lose my job? Will I be seen to be cheating? Will I be given more work to compensate for the efficiency gain?
Instead you’ll want to build out a framework that educates, motivates and rewards those that introduce ways of harnessing AI that impact the green and red columns in the P&L.
I liken this to an octopus - where your AI control centre, closely linked to your senior execs and the company strategy, defines the guardrails for use of AI and introduces the core capabilities.
But down in the tentacles - the individual business units, regions, functions and teams - are where the real value gets created as the power is plugged into the real processes that run your company.
Four images to create AI business value
So there are four steps to walk through:
Connect AI to your core business processes
Align AI use cases to your P&L
Consider new products for new customers
Develop an AI management system that rewards value creation
I’d love to hear how you are approaching AI value creation in your organisation.
Watch the full Ethan Mollik interview: