“What are we doing about AI?” When and How to Adapt your Decision Stack.

Every single organisation I meet is asking itself the same question : “What are we doing about AI?” (And if you’re not, you really should be). We may well be in the midst of a hype cycle around AI but unlike some other recent hype cycles (cough, blockchain) I believe AI will fundamentally change the world, and requires that we adapt our businesses and products. And I’m not alone - Menlo Ventures just released their 2024 State of Generative AI report which shows that enterprise spend in the category has surged to $13.8 billion -- up more than 6x from $2.3 billion last year…

“The reality is it all changes. Every single assumption you might have made about your business is now up for grabs.” - Des Traynor, Intercom

Learning from the Past

This isn’t the first (or the last) massive technology shift, even in relatively recent years, so it’s worth learning lessons from the past. Most recently, the massive shift from desktop to mobile is incredibly instructive.

Apple’s iPhone (arguably the first really useful smartphone) was released in 2007, and in 2010 Google’s Eric Schmidt exhorted developers to focus on mobile over desktop - but even as late as 2015 some people were describing mobile as a fad. Fast forward to today and I think it’s fair to say that is just not true. Today mobile devices account for 72% of internet usage globally, while desktops account for 28% (and shrinking). The companies that saw this shift coming early are still making tremendous gains from it.

Looking back at how mobile impacted most tech companies, they largely followed the same path:

1. Experiment with mobile - eg mobile landing pages but desktop website
2. Optimise for mobile - eg responsive websites that worked just as well on mobile as on desktop
3. Everything is an app No it isn’t - when the app ecosystem exploded it seemed like everything had to be an app and there was a new goldrush, but now it’s more nuanced
4. Mobile first - rethinking your mobile experience from the ground up for user specific use cases on mobile devices

What each step meant was of course different for different businesses. Mobile first doesn’t mean reproducing your entire product in mobile but taking a first-principles step back to what your product needs to deliver to a customer who is primarily interacting with you on mobile. The first companies to take each step broke new ground technically but won PR, customer accolades, and ultimately market share as a reward.

“One big mistake we made in our past was to think of the PC as the hub for everything for all time to come. And today, of course, the high volume device is the six-inch phone. But to think that that's what the future is for all time to come would be to make the same mistake we made in the past.” - Satya Nadella, Microsoft

And it’s not just software - everyone seems to have missed the exponential impact of solar power on our energy production. As Nat Bullard shared on the Exponential View, “every year this decade, the International Energy Agency (IEA) has had to revise its solar deployment forecasts upwards. And not just by small amounts. Actual installations have been as much as three times higher than the agency predicted five years ago.“

So what can we learn from that?

Zoom out and move up the stack

The first inclination when a technology shift such as mobile or AI comes along (once you’ve pulled your head out of the sand) is to tweak your product - the experiment and optimise steps above - and we’re seeing this today with just about every tech company out there sprinkling some magic AI dust on their product. While that can be fun (and let’s face it was enough to drive serious valuation jumps for some startups) it’s important to see the shift early, take a step back, look at the bigger picture, and see how a new technology might affect your whole Decision Stack.

Does your Strategy still Make Sense?

If you zoom out of what you’re working on right now and look at Strategy you’ll quickly realise that AI is changing everything. Just some examples to consider:

• Business models are changing: AI doesn’t scale like software - instead of every single additional user having essentially zero marginal cost, adding expensive AI queries to your product drives a completely different cost structure (and that’s even without discussing the externalities of increased energy use and carbon output)
• Combine that with the potential impact AI is having on your customers’ business and their desire to reduce headcount, and pricing has to change from per-seat to value
• As a potential cost-saver AI can suddenly shift your buyer from a function head to the CFO
• If AI starts automating customer workflows in your product, are they even logging on to your UI anymore? What does that do to your OKRs?
• Even when they are logging on, how does AI change the interface? What tasks can be completely automated away from the user? Which require human approval? Which will users still be focused on?

And that’s just the tip of the iceberg. Any one of these is enough to question a lot of the assumptions that went into your Strategy - taken together they might invalidate the entire thing.

Systematic Impacts can Lead to Perfect Storms

These impacts don’t happen in isolation either. Right now there is a perfect storm of cost-saving happening across the software industry, where tightening budgets and an increasing focus on costs are driving AI adoption. When you zoom out to look at your Strategy, therefore, you can’t do it in isolation from the broader trends in the industry. Using tools like SWOT and PESTLE regularly can help. This systemic impact is the hardest to see while you’re in the middle of it, but the most crucial to understand in order to capitalise on new trends.

As Tim Harford shared in his book and radio series “Things That Made the Modern Economy”, containerisation of cargo started in the 18th century and the shipping containers we take for granted today were invented in the 1950s - but didn’t explode onto the market until the late 1960s when the perfect storm of a dockworker strike, growing international trade, and a war in Vietnam came together to require them.

Does your Vision still Make Sense?

If you zoom out even further it’s worth checking whether the technology shift might even impact your vision/mission - does this technology shift enable a completely different way to achieve the same vision? One startup I coached went through this early on in the AI cycle when they realised everything they had built could become obsolete because AI would bring in a completely different way to solve their mission - the overarching problem customers had. The founder quickly got the whole team to experiment with new ideas for how they could solve their vision with new AI tools. They’re still on this journey but by recognising the impact early, they are now years ahead of their competition.

“Just like every technology that came before it, the winners will be those who can use AI to find differentiated value, to solve problems customers didn't know they had, in ways that traditional approaches can't compete, and faster than the other companies nipping at the heels of the same customer.“ - Jonny Schneider, Author of Understanding Design Thinking, Lean, and Agile

Things you should be doing right now about AI

Based on lessons from the past, and the idea of regularly zooming out and looking up your stack, today you should be:

• Looking at how other leading businesses are reinventing themselves in face of an uncertain future
• Experimenting with AI internally and running hackathons so your whole team get time to learn about these new tools and what they can do
• Understanding where you can utilise generic generative AI and where you might have a data edge that makes building your own tools unlock real value
• Learning about Generative AI tools but also what’s coming next - Agentic AI
• Re-examining all your assumptions about your customer, their problems, and how you solve them
• Asking yourself whether any part of your value proposition can now be solved by the user directly using AI and what that means for your Strategy
• Regularly questioning whether your Objectives, Strategy, and Vision still make sense : or whether you can leapfrog your competition by choosing a completely different strategy now and take the leap from sprinkling AI on your product to going AI-first

What this teaches us about Strategy and your Decision Stack

We are shockingly bad at predicting the future, especially when it comes to massive shifts like mobile, solar, and AI. As Bill Gates said “most people overestimate what can be accomplished in one year and underestimate what can be accomplished in ten years.” But just like strategy, studying and predicting the future isn’t an innate skill but a muscle you can build through practice and iteration. The key is to regularly take time to zoom out of your day to day and look at the rest of your Decision Stack to make sure that you’re focused on the right things and adapting to changing market conditions.

Strategy, therefore, is never “done” - as Clayton Christensen said “Most people think of strategy as an event, but when we run into unanticipated opportunities and threats, we have to respond“ or put another way, as Mike Tyson famously said “everyone has a plan until they get punched in the face”.

It’s critical to be able to adapt quickly, so always making sure that you leave room for big bets and new ideas means you’re more likely to seize on new opportunities as they come up. AI represents an enormous transformative opportunity, and organisations that zoom out, adapt their strategies and even visions holistically while embracing rapid experimentation will be best positioned to thrive.

How are you adapting?