Ai and Agile 2.0

End-to-end iterative product creation is finally here

I have been discussing the future of agile methods with one of my dear old colleagues. We have been toying with the question of what is the future of agile methods going forward. We wanted to form our own opinions on what is “agile 2.0”. 

I know agile 2.0 sounds like a fad and not necessary. It can be rightly questioned whether such label ever will produce anything anybody wishes to stand behind. To me, the agile 2.0 has been a good label because it kept me thinking. It was not evident what agile 2.0 should be for me. Sometimes when you think long enough, your brain just eventually forms the necessary connections. Then I started believing.

I ended up believing the following about agile 2.0. The missing ingredient was the large language models (LLMs) and advances in AI. 

Agile didn’t become agile yet

Design thinking is fully iterative. Strategy is future scenarios and continuous refinement with OKRs. Growth hacking and marketing is experimental. Exploratory testing is continuous and iterative. Making software in teams is iterative and incremental. 

Why has all this iterative niceness not come together in most companies as a really cross-functional, super-fast and iterative process? 

I mean the ingredients have been known for decades now. The aim has been there for ages. I think the reason is just still that we didn’t know how to be quick enough. Experiments took too long. We didn’t have enough time to talk to enough customers. We didn’t have the time to synthesise what we learned. 

I’m not saying we didn’t want. Time is just a mean adversary. 

AI and transaction costs of doing the right thing

Agile 2.0 to me is doing those right things that we failed to do in the past. 

I’m a believer in AI and a witness in the time when large language models and agents cut through the time it takes to create something new. This applies to both analysis of the problem domain but also generation of solutions. You can just vibe code a prototype in 15 minutes. You can summarise interview results in 5 minutes. 

The speed of going from zero to something certainly is not the same as before. What we have at hand is indeed something totally new. The way we work will change in very rapid pace over the next months - not years. 

Doing the right thing is no longer constrained by the time in the calendar. 

Small teams, bigger collective results. Smaller teams, righter products

Cross-functional end-to-end innovation will be the norm.

Ai is said to be the great leveller. AI lifts people’s competence on domains they don’t know. AI also drives up infrastructure-like services for marketing, legal and all kinds of specialities that used to require full functions in scale-up companies. 

There is absolutely no need to wait for marketing team anymore. Or any other team.

We are not there yet - except in few brave leaders in pioneer companies. The linear model of ideate, build, market and sell is collapsing into a single team - single iteration. 

AI is for the super-iterative work

In my opinion only the iterative way works with AI. 

We will have smaller teams that can do more. Eventually we have large-scale product launches with just a handful of people. Using the roadmap as coordination tool is soon very old-skool. Less need for predictable roadmap. The trends of right outcomes reached and direction matter more.

So stretch your thinking ahead - imagine the future. With AI, it will be the dreamers and definers who will win. AI output has an element of randomness. That’s part of what makes the technology work. The AI output cannot be fully controlled ai priori. That is the human magic that we will still need. 

Humans will be the one giving direction. There is also something about the competent product person who can curate the right stuff from the endless stream of possibilities that the AI already creates. 

We will need to have righter products. This rightness will be created in hyper-iteration with small teams in direct collaboration with the customers. 

To me this is finally the agile 2.0 - It is about time that arrives.