Wrap Up XP2024: ideas, conversations and actions to drive Agile evolution

How is your organisation structured to support the development of large and complex products and services? And how can you improve the process to make it more effective? These are the questions Pierluigi Pugliese's talk focuses on, bringing his experience of coaching large organisations. A typical model of these organisations is a corporate structure based on an accumulated history of creating local silos. The result is that people work on 'products' that do not deliver value directly to customers. Sure, you can introduce agility into any team, but the overall effectiveness of the company will not change much. But where to start the change? From the 'definition of product', the role of the product owner and the principles that can guide the transformation of large software development teams.

"Individuals and interactions only through processes and tools" - with this new perspective on the first value of the Agile Manifesto, Alberto Brandolini discussed the messy scenario in which software companies find themselves nowadays. Organisations have undergone enormous changes in recent years, while Agile has gone from a narrow niche to a pop mainstream. However, the new emerging reality has not yet stabilised, with tensions between full remote work and a return to the office. In this scenario, old habits and tools can no longer be taken for granted and the value of context-free contributions is close to zero. Companies must consider that working tools have become more important than in the past and that they must be a choice of the team itself and not a decision from above. In remote working, the dynamics clearly change compared to face-to-face working: individuality becomes a predominant concept compared to the team. It is time to make small changes with intentionality to make a big difference.  

"I feel a great disturbance in the Force". This is not Obi-Wan speaking, but Diana Larsen, one of the leading thought leaders in Agile, author of several reference books in the field (the new version of the most famous book, "Agile Retrospectives", has recently come out, and the latest publication "Lead without Blame: Building Resilient Learning Teams" is worth mentioning). Indeed, several trends are spreading in these early months of 2024 that say that agile is out of fashion, that it is no longer needed, "Agile is dead". But is it really no longer necessary to have quality software, motivated people, and the ability to adapt quickly and effectively to change? Probably not. Agile is not dead, but there is a need to rethink it. Indeed, now it is no longer a movement of a few rebellious engineers, it is not a solution for a niche of users: we are starting to enter the 'early majority' phase of agile and we have to face the fact that there are different points of view in a community that is getting bigger every day. We have to accept that there are different declinations of agile depending on different business needs and application contexts. And have the courage to be agile not only in the short-term operation of project teams, but also in the long-term management of our companies' strategies.

Marco Fracassi explored the topic of emergent design, questioning how it integrates with the concept of architecture and frameworks that are increasingly present in code. With a wealth of experience in different projects and in dealing with numerous legacy code bases, Marco identified common and recurring patterns. During his talk, he shared his experience on how architecture and emerging design can combine to progressively improve old code bases.

Hendrik Esser, Agile practioner, manager at Ericsson, began with a provocation: in a perfectly predictable context, waterfall is the best project management methodology. In fact, it is a mistake to think that a particular methodology, as SCRUM can be, is the solution to all evils: depending on the characteristics of the context, in terms of organisation and life cycle phase, different methodologies should be chosen. This interesting model takes up a model proposed by Geert Claes of end-to-end agility, which identifies different possible methodological areas depending on the context and the market maturity of the product, taken up and reworked by Stefano Mainetti, Executive Chairman of adesso.it.

Jakub Perlak, a researcher at the AGH University of Krakow in Poland, drew a matrix that maps influence relationships within the team on two axes: centrality and density. Dense, centralised relationships towards one person characterise command and control hierarchy scenarios. Decentralised but weakly dense relationships characterise weakly cohesive teams. Dense and decentralised relationships characterise shared leadership scenarios. Jakub in particular has scientifically measured the level of relationships using social network analysis. Potentially there are plugins in Microsoft Office 365 that bring out the data collected on teams and outlooks and provide interesting visual outputs in this regard. The model presented has many similarities with the model shared by Stefano Mainetti, taken from a drawing by Henrik Kniberg: Autonomy and alignment are not at opposite ends of a spectrum. If we put them on two axes, it is easy to see that the more solid the basis of alignment between people, the more autonomy can be granted.

Thomas Loeber addressed the question of which aspects require advance planning in Agile, focusing on the concept of 'critical junctures'. These decision points influence future trajectories, some of which are more desirable than others. During the presentation, the main causal mechanisms that can lock teams into certain trajectories were discussed. Through practical examples in the field of data and machine learning systems, the speaker illustrated how to avoid these traps. Furthermore, he analysed how a team's level of investment in design practices can lead it down different paths, defined as 'the high road' and 'the low road'. Finally, he showed how a team's approach to problem solving is increasingly self-reinforcing and what we can do to prevent teams from getting stuck on the 'low road'.

The use of agile software delivery methods involving multiple stakeholders and empowered teams can introduce additional complexity and variability into the decision-making process. Although effective group decision-making is important for successful software delivery, group decision-making in the agile context has received relatively little attention over the years. Pritam Chita explained group decision-making, as well as issues such as groupthink and congnitive biases that can impact decision-making in an Agile context, introducing the concepts behind Applied Critical Thinking.