tl;dr Allen Holub’s YOW! Brisbane talk explored why organisations resist meaningful change (Larman’s Law) and offered strategies, like experimentation and delivering value incrementally, to overcome these barriers.

Introduction

I recently attended the YOW! Brisbane 2024 developer conference and thought I would write up one of the more enlightening talks I saw there. Allen Holub’s “Getting Buy-In: Overcoming Larman’s Law”.

Allen Holub is known for his dislike of what has become of modern “Agile” software development and is an advocate of a more pragmatic approach that is more in line with the orginal manifesto for agile software development. He’s also a proponent of the #NoEstimates movement. I find myself, more often than not, agreeing with a lot of what he has to say. However, I’ve found that his suggestions are impractical for the consulting environment I work in. For example, no corporate software project is ever going to get off the ground without some form of estimation of time and budget.

So of all the talks I attended at YOW Brisbane, this was the one I was most looking forward to. He began by explaining some of the development practices that he advocates for and then discussed why organisations often resist them and how to overcome that resistance. So let’s get into it.

Practices Holub Recommends

Holub suggests several practices to improve software delivery. While effective, many of these challenge traditional corporate norms and often face resistance from management:

  • Mob programming: Mob programming, Holub argues, is highly effective but often misunderstood as “multiple people watching one person work”. True mob programming involves the entire team collaborating on the same task in real-time.
  • Reducing Work In Progress (WIP): Limiting the number of tasks being worked on simultaneously improves productivity and reduces waste, but this simple concept can be hard for companies to embrace.
  • Delivering in small increments: Holub advocates delivering value in small chunks with short feedback loops instead of working on large projects over long periods. However, this shift in mindset can be seen as risky by management. Despite the opposite being true.
  • Regular user feedback: He highlights the importance of delivering value to customers as soon as possible and the importance of getting regular user feedback, which may not align with traditional development cycles.
  • Adapting processes to teams: Holub insists that teams should adapt processes to suit their needs, as they are the ones best positioned to know what works for them.

Larman’s Law

When trying to adopt new ways of working at your organisation, Holub explains that companies tend to avoid fundamental changes to their processes and instead opt for superficial changes, like relabelling existing roles without changing the way people work. For example, project managers may be given the title of “scrum master” without being trained or empowered to act differently. The underlying processes and management structures often remain the same. Companies often implement ‘fake’ Agile processes, where they merely put new labels on existing practices without real change.

“Organisations are implicitly optimised to avoid changing the status quo” - Larman’s Law

Larman’s Law states that organisations are optimised to avoid changing the status quo. This resistance is often driven by middle management, who have incentives to maintain their roles and current power structures. Holub emphasised that this resistance is not always malicious but often stems from a discomfort with learning new skills.

Overcoming Larman’s Law

Holub offered several strategies for breaking through and challenging this resistance:

  • Bottom-up experimentation: Teams should experiment with new ways of working on their own and do so in safe spaces to test ideas without fear of penalty. A “skunk works” approach.
  • Top-down business cases: Presenting changes in terms of financial impact rather than abstract concepts like ‘agility’ is key. Speak the language of business to upper management and address their fears by showing how changes reduce risk. For significant change to occur, it’s essential to have the support of upper management.
  • Demonstrate, don’t just talk: Actions speak louder than words. Holub shared an example of a team visibly working in a new way, which sparked curiosity and interest from others and demonstrating the benefits of new practices to win over sceptics.
  • Don’t ask for permission: Start small experiments without waiting for approval. Holub’s advice is to “ask for forgiveness rather than permission”.

“We need a way to talk to management in a way that makes sense to them, in terms of cost and benefit.” - Allen Holub

Alternatives to Estimation

As I’d hoped, Holub also discussed alternatives to traditional estimation, which he believes is often a waste of time. According to him, estimation stems from upper management’s fear of risk. Managers use estimates to feel in control when processes are opaque, but these estimates are usually inaccurate and can lead to poor decisions, such as cutting corners to meet artificial deadlines.

Instead, Holub recommends focusing on delivering value to customers as soon as possible through small, incremental releases. He argues that when stakeholders can see tangible progress regularly, the reliance on upfront estimates diminishes. While this approach is logical to me, it still doesn’t address the one question I continually have to answer at the start of any project which is “how long will it take and how much will it cost?”. I’m still not sure how to get around this in a corporate environment.

Conclusion

As a long-time follower and admirer of his work and opinions, it was great to see Allen speak in real life. I don’t think I was aware of Larman’s Law until this talk and I came away with some new practical ideas on how to push back against this resistance to change. While I find his argument against estimation compelling, I still believe that it is all but impossible for me to avoid doing this any time soon. Nonetheless, Holub’s message was clear: small, incremental changes, combined with a focus on delivering real value, can gradually transform even the most resistant organisations. His blend of humour and insight made this one of the most enjoyable and enlightening talks of the conference.

Thanks for reading!