The Diagnosis and Treatment of Bimodal IT
May 25, 2016
Is it just me, or does Bimodal IT sound like a mental health condition? Unfortunate name aside, it has been adopted by companies reluctant to embrace agile but looking for a halfway-house / best-of-both-worlds solution.
In my last post I reviewed some of the issues I see with Bimodal IT; these include promoting a segregation of techniques when companies really need integration and recommending sequential lifecycles for complex projects. While it is easy to poke holes in ideas, it is more useful and rewarding to fix and improve things. So, let’s help organizations using Bimodal IT improve.
First, we should acknowledge the elegance of its simple design and understand why organizations are drawn to it. The simplicity of an If-This-Then “A”, If-Not-Then “B” approach is appealing and it allows companies to try agile-like approaches without making a full commitment. There is a refreshing clarity and simplicity in a simple two-way model. However, true to its namesake personality disorder, organizations using Bimodal IT exhibit large swings in the execution approach that are not natural or optimal.
Diagnosis: Tyranny of the OR vs. the Genius of the AND
In the book “Good to Great” Jim Collins popularized the idea of the “Tyranny of the OR vs. the Genius of the AND” to explain the problems of being forced to choose from alternatives and the potential in choosing a better third alternative – even if it takes more effort.
The “Tyranny of the OR” part describes having to choose from two seemingly contradictory strategies – either Mode 1 which is traditional and sequential OR Mode 2 which is exploratory and nonlinear. The “Genius of the AND” part refers to instead embracing both ends of the continuum and simultaneously making the best decision for the unique circumstance at hand. Most organizations are ruled by the tyranny of the “OR”, whereas Great organizations find a third way to satisfy both and leverage the Genius of the “AND”.
Jim Collins linked the ability to leverage “AND” thinking to high performing companies, but this third alternative or “Middle Way” has been around for a long time. I wrote about it in 2008 here, and my favorite quote relating to it is: “The test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function.” - F. Scott Fitzgerald.
Treatment: Mix Models, Engage the teams and Innovate
An example of applying the AND mentality to Bimodal IT would be to execute a couple of Mode 1 and Mode 2 projects and then get the teams together for an improvement workshop. We could ask them about their experiences and suggestions for cross-pollination of the best techniques. Maybe Mode 1 projects could benefit from a monthly Show-and-tell review of project outputs and planned work for the next period with the wider project community? Maybe Mode 2 projects would benefit from the development of RACI charts before distributing work between team members and part-time supporting roles?
I am not suggesting these are universal enhancements Mode 1 and Mode 2 projects, they are just examples of things that might be suggested. The real power of the process comes from getting people thinking about how to improve the process and caring about the outputs. Giving people input into how we undertake work and the ability to improve it moves them from un-invested-followers of a process to engaged-workers with some autonomy and say in how things get done. Guess which group enjoys their work more? Guess which group tries harder to overcome problems and deliver results?
I am in favor of using established and proven development approaches, they leverage good practice and help prevent common pitfalls. However, since organizations vary in function, organization and culture it is naïve to assume different companies should use the same one (or two) processes to execute their projects. The impacts of failure in air traffic control are quite different from pop-culture web sites and they should use different development approaches.
As Alistair Cockburn and Jim Highsmith have been saying for decades, we really need a methodology per project. Or as the Declaration Of Interdependance (DOI) say an approach that is "Situationally Specific". If this all sounds too hard or complicated it need not be. There are lots of free frameworks available to engage their team members in continuous improvement of their methodology. Doing so also increases ownership, support and engagement.
The continuous improvement model used could be one already in place at the organization or one that best fits with the mindset or culture. It could be PDCA, Six Sigma or Kaizen, they all share six common principles.
Gartner did a great job creating a framework that is appealing and accessible to organizations slow to adopt adaptive lifecycles. If they were to now follow it up with some guidelines for tailoring and evolution within organizations adopting it they would have a winning strategy for engaging participants and driving better results.