Tuesday, November 29, 2011

Describing Agile as a Set of Process Patterns

A couple of us are Thinkimg about how to visualize our approach to project delivery within the Deloitte LEAN crew (yes people at big consulting forms document their methodologies amd tools, even the guys with a more agile bent).

One universal truth for us is that we have never use the same set of techniques twice. Sure we reuse select practices, but they we always being tweaked, amd something else is always getting mixed in. Experience ranges from RUP, to all mainstream flavours of agile, to Kanban amd Lean, with us just beginning to play with Lean Startup methods.

With that in mind I came up with the following illustration, loosely based on the original GOF patterns diagram at the beginning of the book. All the diagram may seem overwhelming, the beauty of it is that you can start anywhere you want, and likewise stop, depending on your context.

Saturday, November 26, 2011

Applying Lean Startup techniques to Kanban (or any other) Organizational Transformations

Eric Ries definition of a startup is as follows:

A startup is a human institution designed to deliver a new product or service under conditions of extreme uncertainty.

By this definition the enterprise wide Kanban transformation that I am helping a client with can be deemed a startup initiative.One that could be optimized by lean startup standards.

Eric talks about how Lean startup techniques like innovative accounting, validated learning, analyzing engines of growth, etc can be applied to organizational change initiatives. These kinds of change initiatives are by definition wrought with extreme uncertainty. So I've decided to apply a couple of concepts from the LeanStartup toolkit to see of the model applies.

The notion of wether to pursue or pivot on a strategy is one of my favorites coming from Eric's great book The Lean Startup. Startups are constantly evaluating their progress using useful metrics such as cohort analysis to determine wether to continue with a strategy (pursue), or keep one aspect of the strategy the same, and fundamentally shift another aspect (pivot). In this case our objective was to grow Kanban usage, and maximize growth of active users for Kanban enterprise wide.

I ve mapped out our Kanban transformation, including all it's twists and turns below, and it looks like we have been following aspects of the pursuit or pivot model without realizing that is what it was.

  • We also made many of the mistakes that many startups make according to Eric.
  • We spent way to much time designing a solution, we actually knew this would be a problem, but are hands were a little tied by the client, That being said I was unprepared for how unhelpful the upfront design activities actually were
  • We took a long time to make our first pivot, had we been armed with the model, we probably would have pivoted as soon as we saw the transformation was not going at the pace it needed to
  • We could have been tracking the relationship between specific activities and Kanban adoption,and set up all activites using validate learning, this would have accelerated validated learning, and identified the need to pivot faster
  • That being said our ability to pivot did increase dramatically over time. Our first pivot took 5 months, and then 3 months, then less than 2.
I think the Lean Startup model has huge application inside the enterprise, managing the many aspects of IT that contains a high degree of uncertainty.

Our next steps are to start tracking our initiatives and mapping them to non vanity metrics. We think tracking cohort users, and their usage of specific Kanban features is the way to go. We can do this thanks to the fact that all Kanban adopters are sing an electronic tools, LeankitKanban, as the tool of choice