The Agile Approach is Actually a Varied Approach

In my last post I discussed the conditions that drive organizations towards agility, using an adaptation of the Stacey Matrix (from Strategic Management and Organisational Dynamics by Ralph.D. Stacey). In this post I’ve annotated this diagram in different ways, highlighting how different approaches are required to deal with each type of work:

With predictive work, the path to success is clear. A perfectly valid approach in these circumstances is to plan the work and work the plan. This is an ordered approach where detailed planning helps to ensure that all of the bases are covered and that there are people available with the necessary skills and capacity to perform the work.

The approach to emergent work takes a different path out of necessity. The greater levels of uncertainty demand use of an adaptive approach, making use of a series of small, frequent experiments in order to validate our assumptions about what is truly valuable and how to best go about delivering that value. These are our all-important feedback loops designed to inform and guide us.

A key implication with an adaptive approach is that we focus on defining and delivering against a desired outcome without a defining a long-range, detailed plan up front. We keep our options open and seek to uncover the best possible solution by iterating over the problem, incrementally building value.

Waste will occur if we use an ordered approach to emergent work because the greater levels of uncertainty negate our ability to predict very far into the future. Striving to define a long-range, detailed plan up front forces us to make decisions based on untested assumptions. This can lead to cost overruns and late deliveries due to re-work if those assumptions are found to be invalid.

This does not mean that detailed planning is eliminated! Instead, detailed planning is spread out across those previously mentioned, small experiments that are designed to contribute in some way towards an explicit outcome: we create a hypothesis that articulates the measurable benefit we are striving to achieve, run a quick test, evaluate the results, and if required, we adjust and conduct another experiment to further validate our understanding. If no adjustment is needed, we press on with our next “experimental increment.”

And of course, we must deal with chaotic work, which are those urgent, unplanned issues that arise. The approach best suited here is to triage each issue and respond accordingly. Some issues, like those that are preventing or seriously hampering day-to-day business operations, require an immediate response and will interrupt other, planned work in progress. Lesser-critical issues can be prioritized and dealt with alongside other planned work, with the first response being a communication back to the reporter regarding the disposition of the issue so that they aren’t left wondering what is going on.

Taking a step back, we can see that for each approach for each type of work, there is a primary objective:

A key message in all of this: Achieving optimal results is a function of each approach being fit for the purpose. 

Although agility is intended to help us deal with the increasing amount of emergent work present in today’s globally-competitive climate with high rates of change, agile teams actually make use of all of these approaches in their day-to-day work. They triage and address production issues. Sprint planning and execution (assuming the use of Scrum, the most popular team-based framework) is all about creating and working a detailed plan to deliver. And teams inspect and adapt leveraging short feedback cycles.