Re-Read Saturday, Project to Product, Week 5, Chapter 4: Capturing Flow Metrics

As a reminder, the Flow Framework defines four flow items (things we are interested in measuring in this framework). They are:

Features
Defects
Risks
Debts

Using the framework is premised on a lot of things, however, two of the most critical are value streams and goals for how those value streams are going to operate. Everything starts with value streams. Without a clear understanding of how work is organized to deliver value to the market, organizations are operating blindly.

The second critical item is the goal for what will be delivered and worked on. The distribution of flow items is the quantitative instantiation of both where you are to today and where you should be heading. Kersten defines Flow Distribution as:

The proportion of each flow item within a value stream, adjusted depending on the needs of each stream to maximize business value.

Each value stream generates a mix of the four flow items: a distribution. How that actual distribution compares to the target (plan) distribution provides input into how the value stream is performing and whether action is required. A value stream whose planned output is 80 features, but spends half its time on defects is off track. Understanding the distribution of flow items and comparing that to a target helps to identify the different contexts impacting each value stream.

As I noted while re-reading earlier chapters, most organizations struggle with defining value streams, therefore they don’t know what the impact is of changing the mix of the flow distribution. Again, flying blind.

The four components of the flow distribution reflect the investment pattern within the value stream. Changing where effort and money are spent will change the distribution. The outcome, however, is not a linear relationship over the long run. For example, a few years ago I was asked to help an organization that was grinding. They had been following a strategy of rapidly building the feature base of their flagship product. The intent was to continue that approach, and the product management team had a high gloss product plan with feature drops for the next few years (all of this had been shared with analysts). At least internally, those plans were becoming more aspirational than probable. It did not take long to find that a huge amount of technical debt had been accrued and that the product teams were dealing with large numbers of defects. Neither technical debt or defects were measured. All of the elements of the Flow Framework need to be measured. It is much akin to trying to drive on the highway without a functional speedometer. You can gauge progress visually by looking at other cars around you but who knows whether you are all 10 miles an hour over the speed limit or not. What my clients had to wrestle with was that the flow distribution was a zero-sum game. The focus on features had to come at the expense of something else. Even adding people and resources, making the pie bigger, doesn’t necessarily change how the pie is sliced over the long run. Changing the distribution requires changing the goal and the behaviors of the people in the value stream.

A final note on Chapter 4, the four flow items have to be tied explicitly to business value. Beware of measuring proxies; they can lead you astray.

Buy a copy today https://amzn.to/2WzvPac