Project To Product
Leadership

Project To Product

Mik Kersten
Read 8 August 2020

Review

The Flow Framework is a compelling frame for a problem that most technology organizations genuinely have. I came to this expecting a practical guide I could run inside a large organization, and I didn't quite get that. Kersten is precise on the diagnosis: project-oriented management creates the wrong visibility, optimizes for the wrong things, and breaks down at scale in ways that Agile and DevOps alone haven't fixed. The four flow item types (features, defects, risks, debts) and the metrics built around them give a leadership team a vocabulary that most never develop. Implementation is another matter. Without a serious organizational commitment, or some external help, landing this inside a large enterprise is an uphill battle. What stayed with me is a sharper understanding of the zero-sum tradeoffs between features, quality, and risk, and why those tradeoffs stay invisible as long as you're running on a project model.

Key Takeaways

The parts worth keeping

The core diagnosis

IT and digital transformation failures trace back to a structural mismatch: organizations run software delivery like a series of projects when they should be running it like a product portfolio. This shows up as three recurring disconnects: project focus over product focus, cost focus over profit focus, and timeline focus over business value delivery.

The three epiphanies

  1. Software productivity declines as software scales because the architecture and the value stream become misaligned. Realigning daily work around the value stream instead of the architecture increases productivity.
  2. Disconnected software value streams are the primary bottleneck to software productivity at scale. These disconnects are caused by the misapplication of the project management model.
  3. Software value streams aren't linear manufacturing processes. They are complex collaboration networks that need to be aligned to products.

The Flow Framework

The Flow Framework connects business and technology by making value stream delivery visible and measurable. It has three components: a product-oriented mindset (treating software as evolving products, not projects), Value Stream Networks (the infrastructure that provides visibility and automation), and four MECE flow items that define what gets tracked.

The four flow items for any software value stream:

Flow ItemWhat it is
FeaturesUnits of functionality delivering business value
DefectsBugs, quality issues, escaped problems
RisksSecurity, regulatory, and compliance work
DebtsTechnical debt that impairs future flow

The five flow metrics

Business results per value stream

Each value stream should be tracked against four categories of business results:

CategoryExample metrics
ValueRevenue, MRR, 7-day active users
CostTotal staff, operations, and infrastructure spend
QualityEscaped defects, NPS, retention
HappinesseNPS, employee engagement

Four questions everyone in the organization should be able to answer

Software vs manufacturing: key differences

Five sources of waste (Dominica DeGrandis, Making Work Visible)

  1. Too much work in process: overloading teams leads to inefficiency and reduced flow.
  2. Unknown dependencies: unrecognized dependencies between teams, components, and products generate unplanned work.
  3. Unplanned work: emergencies and rework disrupt planned activities and introduce bottlenecks.
  4. Conflicting priorities: misaligned goals across the business impair decision-making and delivery.
  5. Neglected work: deferred technical debt and zombie projects accumulate and hinder flow.