OUR THOUGHTSStrategic Advisory
Unlocking the power of flow metrics
Posted by Steven Gibson, Don Smith, Megan Barnes . Sep 26.24
Our product landscape has become increasingly customer-centric. With the speed of market disruptors, organisations are constantly seeking ways to improve their ability to deliver value to customers and to pivot and respond. One powerful tool that has gained traction in recent years is the use of flow metrics. However, unlocking the power of flow metrics requires a holistic approach that goes beyond just measuring single team performance. In this article, we’ll explore the challenges and offer some guidance for leveraging flow metrics to optimise your entire value stream.
The importance of end-to-end flow metrics
When implementing flow metrics, it’s important to consider the entire value stream, not just the engineering portion. This means tracking the flow of work from initial ideation to delivery to the customer. By taking this comprehensive approach, organisations can identify bottlenecks and opportunities for improvement across the entire process, maximising the value of investment to optimise flow in the value stream.
One of the key challenges in implementing end-to-end flow metrics is getting buy-in from enough stakeholders involved in the value stream. This can be particularly difficult when different teams have their own established processes and budgets, leading to local optimisations, which challenge funding and coordinating improvements across teams. To overcome this, it’s essential to communicate the benefits of flow metrics and how they can help improve overall efficiency and value delivery.
Defining the start and end points of flow
A question that always comes up when implementing flow metrics is determining when the ‘timer’ starts and stops for each piece of work. This decision can have a significant impact on the insights gained from the metrics and should be carefully considered.
For the starting point, organisations need to decide whether to begin tracking work from ideation, the start of feasibility studies, or another trigger that makes sense for the organisation. If the timer is only started when work is handed off to engineering teams, improvement opportunities are often missed. The choice of when to start the timer should be as close to ideation as possible to capture high-value improvement opportunities. It also needs to align with the organisation’s goals and not stretch too far outside the sphere of control of the value stream.
Similarly, the endpoint of flow can vary. Often the timer stops either:
- When all stories in a backlog are ‘Done’
- When a feature is deployed to production
- When a feature flag is turned on
- When the deadline date arrives
The main point to recognise is that end-to-end flow may expand further than the bounds of product development teams so it’s important to have clear conversations with stakeholders to agree on these definitions upfront. This ensures that everyone is aligned and that the metrics provide meaningful insights.
Ideally, the timer stops when value is delivered to the customer and after feedback has been collected. However, efforts outside of a development team may be required after deployment, such as the release process, so being pragmatic about when to start can help build confidence before tackling larger systemic bottlenecks that span multiple value streams and centralised functions.
Balancing portfolio management and flow metrics
Another challenge in implementing flow metrics is striking the right balance between maintaining a healthy backlog and avoiding unnecessary work. While it’s important to visualise all work in the system for transparency, creating tickets for every potential future backlog item can lead to bloated backlogs, increased coordination costs and cognitive load.
A good rule of thumb is to only create backlog items for work that has a reasonable chance of being addressed within the next six months. This just-in-time approach helps maintain focus on near-term priorities, while still providing visibility into upcoming work.
For longer-term planning, consider using higher-level artefacts like Epics or Features rather than detailed user stories. This allows for strategic planning without overwhelming backlogs with low-level details that are likely to change before implementation.
Looking beyond team-level optimisation
Optimising team processes alone may not yield the most significant improvements in flow. One of the most valuable insights from implementing end-to-end flow metrics is in their ability to pinpoint persistent bottlenecks, helping identify the biggest opportunities for reducing flow time anywhere in the value stream.
For example, the time spent on discovery, feasibility studies or market research may dwarf engineering implementation time. Similarly, significant delays may be caused by the process of releasing and rolling out new features to customers.
Once organisations have built momentum and habits around flow metrics, visualising the entire value stream helps identify bottlenecks and focuses improvement efforts where they’ll have the most impact. This might involve streamlining the product discovery process, improving release processes, enhancing communication between teams or addressing organisational policies, approval processes and decision-making.
Linking flow metrics to business outcomes
While flow metrics are excellent for measuring your ability to build the thing right, they don’t inherently measure how well you are building the right thing. To truly optimise your value stream, it’s essential to link flow metrics to business outcomes.
This means tracking not just how quickly work moves through your system, but also the impact that work has on key business metrics driven by the strategic portfolio or organisational OKRs, KPIs or goals. The Flow Framework® categorises business results into four areas – new business value, reduced costs, quality and happiness. It’s a helpful tool for connecting flow metrics and business outcomes.
By focusing on business outcomes, work can be prioritised based on value impact, even if it means having less in your backlog.
Avoiding the pitfalls of metric gaming
As with any metric-driven approach, there’s always a risk of groups trying to ‘game’ the system to improve their numbers without actually improving the underlying processes. This can lead to misleading data and suboptimal decision-making.
To avoid this pitfall, it’s important to foster a culture of transparency and continuous improvement. Emphasise that the purpose of flow metrics is to gain insights and drive real improvements, not to achieve arbitrary targets. Encourage both leadership and teams to use the metrics as a tool for learning and optimisation rather than a performance evaluation mechanism.
Some strategies to prevent metric gaming include:
- Focus on trends rather than absolute numbers. Look at how metrics change over time rather than fixating on specific targets.
- Use multiple metrics for a holistic view of your system. Relying on a single metric can lead to unintended consequences. Use a balanced set of metrics to get more comprehensive improvement results.
- Encourage open discussion about challenges. Create an environment where teams feel comfortable discussing obstacles and bottlenecks without fear of repercussion.
- Align metrics with business outcomes. By tying flow metrics to actual business results, you reduce the incentive to game the system for its own sake.
- Leadership can value and encourage teams to use these measurements to aid the team’s decision-making rather than use them for judging the team’s performance. If the teams are using these metrics to successfully improve, that is more than sufficient.
Remember, the goal is to represent reality accurately and use that information to drive meaningful improvements. Any attempt to artificially manipulate the metrics ultimately undermines this goal and reduces the value of the flow metric approach.
Summary
Implementing flow metrics across your entire value stream can be a powerful tool for optimising your software delivery process and driving business value. By taking a holistic approach that considers all stages of the value stream – from ideation to customer delivery – organisations can gain valuable insights into their processes and identify opportunities for improvement.
Key takeaways for successful implementation of flow metrics include:
- Seek buy-in from all stakeholders and agree on clear definitions for the start and end points of flow.
- Balance the need for visibility with the practicality of backlog management.
- Look beyond team-level optimisation to identify bottlenecks to the left and right in other areas of the value stream.
- Link flow metrics to business outcomes to ensure you’re optimising for value, not just efficiency.
- Foster a culture that views metrics as a tool for learning and improvement, not just performance evaluation.
By following these key takeaways, organisations can leverage flow metrics to create a more effective, responsive and value-driven software delivery process. Optimising your value stream is an ongoing process and flow metrics provide a valuable compass to guide continuous improvement efforts.
As you embark on or continue your flow metrics journey, keep in mind that the ultimate goal is not perfect metrics, but rather a deeper understanding of your processes and the ability to make data-driven decisions that positively impact your business and customers. With patience, commitment and a willingness to learn and adapt, flow metrics can become a cornerstone of your organisation’s success.
More Ideas
our thoughts
Unpacking the SCARF Model
Posted by John Stephenson . Oct 03.24
Have you ever found yourself or your colleagues reacting strongly to certain situations at work, seemingly out of nowhere? Maybe it was a change in job title, a shift in responsibilities or feeling left out of a meeting. John Stephenson unpacks the SCARF Model, a framework for motivating and understanding such behaviour...
> Readour thoughts
Improving flow with layered test automation
Posted by Gareth Evans . Sep 19.24
The ability to deliver high-quality code quickly and consistently is paramount. Two concepts that have revolutionised our approach to this challenge are the Test Automation Pyramid and the principles of flow and feedback. In this blog, Gareth Evans explores how these ideas intersect and how adopting the Test Automation Pyramid can significantly enhance your development flow and feedback loops.
> Readour thoughts
Value is a flow
Posted by The HYPR Team . Sep 08.24
The HYPR team came together at their latest WorkerBee event and discussed why ‘building the right thing’ can’t be separated from ‘building the thing right’...
> Readour thoughts
Part 1 – Using OKRs to drive strategic growth
Posted by Martin Kearns . Sep 02.24
In part one, Martin Kearns takes lessons from the track and looks at how using OKRs drives strategic growth, as well as leveraging short-term milestones to test assumptions and stay on track.
> Readour thoughts
Event modeling: A socio-technical approach to system design
Posted by Arjan Noordhoek . Aug 29.24
Event modeling is a software design and development approach introduced by Adam Dymitruk. It’s a method for designing and building systems based on capturing and modeling the events that occur within a business domain.
> Read