Value Stream Management: Navigating the Stream of Continuous Delivery”
Introduction to Value Stream Management (VSM)
Challenges that VSM Addresses
At the highest level, it’s the ultimate challenge; how to thrive in the digital age. There’s a complex system of levers needed to digitally transform, underpinned by the two key enablers: cloud and DevOps. Transitioning from a traditional dominator hierarchy, where work is constructed and delivered around projects in a waterfall manner and the system is riddled with dependencies to a DAO with autonomous, multifunctional teams with loosely coupled systems working in small batches and increments is a long and arduous journey. Business as Usual (BAU) has to stop. Time has to be found to unlearn, learn new ways of working, change behaviors, processes, and system architectures.
VSM highlights where to spend that precious time to maximum effect, accelerating the journey. It measures progress on the journey making experiments empirical, giving leaders the confidence positive change is happening. Insights into all value stream steps and traceability between them make the work visible and surface the improvement opportunities. Teams are no longer blind; they know the pace of their work, they know the impact of their efforts. They can make informed decisions about what’s next and understand the outcomes from their customers’ perspectives.
How to Optimize Value Streams
It’s only possible to know for sure that improvements are happening if they are measured. Agile teaches an empirical and experimental approach where we imagine the outcome of our actions in a hypothesis and visit the change at a specific time (or times) in the future to see what actually happened.
When optimizing flow, the lean waste acronym ‘DOWNTIME’ is useful to follow:
Example hypotheses formats used for optimizing VSM flow are:
If we do to remove in our value stream, we think that our cycle time will improve by on
When we automate it will reduce the processing time from to in this
By improving our flow time by <% or multiple> we will be able to deliver more enhancements to our customers, worth in outcomes
Improvements to value stream flow are made by removing:
Waste
Delays within steps and between steps (wait time)
Non-value adding activities
Examples of interventions to optimize digital value streams based on good DevOps practices include:
Shortening the time between generating an idea or receiving it from customer feedback to starting work on it through small batch, autonomous product ownership
Using continuous integration and associated test automation to merge and accelerate software delivery and validation steps
Practicing continuous delivery to accelerate frequency and volume of value outcomes to customers
Shortening the Route to Live (RtL) using cloud, IaC and self-service environment provisioning practices
Using limited blast radius and dark launch techniques such as blue/green deployments and canary releases to deliver faster with less risk
Automating the deploy and release processes using standard configurations and processes to accelerate the step and mitigate interdependency risk
Using cloud-native architectural techniques such as microservices to create loosely coupled environments where system components can be independently built, tested and deployed
Replacing heavyweight Change Approval/Advisory Boards (CABs) with automated checklists built into the CICD pipeline/DevOps toolchain and lightweight peer-based reviews
Replacing heavyweight incident management processes that cause exacerbate unplanned work with swift, intelligent swarming
Employing DevSecOps practices and tools to reduce delays caused by wait time on security teams
Organizing around value streams as multifunctional, autonomous teams to reduce handoffs
In order to measure the impact of these interventions on flow, teams must have realtime data on processing, wait and cycle time available from their preferred toolchain. To achieve that goal, they’ll need a VSM tool, like Plutora.
The Value Stream Manager’s Job
Fully embracing VSM requires the introduction of value stream-specific roles. The most essential role to assign is that of the Value Stream Manager. Without it, there is no fulcrum of accountability for the value stream’s operation. DevOps and therefore VSM promote multifunctional, autonomous teams and also multifunctional people; roles are valued more than job titles. This role then can be assigned to an individual who also carries other roles, for example, Product Owner or Scrum Master.
The duties of the Value Stream Manager include:
Defining the product family by conducting product routing analysis and appropriate groupings
Ensuring that a current state value stream map is created of the end-to-end value stream
Conducting fact-based analysis of the current state map
Preparing future state maps showing what the value stream could look like in the long term
Preparing target state maps that use lean techniques to eliminate waste and improve process value in the short to mid-term
Creating a plan to achieve the future state along with OKRs and KPIs
Ensuring work items are in the value stream backlog and are refined, prioritized and delivered
Implementing a VSMP to automate the generation of maps and improvements
Definition of metrics and enablement of continuous inspection and adaptation
Leading the continuous implementation of the plan
Leading and mobilizing the people inside and outside of the value stream to enable the required changes
Leading the day-to-day activities within the value stream to ensure that current commitments are achieved while improvements are being made
Increasing the ratio of value adding activities to non-value adding activities
Eliminating waste through the end-to-end value stream from idea to realization
Ensuring optimal customer experience, gathering and acting on feedback
Measuring flow and improvements to cycle time through experiments
Managing and breaking dependencies with other value streams
Treating the value stream as a microbusiness and managing the P&L
Myths About VSM
Complex systems of thought and ways of working are always misunderstood and misinterpreted by some of the marketplace, particularly when they exist in their early stages of evolution and VSM is no different. Myths can be useful as they offer a framework for positive definition, discussion, and challenges around the principles and practices. Here are some of the common myths around VSM and their corrections:
Myth 1: Value Stream Management is Value Stream Mapping.
Value Stream Mapping is currently more widely understood and practiced than full Value Stream Management and the two are frequently conflated. Mapping is a sub-discipline of the group of practices that comprise Value Stream Management and an essential component but in digital value streams, it must be combined with organizational change and continuous inspection and adaptation of data from DevOps toolchains to fully deliver on its promises of organizational performance improvement.
Myth 2: Value Stream Management is only for large enterprises.
Anything that has a customer or delivers a product or service has a series of steps or value streams that are followed and completed for the successful delivery of a customer experience that has a value outcome. Even if a business is brand-new or consists of a single value stream, there are benefits to be gained from optimizing flow and feedback to and from the customer.
Myth 3: Value Stream Management and DevOps are incompatible.
The concern derives from the reality that DevOps toolchains are already hugely complex in their inherent heterogeneity and their cross-organization variability. The requirement to maximize the data available from the DevOps toolchain requires the addition of another component, the VSMP, to access the insights essential to optimization. But a VSMP connects all the tools in a way that is unachievable, or incredibly time-consuming, to do manually and makes the insights available so that teams don’t have to do this work themselves. VSM is either next-generation DevOps or DevOps is the intervention handbook that powers VSM. Either way, they don’t conflict but the combination of the two is the most powerful way to delight customers through the fast flow of enhancements that deliver measurable value outcomes.
Myth 4: Value Stream Management creates a blame culture and gives a new stick to management.
Actually, VSM, like DevOps, promotes the cultural imperative of the Distributed Autonomous Organization (DAO). Traditional management doesn’t exist - the value stream is managed, not the people. Leaders’ roles are to help teams self-discover improvements. Value stream teams define their own OKRs and KPIs in line with the organizational vision and they are accountable for continually inspecting, adapting, measuring progress and surfacing and sharing success to make local discoveries global improvements.
Myth 5: Software developers don’t care about VSM.
Every role in the value stream is required to understand their step in the end-to-end value stream. In a digital value stream, software developers are accountable for the most value-adding step. Developers care enormously about customer experience - it’s a sign of their pride in their work and their craft. VSM, like DevOps, cares about employee engagement and burnout. It knows that autonomy, mastery, and purpose are key to happiness and productivity for knowledge workers. Aligning individual purpose with organizational purpose is one of the biggest challenges businesses face today. Providing developers with real-time feedback on the impact of their work on the people that use their creations is a massive enabler for the continual strive for optimal customer experience. And great customer experience equals high organizational performance. Joy is found in progress and success.
Myth 6: Value is indefinable and unmeasurable.
It’s true that value is in the eye of the beholder, but there are many ways to see and measure customer experience, understand value outcomes of changes to digital products and services and learn from this feedback. Lagging indicators like sales, revenue, and profit and intentional indicators like reviews and referrals, conversions, retention, customer journey times, basket size, and bounce rates for example. Value Stream Managers are accountable for the P&L of their value stream through continuous inspection of the impact of improvement experiments on customer experience and behavior and adjusting the costs of sale and production in line with income.
Myth 7: We have too many interconnected value streams to make managing them possible.
Actually, this is one of VSM’s strengths; making the disparate value streams visible, the connections, contentions, and conflicts between them. VSM makes it possible to manage dependencies and shows where to focus effort on breaking them.
Myth 8: DevOps is enough.
Research shows that DevOps evolution is stuttering in many organizations and VSM is the key to unsticking the enormous change required for digital transformation and enabled by DevOps and cloud. Teams struggle to know where to make improvements and can’t see their work or their progress. Leadership lack anything other than narrative to see how their efforts and investments are having an effect. VSMPs make all this visible and DevOps delivers on its promises.
Myth 9: I can do VSM without a VSMP.
Sure you can, but why miss out on using all that data you’ve already invested in collecting and augmenting your teams’ cognitive load by providing them with insights that will improve organizational performance? Automating your value streams, their maps, and being able to continually inspect and adapt them is the optimum improvement position.
Myth 10: We don’t have time to implement VSM.
The real question is, can you afford to not implement VSM? Research shows that VSM adoption correlates with organizational performance so businesses that don’t harness these ways of working will lose pace with those that do. VSM is key to business survival in this chapter of the digital age.