6 Lean Metrics to Increase Speed of Delivery
Want to Move Faster? Don't Focus on Speed
Every team has heard this at one point or another: We need to move faster. In order to gain or maintain a competitive edge, leaders charge teams with the task of completing the same high-quality, high- value work -- just faster.
In an effort to meet the charge, many teams develop unsustainable, bad habits: They work haphazardly, spend less time planning and more time doing, take fewer breaks, and usually, work longer hours. With the goal of “speed” in mind, they work themselves into the ground -- usually without increasing speed in a way that adds value to the customer (at least, not in the long run).
When we focus intently on the goal of moving faster, we often do so at the risk of sacrificing the things that make our work valuable: intention, focus, and ultimately, quality.
To Improve Speed, Optimize for Flow
Luckily, there is a way to deliver work faster without sacrificing quality, focus, morale, and sustainability. To do this, we need to shift our focus away from speed, and commit instead to improving flow.
In Lean thinking, flow refers to the manner in which work progresses through a system. 83% of teams practicing Lean use Kanban to visualize and actively optimize for flow.
Lean teams know that in order to increase speed, they have to optimize for good flow. “Good” flow describes a system in which work moves through at a steady, predictable pace. “Bad” flow is when work stops and starts frequently. Optimizing our processes for a consistent, predictable flow of work is essential for faster and more reliable delivery. Often, our issue isn’t speed, it’s focus. When pressured to move faster, we often work in a way that is more chaotic and disorganized, which actually slows us down.
By aligning team members around a shared understanding of what’s in progress and what’s up next, a well-designed Kanban system can help keep everyone focused on getting prioritized work done as efficiently as possible.
The Power of Lean Metrics
When used in conjunction with a well-designed Kanban system, Lean metrics can provide actionable insight that can help teams answer important questions like:
- How long does it take for us to complete a piece of work once we begin working on it?
- What’s the average cycle time for each type of work that we do?
- Where does work get stuck? How long does it wait between active steps?
- Are we underestimating how much time it takes us to complete a request?
- Are we overloading any particular team member? How can we relieve this burden?
Note: Proceed with Caution
Before you begin analyzing metrics, it’s important to ensure the quality of your data. Make sure your team is using your board in a consistent way before you begin analyzing metrics. Before drawing any conclusions about your data, ask yourself these two questions: What does this metric truly measure? What insights am I trying to learn from this data? The answers to those two questions should be fairly similar. Data is only useful when it is analyzed accurately, in context.
6 Lean Metrics to Increase Speed
WIP (Work-in-Process) Limits
Practically speaking, work cannot add value to the customer, team, or organization unless it’s finished. Tracking work items that are started, but not yet finished, can help teams improve the overall flow of value through the system, and thereby deliver work faster. WIP limits help teams stay focused on completing existing work before starting new work, so they can get work done quickly with greater efficiency.
WIP limits are constraints on how many work items (cards) are actively being worked on at any given time. They can restrict how many cards are allowed in a specific lane or set of lanes, or can be applied to specific individuals. Learn more about WIP limits and why we need them in this post.
In LeanKit, teams can configure their boards to alert team members when the WIP limit is reached for a certain lane or set of lanes (learn how here). If a team member decides to override and exceed the WIP limit, the lanes will turn red. The lanes will stay red until the board returns to an acceptable level of WIP. When the WIP limit is exceeded, this should be a signal to the team to work together to return the board to a healthier level of WIP by working together to finish work on another card as quickly as reasonably possible.
Keep note of how often the WIP limit is exceeded in a given time period, and for what reason. If your team consistently meets or exceeds its WIP limit, dive in to learn whether your team is missing opportunities for collaboration that could help you move work along faster. If your team never reaches its WIP limit, the limit might be too high, which means you won’t experience the benefits that come from limiting WIP.
Note that WIP limits should be slightly painful -- the purpose is to keep us focused on delivering work to “Done” as efficiently as possible, which goes against the chaotic, multitasking ways we’re accustomed to. Learning to implement and respect WIP limits requires teams and the individuals within them to break a lot of bad habits: multitasking, committing to too many things at once, working without regard to team capacity, and more.
If your team wants to improve its speed, it has to commit to working with more discipline than ever before, and WIP limits can help provide that discipline. As a team, read and discuss this post on why we need WIP limits, then use this 5-step guide to get started.
Queues (Wait States)
When teams try to increase their speed of delivery with Kanban, they often try to reduce the amount of time it takes to complete each step in the process. This is based on the (often flawed) assumption that it’s the active work steps that are taking the most time to complete. In reality, work often spends much more time in the wait states between active steps than it spends actively being worked on. Limiting the time that work spends in queues, or wait states, can help teams drastically improve speed.
An efficiency diagram measures the difference between total WIP and the work that’s waiting in queues. It illustrates when the work in queue is growing as a percentage of the total WIP. This can help teams pinpoint where work is likely stuck in a queue, and also investigate what can be done to get work flowing again.
Make sure your team has been using your Kanban board consistently for several weeks before diving into efficiency diagrams, to make sure that you are truly analyzing patterns in your workflow.
In Kanban systems, teams typically use blocker icons to visually communicate that a piece of work cannot move forward in the process. Unlike work items waiting in queues, which are just waiting to be pulled into the next step, blocked work items are typically waiting on an external dependency or some failure condition. Blockers serve a dual purpose: they signal to the team that a piece of work needs immediate attention, and also serve as a valuable flow metric.
As you use your Kanban board, simply count how many items are blocked at any given time and regularly assess how long work items typically stay blocked (by reviewing the card history of each blocked item).
In LeanKit, when you block a card, you are prompted to provide a reason. If your team members are curious as to why a card is blocked, they can hover over the blocker icon and the reason will appear.
As a team, analyze patterns in not only how many items are blocked, but why they are being blocked. Consider the question: Could we reduce the amount of time work spends being blocked? If the most common reason for blocking a piece of work is because you are waiting on external feedback, could you alert the person providing the feedback in advance so that they are ready to review the work as soon as it is ready to be reviewed?
These kinds of process improvements can drastically improve your team’s efficiency and overall flow. Learning to avoid and mitigate blockers can help your team keep work moving through your board with greater efficiency and fewer disruptions.
Lead Time and Cycle Time
When teams receive an order to deliver faster, it’s usually because of an external perception that their lead time is too long. Lead time measures the total time it takes for work to be completed from the moment the work is requested to the time it’s delivered. This includes process time, as well as time that work spends sitting in queues. Work can be delivered internally, to another team, or externally, to a customer.
Long lead times for prioritized work are often an indicator of an inefficient system -- so the question becomes: Why does work take so long?
In most cases, lead time metrics alone cannot provide teams with the insights they need to improve their lead times -- with just lead time metrics, it’s difficult to tell whether it’s the active process steps or the wait states that have the greatest room for improvement. It’s hard to understand which parts of the process usually take the longest. This is where cycle time metrics become very useful.
Cycle time measures the time it takes for work to get from Point A to Point B. It can measure the time between any two points in the process, such as the time between when a piece of work is pulled into the first active lane in a process, and when it reaches the “Review/QA” step.
Often, teams measure the cycle time of individual process steps. These are described by the name of the step: Design Cycle Time, QA Cycle Time, Review Cycle Time, etc. Analyzing each of these process cycle times can help teams gain a better understanding of where work gets stuck, which parts of the process might need more resources, and other opportunities to improve the flow of work.
After using cycle time metrics to determine opportunities for improvement, reviewing lead time metrics can help teams understand the impact of any changes they make to their overall process -- for example, did hiring a second QA engineer (based on data that showed that QA is consistently overburdened) improve our ability to deliver work more predictably?
Analyzing lead time metrics of various work types can also help teams communicate realistic estimates to anyone who requests work from them. For example, if an IT Ops team is performing routine system maintenance, it can measure its past performance to predict the lead time of its current work and provide accurate estimates for customer downtime. Without lead time metrics, any estimates given to the customer would be educated guesses, at best — and without tracking lead time on current work, teams would not know how to improve.
Throughput is the average number of units processed per time unit. In a Kanban system, examples can include “cards per day,” “cards per week,” or “story points per iteration.”
In an ideal Kanban system, a card would represent a standardized “size” of work. While this isn’t entirely possible, especially on teams that perform many different types of work, it is a good goal to aim for. At LeanKit, we try to size our cards so that they move at a consistently fast pace -- once every day or so, if possible. Breaking work into manageable tasks helps team members stay focused on delivering value as quickly as possible, without delving into the dangers of scope creep. Use throughput to gauge your team’s productivity and assess the impact of various improvements on your overall productivity.
Based on your team’s specific goals, choose a few metrics to regularly analyze -- we recommend starting with WIP limits for these reasons. Take some time to understand what a “normal” set of data looks like for your team. Then, brainstorm ways to improve your processes and implement and measure them following the continuous improvement cycle (pictured below).
The charge to “move faster” doesn’t have to be met with anxiety, longer hours, and worsened work quality. Use these 6 Lean metrics to optimize flow, so you can deliver more value with greater efficiency and speed than ever before.