Since I started my business, I’ve been constantly looking for opportunities for improvement. For me, these are optimizing the way we manage the work, improving our delivery speed, figuring out approaches to make my people more engaged and more motivated.
At some point, recognizing the problems that impede our performance has quickly become second nature. But, how do you get there?
What are the approaches to uncover the obstacles that hinder your predictability? Is there a method you can use to help you head in the right direction and keep your focus on the most impactful impediments right now?
Well, there is, in fact, a single straightforward approach that will help you understand where you need to focus your attention to improve the predictability of your delivery system. And today, I’ll share with you the 3 steps I take to reveal what’s slowing us down.
What Makes a Delivery System a Predictable One?
First things first, what is a predictable system, and, moreover, why does this actually matter?
Predictable systems enable you to give a confident answer to the common question “When will this be done?”. They make it possible to create accurate delivery forecasts using your past performance data, thus producing highly reliable service level agreements. And, furthermore, all of this can be achieved with less effort and time than what you’d spend using alternative approaches, like estimating your work.
Additionally, predictable systems prevent excessive work demand, overburden and sacrificing the quality of your work. They help you support a healthy flow of work and stick to your commitments.
There are two components that enable predictable systems – your average WIP and your average age of WIP. And the key here is consistency. Based on Little’s Law, as long as your WIP and your average age of WIP are consistent, you are maintaining a stable delivery system. And the more stable your system is, the more predictable it becomes.
How to Measure the Consistency of WIP and Average Age of WIP
To measure the consistency of your WIP and your average age of WIP, you should observe how the trends of these two flow metrics build over time. This is where the Aging Chart comes into play.
The Aging Chart enables you to track your current tasks in progress. It uses the same visual format as your Kanban board, with each column representing a state in your workflow. The Aging Chart shows how many days a task has already spent in progress.
The WIP widget on the top left side displays the number of tasks in progress for the selected date, as well as how your WIP trends have changed over time. By hovering over the line chart, you’ll be able to observe the number of tasks in progress for each date of the selected period.
The WIP Average Age widget shows the average age of all your work items in progress for the selected date. It also displays how your WIP age trends have built up (so far).
If your WIP and your average age of WIP stay roughly equal from day to day, then their trend lines will be linear; neither increasing nor decreasing over time. And these are the main prerequisites to maintaining a stable system.
How to Uncover the Impediments That Hinder Your Predictability in 3 Steps
There is a 3-step process you can follow to identify the obstacles that hinder your predictability. Let’s explore each of them in detail.
Step #1: Identify the Dates When WIP and WIP Average Age Trends Have Changed Significantly
Now, let’s burn some rubber and analyze the following example.
By hovering over the WIP average age trendline, you will be able to identify the average age of WIP for each day of the selected time period. Here, we see that on Aug 27th, the WIP age of the work was 8 days, which is higher than usual. On Aug 29th, it went back to its normal values, meaning the problem was resolved.
Now, what we want to know is what the problem was and what caused it in the first place.
Step #2: Single Out the Work Items That Caused the Deviations
We’d like to better understand what stands behind that high average age of WIP. To do that, we should go back to Aug 27th and analyze what the board looked like back then. This is where the “Aging replay” control comes in handy.
The “Aging replay” control is positioned on the top of the Controls section on the right sidebar. It enables you to change your basis date to see what your process looked like on a certain date in the past. Use it to go back to any past date and see how your work items were aging back then.
On Aug 27th, there was an item under Development that spent way more time in this process state than 85% of the completed items so far. This had a negative impact on our WIP average age and hindered our predictability.
Here’s the thing, by letting the work age in the process, you not only run the risk of breaking your commitments but also, once the work is eventually completed, it will affect the cycle times of the entire process, by introducing more variability to your delivery times. And the higher the variability of your delivery times, the less predictable your system becomes.
Step #3: Analyze the Root Cause of the Problem
Now that you know which item is hindering your predictability, it’s time to take action to make sure that you don’t end up in the same situation in the future.
Analyze the work and identify the root cause behind the delay, using the 5 Whys approach. Together with your team, think about what could be done to mitigate the risk of delay.
Most of the time, it will be just a matter of slightly tweaking the way you manage your work, either by introducing a new explicit process policy or changing the existing ones. Whatever it is, make sure you achieve an alignment between everyone involved and make it a standard, so that the impediment doesn’t arise again.
In this specific example, it turned out that this work item was blocked by a dependency on the legal team, which is what caused the delay. So, we decided to introduce a dynamic reservation systems to handle dependencies effectively.
To learn more about how to establish dynamic reservation systems and explore the proven roadmap to optimize your workflows for predictability, I’d be thrilled to welcome you to our Sustainable Predictability program!
So here are your action steps: in your next retrospective, go through these 3 steps to identify opportunities for improvement. Create an action plan (remember, always take it one small step at a time) and make a snapshot of your current flow metrics.
Put the plan into action and observe how the changes you’ve implemented are performing. Start your next retrospective by reviewing the results. Compare the snapshots before and after the initiative has been started. If you’ve achieved your goal, make it a standard. Otherwise, troubleshoot what caused the initiative to fail and brainstorm alternatives. Then, repeat the process.
Improvement is not a one-time event. Lasting change happens in increments, over time, with each new improvement building on the one before. Effectively, you’re setting up a system that self-optimizes for predictability and consistent business outcomes.
