What is flow and, most importantly, why does it actually matter? Enabling flow and keeping your attention on the main flow metrics will not only enable you to deliver results in a consistent predictable manner, but it’ll also pave the way for reliable, data-driven decisions.

What is Flow?

Imagine: you’re at the security screening checkpoint at an airport. What does the process look like? Airport security is a standardized system. Step by step, people flow through the activities of that system – prepare for passport check, walk to the conveyor belt, pull out all their electronics, place their bag on the belt, walk through the scanner, pick up the bags at the other end, put all electronic devices back into said bags, and then you’re ready to go.

The system’s capacity, or the number of people that can be screened at any one time, can be affected by many factors — the number of security lanes, the number of available agents, the processing speed of the technology used to perform the screening, just to name a few.

And there are plenty of situations that can interrupt that flow. It could be that one person takes a particularly long time to take their devices out. Or, someone sets off the metal detector and has to go back, empty their pockets and try again. Or, someone with a disability could need additional assistance getting through the line.

The more people move through the system, the higher the likelihood that the process will take longer than usual.

Understanding Flow in Knowledge Work

When your customers make a request, there is a process that you will follow in order to achieve the delivery of that request. Essentially, flow in knowledge work is the movement of the requests through your process.

And our ultimate goal is to optimize flow so that we can improve the way each piece of customer value moves through our system so that it can be delivered to our clients in a more efficient, predictable manner.

Our aim is to enable a predictable delivery workflow that suits our customers’ needs. And the main flow metrics are at our disposal, to help us track the performance of our system.

How to Measure the Flow of Work

When it comes to measuring and evaluating the predictability of your workflow, there are four very simple metrics we recommend starting with. And these metrics are tightly coupled with how efficiently your system is operating. Let’s go through each of them.

Flow analytics

Cycle Time

So, every time you commit to delivering a new customer request, the first question you need to answer is “When will it be done?”. And what our clients actually care about is, when are they going to receive what they’ve requested? What is the elapsed time from the moment they provide the request, up until the moment they will receive it?

In our workflow, we start the timer when the work has been committed to, and we stop the timer when the work has been delivered. The difference between these two points in time is called cycle time.

Cycle time is a measure of how long it takes for an item to flow through your delivery system. Assessing your cycle times and how the trends build over time is an excellent way to evaluate the predictability of your workflow, and to understand whether it behaves the way you expect it to.


In addition to knowing how long it takes to deliver customer value, a lot of agile practitioners are interested in how much work is being delivered for a certain period of time.

This flow metric is especially important if you have a preset release date and you need to know how many items can fit within the time you have in your hands. We call this metric throughput.

Throughput is simply a measure of how many items in our process we’re completing per unit of time – in terms of number of items per day/week/month etc.

Work in Progress

The third flow metric is Work in Progress (WIP). WIP represents the number of items available in your workflow. And that metric influences both how long it takes us to get things done, and how many things we can get done per unit of time.

As such, it is crucial to maintain WIP, and strive to keep it consistent, in order to make sure we make the most out of our flow. Consistent WIP leads to consistent cycle times and consistent throughput.

The main flow metrics are interconnected through the Little’s Law equation. Always keep in mind that changing one of the flow metrics will inevitably affect one (if not both) of the others.

You can use this concept to implement simple strategic tweaks that will improve your delivery speed. For example, reducing the number of items you are working on could have a dramatic impact in terms of how long it’s taking to deliver results. When the WIP is going down, cycle times are going down and throughput is going up, which is exactly what we strive for in order to achieve an efficient delivery workflow.

Average Age of Work in Progress

The last flow metric – and I’d argue the most important one – is average age of work in progress.

According to Little’s Law, maintaining stable delivery workflows depends on two factors – your work in progress and your average age of work in progress. For these two metrics, the key is consistency – predictable systems are determined by your ability to keep both your WIP and the average age of your WIP consistent.

The average age of WIP and cycle time are essentially the same thing, only cycle time is measured against completed tasks, whereas the age of a task is a measure concerning tasks that are still in progress.

Your cycle time metric doesn’t take into account your current WIP. Although you might have a very low average cycle time, if your tasks sit and age in the flow, then eventually get released at a later stage, your average cycle time will go through the roof. This is why it is so important to track your average age of WIP, and regularly compare it against your average cycle time.

Ideally, the numbers should be roughly equal.

It’s always best to look at all of these metrics in conjunction, in the context of how they relate with each other as part of the flow.

For example, getting the wrong things done faster won’t bring any value, and sacrificing the quality of your results for the sake of the speed of delivery itself is something that we would never recommend doing.

Don’t try to optimize any single one of these metrics in isolation, but rather use them together to better understand not only how the work flows through your system, but also as a method of assessing your process performance.

Now, it’s time to take action! Start measuring the main flow metrics and keeping track of how the decisions you make affect your workflow performance. Evaluate how the trends build over time, ideally on a biweekly basis, to understand whether your improvement efforts are paying off.

If your delivery system doesn’t produce the results you are hoping for and you’d like to explore the proven roadmap to optimize your workflows for predictability, I’d be thrilled to welcome you to our Sustainable Predictability program!

Do you find this article valuable?
Rating: 5 stars (10 readers voted)