Secrets of Speed #2: Cycle Time

In a previous post I described the mechanics of throughput — the amount of work that a team can do within a given time period.

While people usually think about throughput when they want a team to move faster most of th time what they really want is a shorter cycle time. Cycle time is the amount of time it takes me to get my stuff – not how much stuff the system produces.

Imagine a coffee shop with a line. If I walk into a coffee shop and see a very long line I’m likely to get discouraged because I think it will take me  long time to get my coffee.

I don't care how many coffees the baristas are making in an hour (throughput) nor do I really want them to hurry up and sacrifice quality (I'm pretty particular about my coffee). What I care about is how long it takes me to get my coffee. Cycle Time is the average amount of time it takes any given item to move through a system and get to done.

In a work environment when someone is upset that a team is moving slowly they are most likely upset that the one or two items they care about aren’t getting done. Unless they are a team member or direct supervisor it’s unlikely that they have attention on all work that the team is doing.

Cycle time is as much about perception as it is about system engineering.

Let’s assume for a moment that your team has done the hard work of learning to limit the number of items that are in the pipeline - or the number of items in production -  and thus maximize throughput. There can still be a perception that your team is moving too slowly by stakeholders outside the team if they aren’t getting the items that they care about in a timely manner.

Increasing throughput might seem like the most logical thing to do, but it’s not. . Instead, you’re likely to get only marginal gains since your team has limited capacity and you’ve already taken steps to maximize this. At worst you’ll end up taking on more than you responsibly can and either product quality will suffer or you’ll begin to stress people on the team – which we all know can lead to attrition and the loss of your best people.

So how do I get my coffee (work item) more quickly? The answer is pretty simple actually: You shorten the line. That may seem obvious, but what does it mean for us at work/mean when it comes to work flow?

In the real world, coffee shops lines are self-regulating. I walk into a coffee shop, look at the line, and make an assessment on the current average cycle time and whether I’ll be waiting based on the time I have.much I value that coffee and how much time I have available. I choose whether to stay or go. Overflow demand at one coffee shop will shift to other shops in the area. This “load balancing” can will also be addressed over time as other coffee shops open to serve the demand. This is how we end up with two Starbucks across the street from each other.

Some businesses however may choose to add capacity essentially splitting the line in two. This is why busy coffee shops have two cashier stations and two espresso machines. They can bring this new capacity online during busy times and let it sit during slower times. Essentially what they are doing is creating two lines where there previously was one.

You’ll also recognize this as an increase of throughput and it works as long as the system you’re working on is relatively simple and easy to understand.

There are other  ways businesses handle the cycle time  and issue of the queue length. Take popular nightclubs, which create a line outside only allowing the lucky few inside at any given time. Done well, this will shorten cycle time and increase the service level inside the club even as it can generate a bit of chaos outside the club.

This is the approach I suggest you take with work – create a more chaotic queue outside of the real queue.

To make this work you first need to acknowledge that when you commit to doing something it is now in your queue. Once I say, “yes, I’ll do that” the requester starts a mental stopwatch. They are waiting and thus judging your speed based on how soon they get that item.

So the only way to keep your queue short and your cycle time fast is through the (tried and true) lean practice of delaying commitment. You say, “not yet” when someone asks you to commit to doing something. It’s easier to hear that than the word “no,” even though in reality you may never commit to the item.

It may sound passive aggressive to say “not yet” with some consistency but, when you delay commitment all sorts of good things happen. Not only do you allow your team the ability to truly focus on the items they are working on, you also create a forcing function that requires the people requesting work from the team to prioritize. And, you give potential work items time to either mature or become irrelevant.  

Delaying commitment means you only commit to things you can get started on in the near term. “Near term” will vary business to business but a good rule of thumb is that anything more than three months in the future in today’s fast moving world is entering the realm of fantasy.

Often requests that seem urgent at first, will not ultimately age well .

Time provides an opportunity to test whether the idea remains important and still exciting as an opportunity. Like with a new relationship, it’s easy to get caught up in that shiny new object, the dazzling and exciting new thing. After the honeymoon wears off you can see whether that idea still holds up, if the initial enthusiasm remains and if it’s still worth pursuing.