When efficiency hurts more than it helps

When we imagine how to use a resource effectively – be that resource a development team, a CPU core, or a port-a-potty – our thoughts usually turn to efficiency. Ideally, the resource gets used at 100% of its capacity: we have enough capacity to serve our needs without generating queues, but not so much that we’re wasting money on idle resources. In practice there are spikes and lulls in traffic, so we should provision enough capacity to handle those spikes when they arrive, but we should always try to minimize the amount of capacity that’s sitting idle.

Except what I just said is bullshit.

In the early chapters of Donald G. Reinertsen’s brain-curdlingly rich Principles of Product Development Flow, I learned a very important and counterintuitive lesson about queueing theory that puts the lie to this naïve aspiration to efficiency-above-all-else. I want to share it with you, because once you understand it you will see the consequences everywhere.

Queueing theory?

Queueing theory is an unreasonably effective discipline that deals with systems in which tasks take time to get processed, and if there are no processors available then a task has to wait its turn in a queue. Sound familiar? That’s because queueing theory can be used to study basically anything.

In its easiest-to-consume form, queueing theory tells us about average quantities in the steady state of a queueing system. Suppose you’re managing a small supermarket with 3 checkout lines. Customers take different, unpredictable amounts of time to finish their shopping. So they arrive at the checkout line at different intervals. We call the interval between two customers reaching the checkout line the arrival interval.

And customers also take different, unpredictable amounts of time to get checked out. The time it takes from when the cashier scans a customer’s first item to when they finish checking that customer out is called the processing time.

Each of these quantities has some variability in it and can’t be predicted in advance for a particular customer. But you can empirically determine the probability distribution of these quantities:

Given just the information we’ve stated so far, queueing theory can answer a lot of questions about your supermarket. Questions like:

• How long on average will a customer have to wait to check out?
• What proportion of customers will arrive at the checkout counter without having to wait in line?
• Can you get away with pulling an employee off one of the registers to go stock shelves? And if you do that, how will you know when you need to re-staff that register?

These sorts of questions are super important in all sorts of systems, and queueing theory provides a shockingly generalizable framework for answering them. Here’s an important theme that shows up in a huge variety of queueing systems:

The closer you get to full capacity utilization, the longer your queues get. If you’re using 100% of capacity all time, your queues grow to infinity.

This is counterintuitive but absolutely true, so let’s think through it.

What happens when you have no idle capacity

What the hell? Isn’t using capacity efficiently how you’re supposed to get rid of queues? Well yes, but it doesn’t work if you do it all the time. You need some buffer capacity.

Let’s think about a generic queueing system with 5 processors. This system’s manager is all about efficiency, so the system operates at 100% capacity all the time. No idle time. That’s ideal, right?

Sure, okay, now what happens when a task gets completed? If we want to make sure we’re always operating at 100% capacity, then there needs to be a task waiting behind that one. Otherwise we’d end up with an idle processor. So our queueing system must look more like this:

In order to operate at 100% capacity all the time, we need to have at least as many tasks queued as there are processors. But wait! That means that when another new task arrives, it has to get in line behind those other tasks in the queue! Here’s what our system might look like a little while later:

Some queues may be longer than others, but no queue is ever empty. This forces the total number of items in the queue to grow without limit. Eventually our system will look like this:

If you don’t quite believe it, I don’t blame you. Go back through the logic and convince yourself. It took me a while to absorb the idea too.

What this means for teams

You can think of a team as a queueing system. Tasks arrive in your queue at random intervals, and they take unpredictable amounts of time to complete. Each member of the team is a processor, and when everybody’s working as hard as they can, the system is at 100% capacity.

That’s what a Taylorist manager would want: everybody working as hard as they can, all the time, with no waste of capacity. But as we’ve seen, in any system with variability, that’s an unachievable goal. The closer you get to full capacity utilization, the faster your queues grow. The longer your queues are, the longer the average task waits in the queue before getting done. It gets bad real fast:

So there are very serious costs to pushing your capacity too hard for too long:

• Your queues get longer, which itself is demotivating. People are less effective when they don’t feel that their work is making a difference (see The Progress Principle)
• The average wait time between a task arriving and a getting done rises linearly with queue length. With long wait times, you hemorrhage value: you commit time and energy to ideas that might not be relevant anymore by the time you get around to them (again: read the crap out of Principles of Product Development Flow)
• Since you’re already operating at or near full capacity, you can’t even deploy extra capacity to knock those queues down: it becomes basically impossible to ever get rid of them.
• The increased wait time in your ticket queue creates long feedback times, nullifying the benefit of agile techniques.

Efficiency isn’t the holy grail

Any queueing system operating at full capacity is gonna build up giant queues. That includes your team. What should you do about it?

Just by being aware that this relationship exists, you can gain a lot of intuition about team dynamics. What I’m taking away from it is this: There’s a tradeoff between how fast your team gets planned work done and how long it takes your team to get around to tasks. This changes the way I think about side projects, and makes me want to find the sweet spot. Let me know what you take away from it.

Kanban Highway: The Least Popular Mario Kart Course

I’ve been reading a really excellent book on product development called The Principles of Product Development Flow, by Donald G. Reinertsen. It’s a very appealing book to me, because it sort of lays down the theoretical and mathematical case for agile product development. And you know that theory is the tea, earl grey, hot to my Jean-Luc Picard.

But as much as I love this book, I just have to bring up this chart that’s in it:

This is the Hindenburg of charts. I can’t even, and it’s not for lack of trying to even. Being horrified by the awfulness of this chart is left as an exercise for the reader, but don’t hold me responsible if this chart gives you ebola.

But despite the utter contempt I feel for that chart, I think the point it’s trying to make is very interesting. So let’s talk about highways.

Highways!

Highways need to be able to get many many people into the city in the morning and out of the city in the evening. So when civil engineers design highways, one of their main concerns is throughput, measured in cars per hour.

Average throughput can be measured in a very straightforward way. First, you figure out the average speed, in miles per hour, of the cars on the highway. The cars are all traveling different speeds depending on what lane they’re in, how old they are, etc. But you don’t care about all that variation: you just need the average.

The other thing you need to calculate is the density of cars on the highway, measured in cars per mile. You take a given length of highway, and you count how many cars are on it, then you repeat. Ta-da! Average car density.

Then you do some math:

$\frac{cars}{hour} = \frac{cars}{mile} \cdot \frac{miles}{hour}$

Easy peasy. But let’s think about what that means. Here’s a super interesting graph of average car speed versus average car speed:

Stay with me. Here’s a graph of average car density versus average car speed:

This makes sense, right? Cars tend to pack together at low speed. That’s called a bumper-to-bumper traffic jam. And when they’re going fast, cars tend to spread out because they need more time to hit the brakes if there’s a problem.

So, going back to our equation, what shape do we get when we multiply a linear equation by another linear equation? That’s right: we get a parabola:

That right there is the throughput curve for the highway (which in the real world isn’t quite so symmetric, but has roughly the same properties). On the left hand side, throughput is low because traffic is stopped in a bumper-to-bumper jam. On the right hand side, throughput is low too: the cars that are on the highway are able to go very fast, but there aren’t enough of them to raise the throughput appreciably.

So already we can pick up a very important lesson: Faster movement does not equate to higher throughput. Up to a point, faster average car speed improves throughput. Then you get up toward the peak of the parabola and it starts having less and less effect. And then you get past the peak, and throughput actually goes down as you increase speed. Very interesting.

Congestion

Now, looking at that throughput curve, you might be tempted to run your highway at the very top, where the highest throughput is. If you can get cars traveling the right average speed, you can maximize throughput thereby minimizing rush hour duration. Sounds great, right?

Well, not so fast. Suppose you’re operating right at the peak, throughput coming out the wazoo. What happens if a couple more cars get on the highway? The traffic’s denser now, so cars have to slow down to accommodate that density. The average speed is therefore reduced, which means we’re now a bit left of our throughput peak. So throughput has been reduced, but cars are still arriving at the same rate, so that’s gonna increase density some more.

This is congestion collapse: a sharp, catastrophic drop in throughput that leads to a traffic jam. It can happen in any queueing system where there’s feedback between throughput and processing speed. It’s why traffic jams tend to start and end all at once rather than gradually appearing and disappearing.

The optimal place for a highway to be is just a little to the right of the throughput peak. This doesn’t hurt throughput much because the curve is pretty flat near the top, but it keeps us away from the dangerous positive feedback loop.

So what does all this have to do with product development workflow?

Kanban Boards Are Just Like Highways

On a kanban board, tickets move from left to right as we make progress on them. If we had a kanban board where tickets moved continuously rather than in discrete steps, we could measure the average speed of tickets on our board in inches per day (or, if we were using the metric system, centimeters per kilosecond):

And we could also measure the density of tickets just like we measured the density of cars, by dividing the board into inch-wide slices and counting the tickets per inch:

Let’s see how seriously we can abuse the analogy between this continuous kanban board and a highway:

• Tickets arrive in our queue at random intervals, just as cars pull onto a highway at random intervals.
• Tickets “travel” at random speeds (in inches/day) because we’re never quite sure how long a given task is going to take. This is just like how cars travel at random speeds (in miles per hour)
• Tickets travel more slowly when there are many tickets to do (because of context switching, interdependencies, blocked resources, etc.), just as cars travel more slowly when they’re packed more densely onto the highway.
• Tickets travel more quickly when there are fewer tickets to do, just as cars travel more quickly when the road ahead of them is open.

There are similarities enough that we can readily mine traffic engineering patterns for help dealing with ticket queues. We end up with a very familiar throughput curve for our kanban board:

And just like with highway traffic, we run the risk of congestion collapse if we stray too close to the left-hand side of this curve. Since kanban boards generally have a limit on the number of tickets in progress, however, our congestion won’t manifest as a board densely packed with tickets. Rather, it will take the form of very long queues of work waiting to start. This is just as bad: longer queues mean longer wait times for incoming work, and long queues don’t go away without a direct effort to smash them.

What we can learn from real-world queues

A kanban board is a queueing system like any other, and the laws of queueing theory are incredibly general in their applicability. So we can learn a lot about managing ticket throughput by looking at the ways in which other queueing systems have been optimized.

Measure your system’s attributes

First off: you need metrics. Use automation to measure and graph, at the very least,

• Number of tickets in queue (waiting to start work)
• Number of tickets in progress
• Number of tickets completed per day (or week)

Productivity metrics smell bad to a lot of people, and I think that’s because they’re so often used by incompetent managers as “proof” that employees could be pulling more weight. But these metrics can be invaluable if you understand the theory that they fit into. You can’t improve without measuring.

Control occupancy to sustain throughput

As we’ve seen, when there are too many tickets in the system, completion times suffer in a self-reinforcing way. If we’re to avoid that, we need to control the number of tickets not just in progress, but occupying the system as a whole. This includes queued tickets.

In some cities (Minneapolis and Los Angeles, for example), highway occupancy is controlled during rush hour by traffic lights on the on-ramp. The light flashes green to let a single car at a time onto the highway, and the frequency at which that happens can be tuned to the current density of traffic. It’s a safeguard against an abrupt increase in density shoving throughput over the peak into congestion collapse.

But how can we control the total occupancy of our ticketing system when tickets arrive at random?

Don’t let long queues linger

If you’re monitoring your queue length, you’ll be able to see when there’s a sharp spike in incoming tickets. When that happens, you need to address it immediately.

For every item in a queue, the average wait time for all work in the system goes up. Very long queues cause very long wait times. And long queues don’t go away by themselves: if tickets are arriving at random intervals, then a long queue is just as likely to grow as it is to shrink.

One way to address a long queue is to provision a bit more capacity as soon as you see the queue forming. Think about supermarkets. When checkout lines are getting a bit too long, the manager will staff one or two extra lanes. All it takes is enough capacity to get the queues back down to normal – it’s not necessary to eliminate them altogether – and then those employees can leave the register and go back to whatever they were doing before.

The other way to address a long queue is to slash requirements. When you see a long queue of tickets forming, spend some time going through it and asking questions like

• Can this ticket be assigned to a different team?
• Can this feature go into a later release?
• Are there any duplicates?
• Can we get increased efficiency by merging any of these tickets into one? (e.g. through automation or reduced context switching)

If you can shave down your queue by eliminating some unnecessary work, your system’s wait times will improve and the required capacity to mop up the queue will be lower.

Provide forecasts of expected wait time

At Disney World, they tell you how long the wait will be for each ride. Do you think they do this because it’s a fun little bit of data? Of course not. It helps them break the feedback loop of congestion.

When the wait for Space Mountain is 15 minutes, you don’t think twice. But when the wait is an hour, you might say to yourself “Eh, maybe I’ll go get lunch now and see if the line’s shorter later.” So these wait time forecasts are a very elegant way to cut down on incoming traffic when occupancy is high. Just like those traffic lights on highway on-ramps.

Why not use Little’s law to make your own forecasts of expected ticket wait time? If you’re tracking the occupancy of your system (including queued tickets) and the average processing rate (in tickets completed per day), it’s just:

$\text{Average Wait Time} = \frac{\text{Occupancy}}{\text{Average Processing Rate}}$

If you display this forecast in a public place, like the home page for your JIRA project, people will think twice when they’re about to submit a ticket. They might say to themselves “If it’s gonna take that many days, I might as well do this task myself” or “The information I’m asking for won’t be useful a week from now, so I guess there’s no point filing this ticket.”

Forecasts like this allow you to shed incoming ticket load when queues are high without having to tell stakeholders “no.”

Queues are everywhere

If you learn a little bit about queueing theory, you’ll see queues everywhere you look. It’s a great lens for solving problems and understanding the world. Try it out.