Sliders: A Guide to Caring About Everything
Above is a talk I gave to the Yammer Engineering team. I’m sharing it publicly because a friend of mine working at an early-stage startup was recently asking me about how you build a culture of results into a company so that this is self-sustaining as the company grows. It’s a great question, and it’s one we’ve been asking ourselves at Yammer from the beginning. The sliders metaphor covered in the talk above has helped us to keep all of our work in the context of what we’re really trying to do: Improve business communication and collaboration permanently. Everything we do should be a means to that end, and at any moment, our focus should be on the things that are most keeping us from that end.
A couple of notes: At one point I say subset when I mean superset. Also, the bottom of the video is cropped a bit and obscures some words on the slides, but I’ve tried to recreate the important images down in the summary below.
A sentiment you’ll hear around Yammer quite often is, “Care about everything all the time.” We want to have as little tolerance as possible for things that get in the way of us shipping a valuable product to users. Whether it’s improving our developer tools, our handoff from design to engineering, or even the signage that helps people find the restroom, we don’t want to allow anything to stay broken because it’s somebody else’s problem.
Still, we acknowledge that we all have a limited cognitive capacity and a finite amount of time to dedicate to fixing things. So we need a way to reason about how best to spend our efforts. Ideally, we want to be sure that all of our time is being spent on the things that will give us the best returns.
So the ‘sliders’ metaphor was born. I imagine the state of the world as having a huge number of dimensions on which we could make improvements — in fact, more dimensions than we have people. In my head, I’ve always pictured the part of a studio or stage mixing board with the sliders, where each slider represents the state of the world along some specific dimension.
An additional advantage of the sliders metaphor is that it gets away from binary thinking about the state of things. Rather than considering anything in a good or bad state, you’re forced to consider that everything exists somewhere on a continuum from good to bad. Engineering teams that use test-driven development need code coverage reports, which seems like something you either have or don’t. The reality is that you might have a tool that generates the reports, but it’s not being run automatically and alerting people. Or it doesn’t cover generated code. Or it can’t cover code linked in a certain way. Or it’s slow. Rather than arguing about what criteria need to be met to consider this acceptable, the sliders metaphor allows us to think of it as something that can always be improved as long as we want to focus on it.
So when do we stop focusing on code coverage reports? We have to think about diminishing returns. If we’re trying to tune a race car, we could spend all of our time and energy working on the exhaust system, but at some point, we’re not going to get any more than infinitesimal performance gains if we don’t stop working on the exhaust and shift our attention to the intake. The following is a visual I use when describing diminishing returns.
The point here is that the things in your organization in the worst state are usually the things you’ll get the biggest results by focusing on. I like to say, “Something is better than nothing, and perfect isn’t much better than very good.” As an engineer, this is a helpful reminder to myself that despite my aesthetic preference for perfection, chasing it often hurts my effectiveness. Back to the car analogy, I need a way to tell myself that the exhaust system isn’t perfect, but that’s okay, and I need to spend time somewhere else.
In fact, once we’re comfortable thinking of different things we could optimize in our organization as a slider, we can consider these sliders together on a mixing board. Some will be very low and some will be very high. This can give us a good way to reason about which to focus on.
Beyond giving us a starting point, imagining the world this way also gives us a way to reason about when to stop and focus somewhere else.
You’ll often hear the phrase “low-hanging fruit” to describe a similar process.
Getting your entire organization on board with this thinking can build a healthy culture of being incremental — of making things better, not perfect. It also breaks the habit of fixating and chasing perfection deep into diminishing returns while big gains go unrealized as everyone optimizes infinitely in a narrow area.
By repeating this process effectively, you’ll eventually get everything up near perfection, but you’ll ensure that in any given moment you were working on the thing that brought you the biggest gains. I realize this requires a lot of flexibility of focus, of talent and brainpower allocation, and of prioritization, and some of the talks we give about Yammer’s culture describe ways you can organize to optimize for that kind of flexibility.
But this kind of incremental approach can also serve as a conflict-resolution approach. Organizations are frequently slowed or even paralyzed as they obsess over agreeing to end states. They want to know that everyone is working toward the same goal, which is admirable. But we at Yammer strongly believe the pace of change in the world is increasing and that it’s getting difficult or impossible to reason about the ideal end state of anything. As we hold off execution and pay the time cost of agreement, the world is accelerating ahead of us.
Instead of spending “agreement cost” arguing about the gap between the different preferred end states, we find the incremental approach allows us to take a step that everyone agrees is closer to their individual and divergent end states than where we currently sit.
As we loop on iterations that improve things, we’ll find that everyone was wrong by some amount about what the perfect end state was. We’ll also find that as time passes, as we learn from our experiences, and as the end state approaches, our ideas of the end state converge. Sure, we’ll wander off course a couple of times, but the cost of course correction will be dramatically reduced if we can cut down the agreement cost required to take each step.
And, if we’re constantly looking around at how this thing we’re iterating on looks in comparison to other things in our organization, maybe we’ll find that we can stop iterating long before we’re at end states all of us thought would be necessary to deliver improvements to our overall effectiveness.
My weekly meeting with my direct reports is in our calendars as “State of the Sliders” as a way of reminding us where to focus. Every few meetings we stop talking about progress on our current initiatives and ask ourselves what we think our “lowest sliders” are.
It does require a cultural change to make your whole team comfortable with focusing on things in the worst state all the time. People like to polish and refine things they’re already proud of. They might feel good (in the short term) focusing on things that are already in a good state. But in the long term, the health and happiness of your organization hinges most on its effectiveness and success. Consider framing this as focusing on the things with the greatest opportunities for making you successful.