Incentives
So a few weeks ago I started this blog full of vim and vigor, ready to commit to building something and putting down to digital text all the lessons I’ve learned and, well…
Then I got a cold. Nothing too bad, but it sapped away some of my energy.
Then the election happened and civil society started to crumble a bit. I’ve got a lot of friends in the federal government (or supported monetarily in their careers by said government) and I watched them fear for their livelihoods. You, dear reader, have had similar experiences I’m sure. Government spending as a whole is quite a lot of our GDP.
(This is to say nothing of the fear that is running through the hearts of every LGBTQ+ American, or their allies, or anyone abroad who benefited from a more “liberal” government in power, which I use loosely.)
That sapped some more energy.
Then my career search went back into high speed, because if you can strike while the iron is hot? You should. I did. Successfully, in fact. But that took even more energy.
And at the end of the day, at the end of many days, I’d look at what I wanted to do and I would promise to get to it tomorrow. And a day later, I’d promise for sure the next day. And more and more until, sure enough, it was early March and a whole month had passed.
Dear reader, I’m sure you can also sympathize.
So why couldn’t I do the things that I wanted to do, like write in this blog? The answer is, it was hard. Which is easier to say and even easier to respond to. Maybe it’s not hard and I’m just not prioritizing it. Or maybe it’s not hard and I’m lazy. Maybe it is hard and instead of writing, I should spend some time finding ways to make it easier?
For two years of my life in my early and mid-twenties, as a somewhat directionless but frustratingly productive person, I found myself having decided that I was going to go to grad school. My reasoning? I liked college. I liked the idea of teaching and researching. I had lived in Japan for a couple of years and realized that translating was not the giving me a good answer to the question “What do I do? Why do I get out of bed in the morning?”
I aimed for international relations programs and poli-sci programs thinking hey, I’m reasonably good at Japanese, I like the political and historical and cultural analysis, let’s do something with that. What I ended up in was a two-year masters degree in Public Policy.
Did I end up working in government or a think tank or in academia? Clearly not.
What the experience gave me was a set of skills and knowledge to look at things that we as engineers do all the time — evaluate complex, interrelated systems and make changes that we hope do better things. And I’ve leaned on those skills time and time again, especially since the real benefit of that knowledge was to see how it related to people, to the complex and interrelated system that is any modern workplace.
(As an aside, do I think it was worth the price tag? Jury is still out, honestly. Do I use the skills? Sure. Were they a necessary condition to get into software? No, but it helped immensely.)
I took one thing above all else from policy school, which I’ve used over and over and over and over (and have told anyone who has genuinely asked over and over and over as well.) It is, to wit:
People follow incentives. They will readily do what it is easy and they will avoid doing what is hard. People can do what is hard but only with significant effort to do so. People can choose not to do what is easy, but again, generally only with significant effort.
This applies so often in the workplace that once you start thinking in terms of incentives, you can’t help but see it everywhere:
When you’re looking for information on a task, do you dive through the documentation to find the company standard or do you ask someone who’s done it before what they did? You probably ask someone else. On the off-chance that you do find the information you’re looking for in your Confluence or your Notion, do you make sure it’s correct, update it if it’s not, and then popularize the use of it within your team? You probably just get your task done and assume you’ll come back to it later. No sweat, we all do it.
Imagine you’ve gone to HR with a minor complaint or to your boss with a minor issue — to fix the issue involves a reasonable degree of effort and the number of people who would benefit would be small (maybe even you). Do they jump in to help or do they do what they can to not have to do the hard thing (maybe that’s telling you it’ll happen next quarter or promising a follow-up that never materializes?) Can we really blame them? Folks are busy. Solving things is hard.
A junior colleague reaches out with a pull request. They’ve been on this task for weeks. The project has run late. Everyone just wants to see it out. There’s a small but meaningful error in their code — flagging it for fixing now means the whole thing is delayed even further. Do you bite the bullet and get through it or do you say that it can be a fast-follow, we’ll do it right after we release? Look, we gotta get this feature out, don’t we?
I want to be clear here that one could characterize everything I’ve said as laziness. I’m not suggesting that people in these situations are lazy — I’m claiming that they are (correctly) following their incentives. They are doing what it is easy and not doing what is hard.
(In fact, this slips all the way into product design. Anti-patterns in design are there to make certain things harder to do so that you do them less — try quitting a gym without schlepping there in person and filling out a paper form! But, if you want to (say) buy tokens to progress in a free-to-play game instead of grinding or watching ads? You’ll find that flow as easy as possible. It’s why payment flows are so heavily optimized to make going from “I want to buy this” to “I’ve bought it” as frictionless as possible (looking at you ApplePay and specifically looking at how you can press a button and pay). It’s why at most SaaS company, there really are two five alarm fires — the app is down, or the payments are down).
So how do we get the behavior we want? We can ask that people put in the (sometimes Herculean) effort to do what is hard. Some people do that. They usually get promoted to Staff or Principal and do three times the amount of work of other Staff or Principal engineers in the world of individuals contributors. Or they become the heavily overburdened managers who have a caseload twice as big as other managers and when they burn out and depart, everyone says “wow, I can’t believe they lasted as long as they did.” But while some people will do that, most won’t and crucially, you generally can’t make that happen, not without some Herculean effort of your own.
Instead, you can change the incentives.
Do you want to see folks using the documentation or updating it? Maybe it’s time for bug bounties for internal documentation or surfacing common tasks faster (FAQs go a long way — your slack channels probably could use one!)
Do you as a leader want to see folks take ownership of minor process improvements or handle issues without HR’s involvement? Make it easy to do — retrospectives (for all their failures) make it easy to bring these things up. 1:1s, done consistently and in a way that empowers folks to be honest and vulnerable with you, can surface these issues ahead of time or cue you into the need for more training or conversations.
Do you as a mentor or senior engineer want to see your junior engineers turning in better work? Do you not want to see things vanish into the backlog (I.e. where fast follows go to die?) Make it easier to do that work. Scope the tasks that junior engineers do well (resolve that ambiguity before they have to). Find ways to avoid fast follows at all (pair programming is a great way to do this). Consider linting and automated testing and whatever you can do to reduce the cognitive load on the things that don’t matter to free up a junior engineer’s ability to focus on what does — solving the ticket at hand, hopefully with a good amount of thought behind it.
You might notice that a common unspoken theme here is that to change incentives, someone has to do that and that, itself, might be hard (and thus might never happen). Folks in the neuro-divergent community often refer to spoons — the limited resource of being able to give a fuck about something. Giving a fuck about something (and actually doing it) is, in effect, doing the hard thing instead of the easy thing.
Writing this blog, for what it’s worth, is one of the hard things. I encourage you to, the next time you want to do a thing and find yourself consistently not doing that thing, consider the incentives you have. What can you change to make the hard thing easier, or what hard things are you doing now that you can stop to give yourself the space to tackle what you want to do?