Setting Goals and Evaluating Performance

Problem

“Performance” in the world of engineering can often be a dirty word. People hear it and question whether or not evaluating engineers in this way is the best thing to do. It is. Setting expectations, changing processes, and improving things for everybody in the company is what every leader should be trying to do.

Everybody wants to set a goal. It ends up being a bit like an oil change. When everything is going okay and the car is running smoothly, you don’t care about whether or not your oil needs to be changed. You’re not talking about it or thinking about it, you’re just driving your car. You really only worry about your oil when something bad happens because you have not changed it. Changing your oil, mentally, now comes with all of this baggage.

Now, suddenly, with performance on the agenda, people panic. They feel as though they need to fix everything right away. This is unfortunate, because it’s much better to put in that initial footwork early on, proactively, before performance becomes a problem at all. Being explicit and transparent with your team about these things keeps the gears turning.

Actions taken

At my first board meeting, I actually ended up having a really difficult conversation with my director. I was told that I needed to increase story points for the team, because that’s what my leadership thought would solve the problems that we were having. I told them that this was not the case at all.

What we needed to improve was the quality of what we were putting out; this would increase our velocity naturally. Sure, increasing the volume of the work being done meant that we were doing more. Relying only on vanity metrics like this, however, doesn’t actually drive real progress in the areas that are really worth caring about. What we were shipping was not of the highest possible quality.

That’s how I kind of found myself in my previous position. My manager, the VP, had left the company, and I got this big, battlefield promotion and ended up taking over all of engineering. My proposal for the six months to follow was to focus on quality above all else. Here are the goals that I’m setting. Here is the standard to be met.

We started seeing some improvement. We had a strong foundation of quality to stand on to start the year off with and we all felt very good about what we were doing together. Engineering, product, and design were all on the same page. These tenets were built into the fabric of our organization, and everybody was finally able to understand what our real mission was. We put the work in early, up-front. This made the rest of it easy.

After finding our footing, the next step was becoming more predictable and consistent in our execution. That’s another part that nobody really talks about. Performance is not just good or bad; what you really want is a pattern that is predictable. That’s what makes a team really strong. Once the bar has been set and you are reaching that standard consistently, you will be in a better position to take a deeper dive in order to optimize and make your team even more effective.

Lessons learned

• OKRs and KPIs are a great way to facilitate this consistent quality. For us, in engineering, a lot of our OKRs involved the releases that we were working on. We can commit to, say, meeting no fewer than 80% of those obligations, collectively, as an engineering team. It’s just a target to aim for; you don’t have to hit the bullseye every single time. If you do, then your bullseye may be too large.
• Engineering performance is really about quality and predictability. That’s the foundation, and to build it, you need to be transparent with your team. You cannot withhold those metrics from them and expect them to succeed. We have a monthly engineering summary that everybody in the company has access to. It helps them feel accountable.
• You need to celebrate all of your wins and build that sense of camaraderie between team members. You also need to create a safe environment for your employees to reconcile with failure and to overcome. We own this holistically, as a group.