Enterprise Engineering Management 102

Bad Measurements of Performance

Enterprise Engineering Management 102

Check out a free preview of the full Enterprise Engineering Management 102 course

The "Bad Measurements of Performance" Lesson is part of the full, Enterprise Engineering Management 102 course featured in this preview video. Here's what you'd learn in this lesson:

Ryan discusses the pitfalls of using certain metrics to measure team performance, such as lines of code, number of commits, time-based metrics, a focus on quantity over quality, and associating card points to hours. Ryan emphasizes that these metrics do not provide meaningful insights into high performance and should not be used for measurement.


Transcript from the "Bad Measurements of Performance" Lesson

>> All right, so, I had to do on like performance of measuring performance in your team. Because there's a lot of bad metrics that you can look at for measuring performance, and I'm sure I'm not covering them all, so definitely audience, people on stream, shout them out. We've all seen the bad ones out there.

Lines of code is not a good measurement, remember that. No, it's not great. That's not efficient doing that or number of commits. All right, if that's it, I'm just gonna commit every little line of code I wrote. And that's a commit, like these are metrics that just don't really give you any insights into high performance and not really something that you should be measuring.

Time-based metrics, time spent in the office. I mean, why? I mean, as long as someone's getting their job done effectively, that's great. A lot of teams have or companies have unlimited vacation. It's I love that. You're not paying attention to the time that people take. People need vacation in order to perform well.

They need that time out of the office. And so, you don't need to measure how much time it is, granted, if they're never there and never delivering, okay, maybe that becomes a problem. But time-based ofbhow much time you spend on something may not be a great metric.
>> One pitfall I've seen with unlimited PTO that I think actually does warrant a time-based metric is making sure people are taking some time off.

I, for example, have an employee who asked me for two weeks off a couple of weeks ago, because he'd only taken three days off so far this year, and that's through my filters and told me that I should probably be checking in on that every quarter with folks and figuring out, who's maybe not taking the breaks that they need, so that I don't have an unexpected two-week off period.

Where I want him to take the time off, he clearly needs to take the time off, but I would rather know about it a little further in advance and also be able to plan for it a little more effectively.
>> Absolutely. I like the, like sometimes companies will do like the minimum, like minimum, is like that's a good one.

It's like it's forcing people to take that time, and actually that is one time I might be paying attention to like time office. This person hasn't taken time off. I'm not watching, but you kinda know when you're like, I feel like it hasn't been and you can ask them, when was the last time you took time off?

You thinking maybe you should at some point and nudging them, but yeah, you can kind of get ahead of that, I love that call out. That is a really good one. Quantity over quality. It's like just cuz someone does a lot of something doesn't really necessarily mean it's great.

Maybe in some cases, but I think you can kinda run down weird areas of trying to measure that for performance. So these are some of the items I had. I'm curious, what are some of the other horror stories you've all seen, or things that managers probably shouldn't be measuring for performance?

Jem, you're smiling.
>> We have a section coming up. I don't wanna steal my own thunder.
>> Okay, that's fair. That is fair. Anything else that people have seen companies maybe pay attention to or track? Yes.
>> Associating card points to hours.
>> That's a good one too.

So if anyone's done like the card points, and figuring out for the tracking to hours. Also, if you're given like 30 hours to do something, oftentimes people fill that time with 30 hours when it might have taken 15 and so thinking about that as tracking to hours, that's a good call out.

Dustin, I saw, yep. Yeah for you in the chat, PRs, your tickets completed, meetings attended, how many stories you've completed in a sprint or how many spill overs you've had? Yes, all good ones. Man, I must be performing really well. I'm in a lot of meetings as an engineering manager.

Yes, I'm doing well. Yes, all those things. It's like that's not something to measure. You might have to be in a lot of meetings. You might not need to be in meetings, it all depends. And that should definitely not be a measurement, number of pull requests you have, it doesn't really tell me much.

Learn Straight from the Experts Who Shape the Modern Web

  • In-depth Courses
  • Industry Leading Experts
  • Learning Paths
  • Live Interactive Workshops
Get Unlimited Access Now