Thread by @bcantrill

So, my thoughts on engineering performance management have always been a bit idiosyncratic, but Matt's tweets today have me reflecting, so... storytime. 1/

https://twitter.com/mattklein123/status/1216486019744202752

An an engineer in a large corporate environment, I had found that performance management never actually improved my own performance. (!) 2/

The things that resulted in my own high performance were the basics: being motivated by the problem; being drawn to the mission; being a part of an incredible team. 3/

Daniel Pink summarizes these motivators concisely: people are motivated by autonomy, mastery, and purpose. 4/

The role of management is in constructing that environment, not micromanaging it. If engineering performance is suffering, it's (likely) a management problem: wrong problem, wrong mission, or wrong team -- or all three. 5/

I elaborated on my thoughts of what was wrong with everything in this talk in 2013 at Surge (RIP): 6/

There are some things I would change in that talk, but I broadly stand by it. In terms of how my thinking has evolved... 7/

I worked with a very open minded director of HR at Joyent; she at first thought I was bonkers, but she was willing to work with me and saw that it worked: the path to highest performance was in stoking intrinsic motivation. 8/

After Samsung acquired Joyent, to their credit, they didn't try to force traditional performance management upon us -- but they also wanted something a little more... formal. 9/

We decided to "formalize" our process by having each engineer engage in a bi-annual reflection, asking them to write down their answers to several questions. 10/

1. What are you most proud of in the last six months?
2. What did you learn?
3. Where did you struggle?
4. What are you anxious about in the coming six months?
5. What are you excited about in the coming six months? 11/

The answers were incredible -- deeply reflective and revealing. It allowed us to sharpen intrinsic motivation, especially with some very surprising answers about what engineers are proud about. 12/

In particular, engineers are not boastful by nature; ask them what they are proud of, and you are likely to learn about a detail of a system that you did not sufficiently appreciate! 13/

My belief was that you could hand the reflections to someone who knew absolutely nothing about the organization or even the technology, and they could (easily) spot the high performers vs. people that we're struggling. 14/

The reflections were a huge win -- and they had nothing to do with OKRs or MBOs or formalized ranking or any other traditional performance management garbage. 15/

So let's not kid ourselves: traditional performance management has nothing to do with improving performance -- and everything to do with middle management claiming responsibility for that performance. 16/

If you want a high performing team, get back to the basics; your job as an engineering leader is to effect Pink's triad in your team: autonomy, mastery, purpose. 17/

The only real performance management to be done is for engineering leaders to ask themselves what they themselves are doing to give their team autonomy, encourage their team's mastery, inspire their team's purpose. 18/

Also, have I mentioned that we're hiring at Oxide? ;) oxide.computer/careers/ 19/19