We Are Group

A huge chunk of my brain has been occupied by annual reviews over the last few weeks, and as a result, I've had a lot of ambient thoughts that are tangentially related to performance evaluation.

And what do I do when I have thoughts?

I scream them into the infinite and unforgiving void that is the internet.

I Am Groot

In every place that I've worked, individual performance evaluation is the norm.

Even in places where there wasn't a well-structured evaluation framework, as a manager, you were still expected to look at an individual's accomplishments and contributions, evaluate them against a set of expectations and either rate them, provide feedback or both.

The core motivation of the business is to ensure there is a good return on investment from the incredibly expensive cost of employing people. It wants to avoid low-performers and encourage high-performers, and evaluation frameworks and processes exist to encourage that outcome.

That doesn't mean there isn't anything in it for the people being evaluated though, because most people want to get better at their job. Being better means being more effective and being more effective means more opportunities, or money, or recognition or whatever other form of growth or progression that people are living for.

So really, individual performance evaluation is kind of a win-win situation. Businesses get better people and people get feedback on how to be better.

Conceptually individual performance evaluation makes sense.

We're used to looking at a group of people and breaking it down into its constituent parts, acknowledging that Bob brings skill X and behaviour Y to the situation while Mary brings skill Z and behaviour K.

The ideal team is greater than the sum of its parts, but it's still made up of parts and those parts are individuals.

So, the normal individual performance evaluation process is to look at the things that can be attributed directly to the person in question. What they did, what they drove, how they helped, who recognised them, etc.

Then to somehow contrast and compare that information to a standardised baseline, or to similar people around them.

Easy, right?

I Am Groot?

The problem with measuring and evaluating people as individuals is that it encourages selfish behaviours, because it's much easier to be able to say, "I did X" than it is to say, "I helped Mary do X".

In such a performance evaluation framework, the first statement is strong. It implies a level of agency and responsibility.

The second statement is weak. It is difficult to measure and difficult to prove. How much did you help Mary? Would she have done it just as well on her own? Perhaps she would have done it better without you?

Now, that's a pretty cynical view of things, I know.

Not everyone defaults to self-interest. Some people just focus on doing a good job and making a difference and don't give much thought to how they are going to prove their awesomeness when it comes to review time.

Also, in a well-designed performance evaluation framework, there are often behavioural guards to prevent the worst of this behaviour, like measuring people against the Atlassian values for example. You can't be a high performer and be a jerk about it, because that means you're not actually a high performer.

Still, when measuring individuals, the system encourages self-interest at a fundamental level, thus you see more of it.

What if there was another way?

We Are Groot!

If you must have ratings, why not rate the entire group exactly the same?

If the group is meeting expectations, then everyone gets a Meets Expectations (ME) rating. If the group is struggling, everyone gets a Does Not Meet Expectations (DN) rating. If the group is smashing it, everyone gets a Greatly Exceeds Expectations (GE) rating. And so on and so forth.

In this context, a group would be a team of engineers and other supporting people + the manager responsible for them.

It might not seem fair, but bear with me for a moment.

The most important thing to the business is results. The delivery of value, whatever that might be, and the creation of meaningful outcomes.

Teams produce better than individuals. A group of people, appropriately bound together as a team, can make amazing things happen in a more robust and reliable way than a single individual.

So why not lean into that.

Why not evaluate the team itself as an atomic entity, and let the team sort out the rest themselves.

If there are low performers for example, it's up to the team how they deal with them. Maybe they invest in the low performer, educating them and raising them up to the required level. Maybe they excise the low performer, removing them from the group for the greater good.

Team dynamics and harmony become more important in this world, because the best teams are ones where people work together to achieve their goals.

It also somewhat enshrines the team itself as an important element at the organisational level, discouraging chopping and changing people around in that weird optimisation game that managers sometimes play.

We Are Groot?

Honestly, I can see so many different issues with a team-based performance evaluation system.

For one, teams are already evaluated on their performance, so it's not like it adds any new information.

For another, I can imagine that high performers will rail against being capped at the performance of their team.

For a third, you'd still need an objective way to either grow or excise low performers within a team, or it's just going to turn into Lord of the Flies.

I could keep going, but I won't.

I still think the idea was worth thinking about though. Sometimes crazy ideas have a kernel of awesomeness in them. Maybe I could start experimenting by giving the team a rating and surfacing that to the organisation, while still rating everyone individually as well.

It might start a new trend.