Peer based performance feedback


#1

Over the past few months, the PeopleOps team have been working on designing a version of “performance management” that fits with the values & organisational style of Status. The goals would be:

  1. Be open & transparent about what we’re working on, and have accomplished.
  2. Provide a system where anyone can provide feedback on our work.
  3. Help us recognise our key contributors, with an understanding of how and why they’re awesome.

To traditional organisations, performance management is where managers assign employee ratings based on their opinion of performance over a historical time period, sometimes including (but often not including) peer feedback. The impact, feedback and outcomes are typically private between the employee and manager.

That won’t work for us. We looked at a lot of tools/software to help us, but concluded that there’s nothing really out there that allows an organisation like ours to understand/manage individual performance as a community.

Next week the People Ops group will be testing a new methodology that we think can achieve our aims and fit within our values - a transparent method of showcasing our accomplishments that invites the community to review and promote individual accomplishments. No one has really done this before, and we know version 1.0 won’t be perfect, but we’re really keen to try and solve this.

We’ll share more specifics next week after our test, but if you have ideas on how we could implement a method of performance management that would fit into our organisation we’d love to hear them here. We’re looking to refine a method over the next 3 weeks, and then launch a 1.0 version shortly after.


#2

One thing other than peer feedback that I think is worth considering is concrete and specific success criteria.

For example perhaps you cryptographically commit or “promise” to do such and such thing by such and such time with such and such resources and get graded at the end based on how closely you kept your promises.

This may encourage people to be more thoughtful and honest in their estimates and promises, as well as helping to develop a reputation for being reliable.


#3

Been thinking in similar terms and definitely agree. Here’s some rough code sketch from a Sunday afternoon a few months ago: https://github.com/oskarth/sitg/blob/master/sitg_v0.sol - I should get back to it and make it actually runnable.

Another idea in this vein is to provide funding for a challenge, and someone can accept it provided they are willing to risk X%, say 10%, or the reward. This leverages private knowledge and is a form of Kelly criterion bet. Can do it socially with trust assumptions at first among core core contributors before making it work in a trust minimized way (evaluators/oracles).


#4

Cool ideas @noman and @oskarth - just out of curiosity, would building something like this take a lot of work?


#5

@ceri To do it fully will be a long process, but there’s a lot we can do do culturally already. You can see the start of this work here:

In other words, we can do it right now :slight_smile: