Kristian Glass - Do I Smell Burning?

Mostly technical things

Lightweight OKRs

If OKRs - “Objectives and Key Results” - are completely new to you, there are probably better introductions. But if you’re looking for a lightweight OKR system that has worked well for me, then here goes.

Many of the problems I’ve seen and heard of with OKRs seem to come from well-intentioned people taking Google’s implementation of OKRs (which includes scoring and assessment and measurement and other complexities and nuances) and misapplying it, copying some parts but not other essential pieces, and just generally having a square-peg-round-hole situation.

I’m sure the Google system works well for Google. Here’s what has worked well for me.

First up: what am I trying to achieve with OKRs? I want a way of aligning people and communicating goals and direction. I don’t want to necessarily specify implementation, inadvertently micromanage, or other problematic things. I want to assess how well we’re doing, but I’m not trying to build a measurement framework.

So, define some Objectives - high-level directional goals and intents. Maybe they’re tangible - a revenue target or a concrete achievement. Maybe they’re more nebulous - “make people happier” is a solid (and worthwhile!) objective.

For each Objective, define some Key Results - concrete tangible achievements that indicate progression towards the Objective. They should specify a result not a task - don’t fall into the trap of specifying implementation. A Key Result for an Objective about revenue might be the launch of a new product you expect to sell well. A Key Result for an Objective of making people happier might be to score better on an annual mood survey.

Once you have these, crack on - that’s it!

The focus here is on Objectives - Key Results are potential but not necessary contributors. They give you something tangible to aim for, but they’re a milestone on a planned journey, not the end destination. If your Objective is to exceed last year’s profits, and a Key Result is something resulting in a massive revenue increase, that shouldn’t stop you from taking advantage of another opportunity for profit that might arise instead.

And that’s it. No “scoring functions”, no “difficulty calibration”. This isn’t a framework for goal-setting and measurement across an organisation. It’s a lightweight tool for alignment and direction.

Encourage people to cascade OKRs down into smaller groups. If you have organisation-wide OKRs then it’s great if divisions can produce OKRs that align within those, and teams producing OKRs that align within those, and individuals producing OKRs that align within those. But it’s not a strict subset. The “parent” OKRs won’t capture everything and shouldn’t try. They should be about what matters at that level.

OKR all the things - big to small. Prioritisation matters, and isn’t touched upon here, but there’s value in explicitly encoding targets of all sizes in this form, and later reflecting on success or where things didn’t work out.

Review periodically - both Objectives and Key Results. Things change and there’s no point keeping irrelevant goals around, and there’s no shame in changing priorities in light of new information and circumstances.

Reflect periodically. How are things going? Are you progressing as you’d like? Can you do more? Should you do less?

Lightweight OKRs - a relatively simple tool for alignment and direction-setting.

With thanks to Stephanie Richardson-Dreyer and Ceri Davis for their insights and review.