I have a complicated relationship with OKRs and that doesn’t seem to be changing any time soon. It is complicated primarily because I don’t have any issue with them in theory – I kind of like them in fact – it is just that I haven’t encountered many environments where the assumptions that OKRs make for success actually exist. Randy wrote about this recently and I enjoyed this post on LinkedIn from Ryan. You just need a lot of [strategic] scaffolding and available data to hang OKRs on for them to really hit the mark – and that is rare at least in public service.

Still they are not going away – and I’m totally on board with teams having consistent Objectives so I persevere 🙂

When I was working with GOVUK one of the Lead Product Managers, Jonathan Nicholls, shared a really great deck on what good OKRs look like. Especially in that GDS world. It really helped me get a handle on how I could get over my prejudices and help my teams with their commitments. 

The talk I gave to the team at MHCLG was my interpretation of Jonathan’s deck with my own particular commentary and annotations…and as such I don’t really feel like I can share it as is as the majority of the IP is not mine.

So I’m just going to share some of my more coherent commentary about OKRs from the session – which might not hang together to be honest…and anyway with all my interference I’m not even sure if they are recognisable as OKRs by the end anyway!

POPULAR CONSPIRACY THEORY:

“OKRs were actually a psyop from Google to slow down potential early stage competitors.”

— WHY HAVE OKRs? —

OKRs are mainly useful for clarity – they provide a consistent set of goals. They help communicate intention upwards, sideways and within the team and they help teams have hard conversations about what is important.

— OBJECTIVES —

For a start I wish the ‘O’ was Outcome not Objective but that is neither here nor there I guess.

Make them inspiring and don’t get hung up on the Quarter timeline – they can last longer – much longer if need be.

I appreciate more narrative than some product people – not just a pithy title – more like NCTs. In fact I think I probably end up getting teams to do NCTs but call them OKRs.

Make sure they do more than pay lip service to your strategy/vision/mission – whatever you actually have. Objectives without some kind of anchor to the big picture struggle to make an impact – or survive the pressure of random asks from stakeholders.

From Jonathan’s deck;

— GOOD OBJECTIVES ARE —

✓ Related to the overall strategy [..but they aren’t a substitute for strategy]

✓ About the end state

✓ Inspiring (and maybe even exciting!)

✓ Concise and clearly worded […but it isn’t Mad Men – you aren’t crafting the perfect ad – use the words you need.]

— KEY RESULTs —

Now officially KRs should be Quantitative and measured over the quarter…but I tend to be comfortable with Qualitative options as well. I’m just looking for something that evidences progress towards meeting the Objectives and sometimes this is better reflected in a different way. 

Part of the challenge is that a lot of the time (most of the time?) teams just don’t have the kind of data available that allows the kind of measurement needed and too often the team gets distracted in finding a way to provide the metrics ahead of delivering the actual Objective. The measuring should not be more effort than meeting the blooming Objective. 

Make the KRs hard but not impossible. They should elicit a bit of wincing when you read them but not any sleepless nights.

From Jonathan’s deck;

— GOOD KEY RESULTS ARE: —

✓ Derived by the team

✓ Achievable but not easy

✓ About outcomes not outputs

✓ Measurable

  […and measurable by the team without relying on others or significant extra work]

✓ Concise and clearly worded

✓ Regularly reviewed – It’s OK if they change

=== PRIORITISATION ===

Your OKRs are your priorities so the tradeoffs are limited to delivery of those 3 to 4 ‘Os’ and you need some kind of method to help you decide what those priorities are. Honestly for product people this is more than half the battle and we all get ourselves twisted up around it.

I’ve been known to refer to the process of prioritisation at scale as like the Game of Thrones;

When you play the Game of Thrones, you win or you die. There is no middle ground.

…but honestly it does not have to be that way.

— 1ST RULE OF PRIORITISATION CLUB —

Don’t use MoSCoW

— 2ND RULE OF PRIORITISATION CLUB —

Don’t use MoSCoW

…but anything else is probably fine as long as you are consistent and transparent.

Commitment to these principles is more important than any particular approach;

– Be transparent

– Be consistent

– Be open

– Be flexible

– No surprises

There are however endless frameworks to choose from and they all have their strengths and weakness – 

https://foldingburritos.com/blog/product-prioritization-techniques/

I like a spin on RICE where I add values for Policy importance and something for Urgency (so UP-RICE?)…but increasingly I like the look of Cost of Delay approaches like John Cutler recommends. He even has a little web app to help with the process.