Setting OKRs can be tough.
At Redgate we’ve been using Objectives & Key Results (OKRs) for around 2 years now. I wouldn’t say we’ve perfected it by any stretch but we’re now at the point where a number of our teams are exploring many different ways of agreeing the outcomes for their work beyond the approaches we have suggested — this is great, it’s a perfect demonstration of continuous improvement where a team moves beyond the agreed process and experiments with new ways of trying things.
But in the meantime, I’ve also been experimenting with techniques to set effective OKRs that don’t result in teams spending days and weeks in meetings trying to agree something that doesn’t feel like an output rather than an outcome.
When we first started using OKRs here the process was lengthy and could be painful. We started by agreeing to set OKRs every quarter. The problem was that because we found it so difficult to agree and reach consensus, many teams would find themselves going weeks into a quarter trying to agree their OKRs, which meant by the time they were signed and the team started measuring them it was time to set the next quarters OKRs.
So about 18 months ago I tried a different approach. Me and my teams took over a meeting room for a day and agreed we would attempt to get them set across one day rather than two to three weeks. It worked, but the day was draining. And for anyone who has seen Linda Risings brilliant talk on ‘Thinking fast and slow’ you’ll realize that spending hours trying to talk and agree something that can be pretty complex, our minds tend to switch off after about 45 minutes. It meant we set ok OKRs, but not brilliant ones that would really drive the outcomes we wanted.
Gradually we reduced the time in these sessions from a day down to half a day, then down to 3 hours. The problem however was that the OKRs still felt rather forced or shaped around the work we were already doing rather than shaping the work we should be doing.
As a result, I tried something different for our latest OKR setting this last week. Below is a template guide of what I did which I want to share with you. I don’t guarantee that this will lead to amazing OKRs being agreed by your teams, but it may provide some useful techniques and ideas for you to try and evolve with your own teams which might at least make it easier to get through this process.
I’d love to hear your feedback on these and see if they did help.
Notes before you begin
I scheduled two 60-minute sessions for this process. One to agree the objective part and the second session to agree the key results for the objectives. Anything longer than this and you’ll start to see people switching off or engagement dropping. You may be able to push to 90 minutes but if you do, I’d recommend a break in here as well.
Session 1 — Setting the objective(s)
2. Recap of your last OKRs
3. The objective setting
a. What makes a good objective?
b. Setting these objective(s)
Session 2 — Setting the key results
1. Recap of objective(s) agreed in last session
2. The key results setting
a. What makes a good key result?
b. Setting the key results
Session 1 template — Setting the objective(s)
1. Icebreaker (5 minutes)
The icebreaker I used for this first session is something called ‘What will you bring?’ It’s a pretty well-known icebreaker technique but useful for asking each of the team to note down on a card what they will add to this session, e.g. I’ll bring challenge, I’ll bring deep technical expertise etc.
Give attendees one minute to write down on a card and then share with the group. Remind the attendees that you are asking each of them to hold each other to these statements.
Why use this technique?
Studies show that people who speak right at the start of a meeting are far more likely to contribute throughout the session than those who say nothing at the start. By asking people to share what they will bring they are both voicing something and asking one another to hold them accountable for what they said they would bring. It’s a useful working agreement style icebreaker to kick-off the session.
2. Recap of last OKRs (5 minutes)
Review the teams last OKRs, use a round-table discussion and spend 5 minutes with the group discussing what worked well, what could have gone better etc. Note down these observations on a flip-chart or whiteboard.
Why use this technique?
Being able to discuss openly and honestly the last OKRs is helpful to promoting safety in the room and useful as a first demonstration of people living up to the icebreaker.
It allows some closure of the previous OKRs and some useful reminders for the next OKRs as to what worked and what worked less well.
3. The objective setting Part 1 — What makes a good objective? (10 minutes)
Ask the group to list the things that make a good objective. List these on the flip-chart/whiteboard, these will become your checklist for the session. As a reminder (And to help prompt the team) good objectives should: -
· Talk about the change we want to see in the world
· Focus on outcomes not outputs
· Must be definite
· Must be measurable
You can also pose some questions in this part of the session, should our objectives be time bound for example?
Why use this technique?
Asking the team to agree the indicators that will tell them that they have set a good objective is important. By producing a checklist, it makes the next part of this session easier to qualify whether an objective is a ‘good’ objective.
4. The objective setting — Part 2 — Agreeing the objective(s) (40 minutes)
Give the attendees post-its and pens and ask them to list potential objectives that the team could work on. Do this as individuals to start with, giving them 3 minutes. They don’t need to worry about wording these objectives perfectly, just a sense of what an objective could be.
Ask the attendees to form into groups of two and share their proposed objectives, removing any duplicates, creating additional ones. Give them 4 minutes to share, remove and agree between each pair.
Ask the attendees to form into groups of four and repeat the above process again, removing duplicates, forming new objectives and agreeing between the group of four. Give them 5 minutes to do this.
Ask the attendees to all come together and repeat the above process again. Ask them to order the final list into a prioritized list of potential objectives. Remove duplicates etc.
Give the group 8 minutes to complete this stage.
Ask the group to place the prioritized list on the flip-chart/whiteboard. Select the top three objectives and ask the team to compare them against the ‘What makes a good objective?’ checklist they agreed at part 3 of this meeting.
For any objectives that do not meet these requirements, drop them to the bottom of the prioritized list. The chances are that some of the proposals will not meet the checklist, the team can either try to reword them at this point OR if there are other compelling suggestions, remove these ones entirely.
Once the team have reviewed all proposals, select the top 2–3 — these will become the teams proposed objectives.
Discuss as a group how to word these top 2–3 proposed objectives into the teams formal objectives (if required).
Conduct a fist of five consensus vote for each of the proposed 2–3 objectives to check the team are comfortable with what has been agreed.
Why use this technique?
1, 2, 4, All is a useful technique for allowing individuals to consider options before bringing them together in increasingly larger groups to reach consensus and check alignment.
Fist of five is a fantastic consensus checking tool for where you want to check a group is in agreement and alignment.
Both of these techniques being used together allows a group to explore a problem space on an individual level before coming together to check their thinking.
Session 2 Template — setting the Key Results
1. Recap of objective(s) agreed in previous session (10 minutes)
Give the group time to discuss via a round-table the objective(s) they agreed in the last session. Pose the following questions: -
(a) Has anything changed since our last session that influences the objective(s) we agreed as a group?
(b) Have we learned anything further that would fundamentally change the objective(s) we agreed?
2. Setting the key results (50 minutes)
First ask the group to form their checklist for what makes a good key result. As a reminder some of the things that should be considered in making a good key result are: -
Posing the question of ‘how are you going to get that done?’
· Measurable by a specific metric
· Can be an activity
· Shouldn’t track work
· Related to the objective
· Can we get a baseline for it?
· Not a binary outcome
The list that the group agrees will become the checklist to measure their key results against in the next activity.
For the next part I used a modified version of Min Specs from Liberating Structures.
Ask the group as individuals to list every possible way to measure the objective(s) they agreed in the last session. This will become the ‘max specs list’ Give the group 5 minutes to do this activity.
Ask the group to come up individually and place all of their suggestions under the max list column on your flip-chart/whiteboard. Remove any duplicates/reword similar suggestions.
Now as a group ask the attendees to gather round the flipchart/whiteboard and discuss the list. Their objective is to select a minimum of 3 of the suggestions and move them to their ‘Min specs column’ by answering the question of
‘What is the absolute minimum set of Key Results we would have to achieve in order to complete this objective?’
Your flipchart/whiteboard should look something like the below: -
As the facilitator, ask the group to start by removing all of the suggestions that obviously do not meet the Key Results checklist they agreed in the first part of this session to try and reduce the list down.
To ensure that proposed key results are ‘good enough’ I recommend using the 5 Why’s technique with the group. Ask them to challenge one another on Why a proposal key result should move to the minimum specs column, if they can continue to explain why it should move then the chances are that it’s a viable key result.
Every couple of minutes ask the team to remove 2/3/4 of the suggestions from their list. This keeps the discussion focused on seeking the absolute minimum list.
Continue asking the team to remove their bottom list of suggestions every few minutes. The aim is to help them get down to a few suggestions which can be tested via 5 Whys in order to move to the minimum spec’s column.
By the end of this part of the session the group will have come up with their final 3 suggested Key Results.
As a group agree how these key results should be worded, checking them against the checklist the group agreed at the start of the session.
Ensure the team are able to confirm a baseline metric to start from and the minimum target they want to reach — you may also wish to add in a ‘gold’ target (Although remember the 80/20 rule!)
Ask the group to run a fist of five against each of the final key results to check they have a consensus.
Why use these techniques?
A modified version of the Min Specs liberating structure proved really useful in the sessions I ran with my teams as everyone in the group had lots of ideas on how we could measure our objectives. Suggestions have never been a problem, the difficult is making sure that suggestions can be measured/are measurable/quantifiable etc.
So, by starting the group off on listing everything we have a good big list of things which we can then start eliminating from.
Getting the group to then come together and discuss the proposals was incredibly useful as we discovered how viable some suggestions were, this led to some suggestions naturally being eliminated whereas others became much more viable suggestions for key results once the group debated them together.
Having formed the checklist at the start of this session we had a very good set of ‘acceptance criteria’ for our key results.
To aid this, using 5 whys meant the team had a technique they could use to question/challenge one another to get to the root of why a key result was important/useful.
Finally, running fist of five to check the consensus/alignment of the group at the end was the final technique I used to confirm everyone was happy to proceed.
As I mention at the start of this post, I don’t suggest this will solve every teams’ difficulties with setting good OKRs, but certainly it aided my teams in our most recent sessions.
I’d love to hear how you get on if you use any of the above or if you have any suggestions for improvements to the template.
If you’d like the word template of this session, you can contact us via our social media or email firstname.lastname@example.org and we’ll be happy to send it to you.