My OKR Checklist
There's so much being said about OKRs that I have to believe most people working in any managerial position at a tech company are already aware of them.
A quick survey of videos / books will tell you: Objectives are qualitative, KRs are quantitative. They're stretch goals, some are aspirational and some are committed. They can be shared, or cascaded, they're better for teams, but can be used for individuals. Do not make them a part of performance management, lest people start sandbagging and limiting innovation. They should be implemented across the organization and have a regular operating cadence.
There are books, trainings and a small ecosystem of tools around OKRs now. My signal for when some "technology" has become mainstream is when there are "experts" and consultants touting how they can help you adopt it and reap benefits like the best tech companies in the world (Google being the de-facto example here).
Here are some do's and don'ts based on my experience when crafting OKRs. Think of this as a checklist used to review OKRs:
(For ease, I'm assuming OKRs being reviewed are set for quarters and at a team level)
1. Is the objective worth the effort?
If your customers (the team's customers) or executive leadership, or the CEO or the board read your Objectives would they say - yes!, its totally worth spending the next quarter in pursuit of this objective. If the answer is no, you probably need to re-craft your Os
2. Are the KRs measurable, multiple times?
Can the team measure this OKR multiple times in the quarter? Can it be affected and moved daily or weekly? If the answer is no, you’ll have no way to measure progress through the quarter.
3. Are the KRs sufficiently insulated?
Is this team's work the primary way to affect the KR? or will some other unrelated team's work also lead to this KR? This is a little counter-intuitive, it seems to discourage collaboration (having multiple teams work on the same KR). If you are in a situation where multiple teams need to have the same KR, the teams should be merged into a cross-functional group. (see point 6 below as well)
4. Can you trust the Input KR?
An “Input” KR implies doing more of something is going to lead to the objective you’re after. If you are considering an Input KR, there should be significant evidence that shows doing more of this activity has lead to this outcome in the past.
5. Can you trust the Output KR?
If the KR is measuring something that the team produces, is there prior context which allows you to believe this output will lead to future outcomes the organization cares about?
6. Is the organization setup to collaborate as needed?
Well-crafted OKRs will regularly span across existing team boundaries. This might show up as shared OKR or at least a shared "O". When these situations happen review how the associated teams operate and whether they have the people / processes / culture to collaborate as needed. You might need to add an additional "Input" or "Output" OKR that enforces collaboration, like a weekly checkin with leaders of the two teams, or a joint "progress review meeting".
7. DO NOT try to cover everything a team does in OKRs
OKRs should only reflect your key strategy and tactics. When reviewing, I often look for smells like “well this isn’t really part of our strategy, its more like keep the lights on”.
8. DO NOT try to math your way into OKRs
Mathematically breaking down a KR and converting each driving metric into an OKR is a seductive trap. It encourage existing silos, gives you a false sense of "completeness" and ultimately no more than modest incremental improvements. Don’t do it.
9. RESIST the temptation to change OKRs within a quarter
This is a hard one. Often teams will encounter something mid-quarter that leads them to believe an OKR is no longer relevant or achievable. This could be because the team has reached its limit in terms of being able to innovate. Bring in your best problem solvers to work with the team for a short period of time, host open brainstorming sessions, do something to shake up the cobwebs. If after all that the OKR still seems immovable, add a new one, don't change an existing one.
10. Every OKR should be designed to "stretch" the team
Stretch is not about working harder or even faster. Its about creating a challenge for the team that is big enough to not be boring; and not so big that it leads the team into anxiety.
More important than anything else - Write your OKRs, Commit, and try your best. You might surprise yourself.
To learn more about OKRs:
Book: Measure What Matters and its companion site - The book is great for inspiration, the website for some practical how-to guides. I personally found the website far more useful than the book.
Book: Radical Focus (I read the first edition) - A story based book, that shows one way to use OKRs to align on work in the early days of a company.
Self-paced online course - If you prefer a video format.
