As 2018 is starting and many teams kick off the year with setting new goals, many choose OKRs – Objectives and Key Results – as the foundation. At Weekdone, we've worked with thousands of companies as their OKR software provider, but also a consultant.
At the same time, we're big fans of independent OKR coaches. One of the best in the world is Felipe Castro. He is a Goal Hacker. He helps organizations transform how they use goals by adopting OKR, the Silicon Valley framework for goal setting. He created the Set-Align-Achieve cycle, a simple method to avoid OKR’s most common pitfalls.
We asked Felipe a few questions on OKRs, what works and what does not. Enjoy!
You've worked with OKRs for years. What would be the main things you'd suggest someone getting started with them in 2018 to be successful in the process?
After training thousands of people around the world in OKR, I created the Set-Align-Achieve cycle, a simple method to avoid OKR's most common mistakes.
If you are getting started with OKR, make sure you follow the three steps:
- Set OKRs that measure impact instead of tasks,
- Align them with other teams to solve interdependencies,
- Create a weekly follow-through cadence to Achieve them.
What's your take on not using individual OKRs, and using just company, department or team ones?
My advice is: simplify. Use as few OKR levels as possible. I have worked with several mid-sized organizations (with over 1,000 employees) that had only two levels: company and team. This approach makes the whole process simpler and faster.
Individual OKRs are not for everyone and should never be required. Good OKRs follow good metrics, and for many roles, the indicator belongs to the team. In fact, companies such as Twitter and Spotify abandoned individual OKRs.
Many want to use project completion stages or milestones as their KRs. Is that OKR, or should it be avoided at any cost?
The question companies have to answer is: "Do we want to make decisions based on opinions or actual data?" What happens in most organizations is that someone writes a business case full of assumptions and then team executes a project. But they never measure if the project delivered the benefits promised in the business case. There is zero accountability for results. This planning theatre creates vast amounts of waste since projects often fail or are canceled.
The main benefit of OKR is to create a culture that is data-driven and results-focused. If you want to track projects, regardless of the value they deliver, OKR is not for you. There are better tools for that.
What do you think of the topic of using just one objective initially for each person or team? Should people try that?
You should always try to reduce the number of Key Results. I care more about that, as sometimes splitting one Objective in two makes them more memorable. Having up to five Key Results for each team is excellent. But to create cross-team alignment, you may need to go over that to use shared OKRs.
Even so, no team should have more than 10 KRs – and I am not saying that every team can have 10, the fewer, the better.
Paraphrasing Jason Fried and DHH, make each Key Result work hard to be adopted. Make each KR prove itself and show that it's a survivor.
Anything you'd suggest someone to avoid when they do their first quarter of OKRs?
Don't follow the Google Ventures OKR video. I have a huge admiration for Rick Klau, and his video was crucial to promote OKR, but it was never meant to be the reference to it. Many people try to follow the video to the letter, and they fail horribly. Klau recently corrected the video on Twitter, but they never made an updated version. If you want to learn about OKR, there are way better sources, such as my Beginner's Guide to OKR.
For more information about the work Felipe Castro does with making OKRs a success, see his website.