@Written in early 2019 - there are 2 unfinished sections, but I decided to publish anyway 🤷♂️
We started really using OKRs on the whole company at the beginning of 2019. After completing our Series B round on Q4 2018, we started planning how to grow Liv Up team from 150 people in October to more than 220 in February (today we're 350 people).
The main challenge of growing teams this fast is usually around communication.
How do we make the company strategy clear for everyone?
How do we assure everyone has enough context to make good decisions with autonomy?
How do we align every team's efforts so we're all going on the same direction?
How do we connect every team's work to the overall mission and purpose of the company?
OKRs is a common tool used to attack this challenge and help answer these questions.
But even though OKRs have gained a lot of popularity over the last few years and much has been written about the system, it's still quite hard to start using it. Most of the examples are theoretical or from big companies and some examples confuse more than they help.
We did our homework, read about it and talked to people who implemented it successfully on their or other people's companies (thanks to Kaszek's, Endeavor's and Google Launchpad's networks), specially startups around 100~500 people in Brazil.
The idea of this post is to share our current understanding of OKRs for our specific context at Liv Up after all this research.
Why use OKRs → empower the team
- To make the company's vision and goals clear, enabling autonomy & responsibility
- Give context so people know and understand the direction we want to go together (scale business vision)
- Give context so people make good decisions on their on (scale decision-making)
- To make tangible how the company's purpose is aligned to growth, motivating and inspiring
- Connect our mission to the day-to-day of our teams
Format
Objective: qualitative phrase (inspiring when possible) explaining what we want to achieve.
Key Results: quantifiable/measurable goals that translate what success of the Objective looks like, given our strategy.
If we know the Results we want, we can think of the Objective as the answer to "why do we want to achieve that?".
Example:
O: Be recognized as an inspiring brand in Brazil
KRs:
- 100.000 active customers per month by the end of Q1
- 80% of ingredients being organic by the end of Q1
Hire a branding agency to revisit our branding strategy
→ That's a "how", a project. Not a result.
Q: "Why do we want 100.00 active customers and raise organics to 80%?"
A: "Because we want to be recognized as an inspiring brand AND we believe that if more people know and shop Liv Up and we use more organic ingredients, we have high chances of achieving that."
This example is of a Company OKR. The first KR would be part of a Growth Team OKR (something like "Make more people try Liv Up") and the second from a Consumption Team OKR ("Make our production more sustainable"). That Growth OKR could unfold into User Acquisition OKRs, that could eventually unfold into team members' OKRs ("Raise Liv Up presence on Instagram").
The Objectives therefore unfold the company strategy (we believe that to grow we need to have a bigger presence on Instagram), i.e., point a direction, but don't necessarily tell us how to do that (an ideal KR would be something like "raise in 20% the revenues from influencers", not something like "close 5 deals with big influencers" - what we need to achieve, not how to achieve it).
That's the main difference between OKRs and Roadmap. OKRs tell us where we want to be by the end of the quarter and why. Roadmap tells us how, and can be more flexible (e.g. monthly).
Implementation
Gradual rollout
The benchmark is that it takes 4 to 5 quarters to implement successfully across the whole company and to do it step by step, so people understand how it works and get used to it:
Q1: Company OKRs
Q2: Company OKRs & Teams OKRs
Q3: Company OKRs & Teams OKRs & Senior leadership individual OKRs
Q4: Company OKRs & Teams OKRs & Individual OKRs
The rollout strategy is: every time we're doing one of these for the first time, start top down. Then gradually build with the teams.
(we actually did that only to Company OKRs, Teams OKRs we built collaboratively from the first time. And we're not doing Individual OKRs, at least not this year)
Coming up with the Os and the KRs
← explain how the process worked inside teams and how to ensure alignment →
Cross team OKRs (in portuguese)
Good KRs x Bad KRs
North star post Reforge ← everything in between → Projects
Common pitfalls
- Confusing OKRs with performance review
For OKRs to work properly, it must be clear for the team how individual performance is evaluated. That's a hard requirement. People should know how they're promoted or get raises.
→ OKRs are not for that.
When it's tied to performance, people put lower goals they know they'll hit, so they don't have incentives to innovate and are actually rewarded for that. The methodology says OKRs should target 60-70% of the Key Result to stretch and make people think out of the box.
Here at Liv Up, our 100% is already stretchy. We have a strong culture of bold goals and growing ourselves to achieve them. So our OKRs are made to be hit 100%.
- Trying to implement all at once
People need to get used to it, so start simple and make everyone feel how it works. That way they'll understand the format and the goal of the framework, and will come up with better OKRs for their teams or themselves.
- Focusing on Outputs instead of Outcomes
The idea of OKRs is to avoid projects and focus on results. Not on how to achieve results, but in metrics that translate what success looks like for a certain objective.
Projects go into the roadmap. They're two sides of the same coin. The roadmap contains the big projects we'll focus on this quarter/year, while OKRs contain the direction we're going. One ends up influencing the other, it's not a one-way road.
- Set and forget
OKRs are meant to tell you where the company wants to go so we can prioritize our efforts and point our shared creativity to the same direction, while also making explicit whether we're getting there or not.
If we set and don't keep track of them, how do we know if we're making any advance?
Articles
Added 2024: What if you can't measure the outcomes?