Back to resources

Establishing Annual OKRs For Your Team

Goal Setting

26 February, 2020

Paritosh Aggarwal

Paritosh Aggarwal

Engineering Manager at Airbnb

Paritosh Aggarwal, Engineering Manager at Airbnb, speaks on why he chose to execute annual OKRs for his team, how he did so, and the benefits he has seen from it.

Problem

I was working with a team that did not conceptually understand what they wanted to achieve long-term. They were very much involved in the day-to-day activities and everyday demand of what needed to get done. There was a lot of focused effort on Sprint planning but with no clear direction or laid out roadmap. As a result, I decided to initiate an annual objective planning scheme.

Actions taken

In short, this meant setting yearly OKRs- which were in addition to the quarterly OKRs. The guidelines for the annual OKRs were as follows: Inspirational objectives which people could relate to;

  • The goals should be ambitious;
  • They should also be widely important, for example things that are existential and critical to the success of the team;
  • The key results needed to be specific and measurable so that you could grade yourself, or so that anyone could objectively grade how the team performs on those specific measurements.

To establish specific objectives I held a group session with the team and asked them to identify key work streams. Everyone determined their main priorities, the “what’ part of “What is the team doing?” Then I posted a discussion around who was responsible for these things, specifically which teams. I wanted to make sure that we were not going above three or four work streams (or teams) because these would essentially capture the objective portion of the OKR. Ultimately, we landed on three very important goals, or objectives, from this exercise: reliability, platform capabilities, and usability, and from these three goals stemmed out our three major teams.

The next step was to establish the key results (KRs) for these objectives. This process was a bit more difficult because we found it hard to precisely measure how to hit certain KRs. But eventually we managed.

  • The KR for reliability was the easiest because we wanted to make sure our uptime was above a certain threshold. We had a maximum 1-hour downtime per quarter for our services which led us to come up with a time goal of 99.95%. This is what success would look like the following year if we took out all of the staffing problems and all of the short-term tactical problems. This is what we wanted to achieve.
  • We knew we wanted to ship two new services as our objective for platform capabilities, therefore we had to launch and enable two new use cases. For both of those two use cases we had key results around adding X new features. As a result, we had KRs for new services, new use cases, and new features, but they were simultaneously all tied together.
  • The third objective, usability, dealt with having a KR that established an NPS service, a customer satisfaction survey. The survey results should have a base rate of say 4 out 5, for example. Specific and measurable. Additionally, we wanted to enable non-engineers to achieve certain aspects of the product itself without any engineering help, thus the KR would be a self service where people could create 100% of the campaign independently.

I wrote up the OKRs with the three objectives/themes/teams and the key results. The team as a whole felt good about them. It also energized individuals. Then, we did a sharing exercise with the leadership team, a buy-in process of sorts. I had to give them context around the goals, priorities for the team, and the sequence of one goal over another. They were interested in the framework, teams that were devised from the objectives, and the exact outcomes that we set for ourselves. As a result, we wrote a company-wide memo-style document so people would have an external narrative of what it was we were doing, how it contributed to the platform, and what we wanted to achieve by the following year’s end.

Lessons learned

  • OKRs are a great goal setting framework because often times there is a lack of clarity on whether a team is hitting a goal or not. Yet, with OKRs you are forced to sign yourselves up for a clear objective that will result in a precise measurable outcome. This leads to fantastic accountability and clarity from the team on what is to be done and what they have to hit.
  • There should be a discussion around whether to set conservative or ambitious goals. I’m always in favor of going for it and setting ambitious goals but this will depend on the state of your team. In this situation, my team’s morale was good and we had a centered direction, therefore striving for more helped push people beyond what they were capable of and allowed them to achieve that long-reaching goal.
  • Due to the fact that we spent dedicated time with the team to come up with these goals and because we generated specific outcomes, the team felt invested and connected to the essence of the goals. It gave them the opportunity to ask questions and challenge the objectives while also paving the way for easy buy-in from individuals.
  • The OKRs that we established were relatable and easy to understand. So much so that when someone new came into our work stream, or when somebody asked a peer a question, the conversation would ultimately return back to the objectives and key results. Individuals were able to see for themselves the larger picture, why certain goals were important, and the direction that we were headed. The OKRs acted as a self-alignment tool.
  • Three is a good number to stick to when establishing annual OKRs. Any more and there’s suddenly too many goals and none of them are super important. Too little and people will hyper-focus on them. Preferably three OKRs, three initiatives. They are easy to remember and you can incorporate the well-roundedness of what your organization wants to achieve.
  • Another by-product of the annual OKRs was that we were able to get a very clear handle on what success looked like. It gave the team extreme clarity on priorities. It also empowered everyone on the team to ask questions and determine whether a task was contributing to our goals. Instead of picking up random work from here and there, individuals were really focused on the objectives and what we wanted to achieve by the end of the year.

Discover Plato

Scale your coaching effort for your engineering and product teams
Develop yourself to become a stronger engineering / product leader


Related stories

Providing Better One-on-Ones
21 July

Jibin Jose, Senior Engineering Manager at Booking.com, takes pride in his approach to managing each member of his team on an individual basis.

Meetings
Feedback
Coaching / Training / Mentorship
Goal Setting
Jibin Jose Jose

Jibin Jose Jose

Senior Engineering Manager at Booking.com

Setting Goals and Evaluating Performance
21 July

Trey Tacon, Senior Director of Engineering at TeamSnap, holds his engineering staff to a very high standard, using OKRs to connect the dots between their work and the company’s mission as a whole.

Performance
Goal Setting
Trey Tacon

Trey Tacon

Senior Director of Engineering at TeamSnap

Developing a Strategic Plan
15 July

Joao Esteves, Director of Platform Engineering at Workhuman, reiterates his effort to organize the team’s vision and mission to bring a great deal of stability.

Strategy
Alignment
Mission / Vision / Charter
Goal Setting
Joao Esteves

Joao Esteves

Director of Platform Engineering at Workhuman

Finding Unity Through a Common Mission
21 July

Sameer Khanna, Senior Vice President of Engineering at Pager, talks about one of the most influential turning points of his career: leaving the frivolous world of e-commerce in order to make a difference somewhere more important.

Alignment
Mission / Vision / Charter
Goal Setting
Sameer Khanna

Sameer Khanna

SVP of Engineering at Pager

Optimizing for Results Instead of Velocity
17 June

Jose Muniz, Vice President of engineering at People.ai, has refined the product development process to a near science, focusing not on speed of output, but, rather, on delivering results.

Dev Processes
Alignment
Goal Setting
Jose Muniz

Jose Muniz

VP of Engineering at People.ai

You're a great engineer.
Become a great engineering leader.

Plato (platohq.com) is the world's biggest mentorship platform for engineering managers & product managers. We've curated a community of mentors who are the tech industry's best engineering & product leaders from companies like Facebook, Lyft, Slack, Airbnb, Gusto, and more.