OKRs as a goal-setting framework

At one of my previous jobs, a mechanism of OKRs (Objectives and key results) was the main goal-setting framework. Every year at a company level a main goal, a company objective, was defined, and each team quarterly defined their Objectives and Key Results towards the main goal.

During that time I love this approach, which you by the way can also apply to personal goals. Today, I want to share my reflection on the framework and inspiration to give it a shot.

What OKR is

The concept is very simple. If you are already familiar with it, you can safely skip the section.

As I already mentioned, this is a goal-setting framework, designed to help structure a path from A (the current state) to B (a desirable state) of a project, product, personal development, etc. It is meant to motivate and guide a person or a team toward a goal.

The concept (OKRs) was popularized by Andrew Grove back in the 1970s and it contains 2 parts:

1. Objectives (Where do I want to go?)
2. Key Results, or, simply put, milestones (How will I pace myself to see if I am getting there? Measurable success criteria used to track the achievement of that goal)

To be more specific, let's imagine that we work in a marketing department and we want, that more people will use our product. One of our OKRs next quarter could be:

Ex. 1: Objective: Increase brand awareness

• Key Result: Drive 1M web visitors
• Key Result: Increase social media following by 10x
• Key Result: Recruit and onboard 1,000 community members

Another example — we work at a company, that wants to commit to sustainability:

Ex. 2: Objective: Create the lowest carbon footprint in our industry

• Key Result: Supply chain and shipping infrastructure 100% zero waste
• Key Result: 25% of material is compostable
• Key Result: 75% of material is biodegradable

As you can see, an Objective should be broad and ambitious, but to reach the objective you should have obvious and measurable criteria. Measurability is key and it should be an indicator if a person/a team is on track with the main goal. In 99% of cases, measurability expresses in numbers — 1M web visitors, 100% zero waste, 25%/75% of material, etc.

But don’t scope OKRs only to a job. You can apply OKRs to personal development. For example:

Objective: Run 10 km in under 50 minutes in 2 months.

• Key Result: Go for a run 3 times a week for at least 30 minutes
• Key Result: Increase the distance of run by 1 km every week until you reach 10 km
• Key Result: Increase speed by 5 seconds every week

As a general rule, a number of objectives shouldn’t be huge. A team should know what they do and shouldn’t lose focus. At the same time, a number of KRs shouldn’t be large. As a rule of thumb, 3–5 key results are enough.

Retrospective and grading OKRs

The Andy Grove method of grading OKRs is a simple “yes” or “no” approach: Did you meet it? Or did you not?

For my personal taste, a system with 3 “colors”: red — 0.0, yellow — 0.5, and green — 1.0, which is applied by many companies as well, reflects a path better. The purpose to have a partially achieved key result (yellow color) is:

1. Demonstrate a work, that was done toward an objective. States “Drive 1M web visitors“ and “Drive 0.5M web visitors” are not the same and even a partially achieved key result sheds some light on a potential to achieve the objective (increase brand awareness) by means of further increasing the key metric (millions of web visitors).
2. Metrics, that are put in KRs sometimes only hypotheses. They reflect expectations - what value (an increase in social media following) we should have in order to achieve the objective. In reality, the objective can be achieved by a greater (or a lesser) value of the KR. So we should consider steps, that are taken on the road to the objective.

After we grade each KR, which is simple and shouldn’t have ambiguity, we can estimate an overall score of the objective as a simple average of key results scores. A magic number here is 0.7 (or 70%), so don’t be upset if, at the end of a quarter, you see a lesser number.

A motivation for 70% is to have goals (and key results) ambitious enough. They shouldn’t be completed 100%, otherwise, it means that during the planning we were not accurate in the numbers or we didn’t stretch ourselves harder (By all means, performance of individuals should not depend on OKRs scores. It’s a marker for a product, a company, not for a person).

Frequency of OKRs

Generally and is advised by Andy Grove himself (his speech on youtube) to run an annual and a quarter version of OKRs.

The annual version is done in conjunction with the planning process, when each entity (a team in the case of a company) puts together a set of annual objectives and key results, going with them, that are evaluated and modified, if required, on the next planning.

The more significant is on a quarterly basis. A quarter is long enough to make progress, that can be felt on an organization or company level, and at the same time is short to make a conclusion about the progress and adjust OKRs next quarter. At our company within a team, we reserved a couple of weeks from a quarter to formulate OKR toward a company mission. All objectives and key results were reviewed carefully and picky by a top management team in order to avoid ambiguity and ensure alignment between teams toward the vision (with what to expect from each team).

How do OKRs help?

I prefer thinking about OKRs as a missing link between a business and a team's efforts. OKRs connect a business mission (that is, all in all, money, users, customers) with single-team projects. This is why they are important and should be considered thoughtfully during each planning.

Back then, our team could write down an extensive list of projects (10–20, easily), that we can implement. This is the reality — you always have tons of ideas in mind, but you shouldn’t follow them right away. You step back and see in a broader picture where it leads you. Only projects, that contribute the most to chosen key results should be selected and executed. This is so simple, but many people easily forget it and pick the most attractive projects, trying to fit the data to the desired.

First — always should be an objective. A clear and simple phrase of how a team helps the whole company. Second — key results, concrete and specific measurements, that cannot be interpretable differently by different people. And projects and ideas — at the end, thinking about how to help a business in words and metrics, that are understandable by the business.

Does the OKRs framework work?

Enough has already been said about what it is and how it can help a company, but the central question is — does it really work?

To answer the question, let me make an analogy. When you start learning a new programming language or a new framework you unavoidably make mistakes and use constructions, that don’t fit well for a use case. So is the OKRs framework, as it requires time and effort to master.

At our company, we made about 4–6 iterations to be at the point, where our objectives were challenging and apparent, and key results, that indeed reflect milestones and didn’t create ambiguity. 4–6 iterations sum up in more than 1 year of time but don’t be disappointed, as the approach still leads to the goal, only milestones are less visible and sometimes it’s hard to tell if you pass it or not during a retrospective, that you adjust quarters that come.

The hardest part for us was to come up with numbers of key results. Let's consider the example above:

Objective: Create the lowest carbon footprint in our industry

• Key Result: Supply chain and shipping infrastructure 100% zero waste

Why 100% and not 80%? Maybe with a lower “zero waste” percentage we can achieve the objective too. The question is hard to answer and this is where OKRs hide all complexity.

To sum up, I believe, that OKRs is a powerful tool, that requires strictness and precision in wording at the start, but it does its work in setting a goal and keeping a team focused business-wise.

“Google does it”

I hope at this point you already want to try this approach at least for your personal goals, such as learning to play guitar, getting promoted or even starting a new business, and consider reading other resources on the Internet with more examples to structure the journey (Consider reading a classical High Output Management by Andrew Grove).

I would like to warn against worshiping the framework though. Too often, when someone explains OKRs, he mentions Google and how this idea helped Google to grow 10x, 100x, or even 10…0x times faster. But Google is simply an outlier in all cases. Did Google grow a lot because of OKRs or OKRs are popularized because of Google's success?