OKR’s history goes back to 1954, when Peter Drucker invented the Management by Objectives approach. In 1968, Andy Grove co-founded Intel, where he developed MBO into the OKR model as we see it today. John Doerr, who joined Intel in 1974, learned the OKR framework and went on to join Kleiner Perkins Caufield & Byers (KPCB), which was one of Google’s first investors. John Doerr became an adviser to Google in its early days and introduced OKR to Google’s founders, Larry Page and Sergey Brin, who implemented OKR at Google, which still uses the OKR methodology today.

The concept

OKR stands for Objectives and Key Results. The OKR methodology is a popular strategy implementation framework. The benefits include improved focus, transparency, and alignment required to organize and move employees to achieving shared, bold corporate goal. An OKR consists of an Objective, which defines a goal to be achieved, and up to 5 Key Results, which measure progress towards the Objective. Each OKR can also have Initiatives that describe the work required to drive progress on the Key Results. Usually, the Objective is qualitative, while the key results are quantitative. Companies usually set objectives for a specific time-period, usually a quarter. They track the key results and metrics to ascertain whether they have achieved the objective by the end of that time.

As consultants, you need to understand your client’s mission before you recommend the OKR methodology; it is critical that your client has a mission. Without a sense of purpose and mechanisms to achieve it, anything you recommend is equally OK because your client’s employees and business lines will have different interpretations of the purpose. However, if your client does have an overarching mission and you have determined it, methodically challenge it by recommending stretch, but realistic objectives.

A proper goal has to describe both what you will achieve and how you are going to measure its achievement. The key words here are “as measured by,” since measurement is what makes a goal a goal. Without it, you do not have a goal, all you have is a desire. Doerr’s formula is the best way to explain the structure of an OKR:

I will (Objective) as measured by (a set of Key Results).

Objectives

Start by defining 3-5 key objectives on company, team or personal levels. Objectives should be ambitious and qualitative. The objective or goal is designed to compel action. While CEOs and VCs may rejoice over a 10% customer conversion gain, most employees are excited by a sense of vision, meaning and progress. As consultants, use the language of your client. If your client uses aggressive terminologies, such as crush, eliminate, or drive out, use that wording. On the other hand, for clients that use balanced nomenclature, use the appropriate terms in-line with their terminologies. An example of a good objective is:

Launch a new website to delight our customers

Key Results

Under each objective, define 3, 5 or 7 measurable results that the organization should achieve. Fewer is better, but ensure that there are no more than 7 key results per objective. Key results should be quantifiable, achievable, lead to objective grading, be difficult, but not impossible, and, and SMART (Specific, Measurable, Realistic, and Time-bound). OKR results can be based on growth, performance, revenue or engagement. The results are often numerical, but they can also be categorical. You can use a binary 0 or 1 to evaluate whether you can achieve a result or otherwise. Key results help answer the question – “How would we know if we met our objective?” An objective, such as “Launch a new website to delight our customers” might have Key results, such as:

1. Interview 20 existing customers on what they would like to see on the website
2. Reduce customer cancellations from 10% to 4%
3. Get 60% of users to come back twice in one week
4. Recommendation score of 8

With Key results, you can ensure that you have potentially opposing forces such as growth and performance, or revenue and quality and balance these drivers. Key results should be difficult, but not impossible. One great way to do this is to set a confidence level of five of ten. A confidence level of one means that the result cannot be achieved, while a confidence level of ten is also known as sandbagging. You are looking for a sweet spot, where you are pushing yourself and your team to do bigger things, where you have a 50% shot at failing. Google uses a scale of 0 – 1.0

In his video on OKRs, Rick Klau mentions that:

Objectives are ambitious, and should feel somewhat uncomfortable. The “sweet spot” for an OKR grade is .6 — .7; if someone consistently gets 1.0, their OKRs aren’t ambitious enough. If you get 1s, you’re not crushing it, you’re sandbagging.

Moonshots and Roofshots

The type of OKRs that Klau is describing are called Moonshots. Moonshots are:

• Stretch goals
• Just beyond the threshold of what seems possible
• Success means achieving 60-70%

In practice there is also a second type of OKR, the Roofshots. Roofshots are:

• Goals that are hard but achievable
• Success means achieving 100%

Organizational Maturity

Moonshots are a foundational building block of OKR, but they require a lot of organizational maturity. In my experience, moonshots can cause a few issues:

1. They can demotivate the team. People like to beat goals. Only achieving 60% of the OKRs can demotivate a lot of them, especially in the beginning
2. Lack of accountability and commitment. Moonshots can be misinterpreted by some, creating a culture in which you don’t have to reach your goals: “Hey, it doesn’t matter. It’s just a stretch goal”
3. Alignment issues. Especially when using activity-based Key Results, moonshots can cause alignment issues between interdependent teams. One team needs something from another but the second one is unable to deliver it since it was a stretch

Roofshots exist to overcome these issues. Several teams inside Google use roofshots, usually mixed with moonshots. To develop a results-focused culture, begin by focusing on beating goals through roofshots. Later, when the culture matures, you can evolve to moonshots to push the organization forward.

OKRs cascade

The Leadership should set overarching OKRs, and then each team should determine how their OKR leads to the overall objective. In traditional organizations, goals cascade. The cascading model is a residue of a command & control mindset in which objectives and decisions simply flow downwards the organizational hierarchy from the top. However, this approach takes way too much time and is inefficient. As James Harvey wrote:

The traditional model is a top-down approach and often takes too long to achieve alignment. Direct reports are often dependent on the completion of their supervisor’s goals before they can begin building their own goal plan.

OKRs should be set in a parallel process in which teams define OKRs that are linked to the organization objectives and validated by managers, in a process that is simultaneously bottom-up and top-down. From the company OKRs, the teams can get a clear direction and understand how they can contribute to reaching those OKRs. Each team then defines a set of tactical OKRs for the quarter that contribute to the strategic OKRs and that roughly align with them. Teams’ OKRs don’t have to be 100% aligned with the company’s OKRs since they may also choose to include a local OKR. They can focus on a single key result or try to support the entire set. For example, engineering might decide customer satisfaction is highly correlated with their website’s page load time, so they may set a goal, such as:

The website should load under 3 seconds

Some teams can easily align their OKRs with the company OKRs, while others may have to dig a little more to make sure they are supporting the right objectives. Much of the value in OKRs comes from the conversations on what matters, how that’s measured and what it means for some teams who are often used to working with their own standards.

As Laszlo Bock, Google’s former VP of People Operations wrote in his book Work Rules!:

On the topic of goals, the academic research agrees with your intuition: Having goals improves performance. Spending hours cascading goals up and down the company, however, does not. It takes way too much time and it’s too hard to make sure all the goals line up. We have a market-based approach, where over time our goals all converge, because the top OKRs are known and everyone else’s OKRs are visible. Teams that are grossly out of alignment stand out, and the few major initiatives that touch everyone are easy enough to manage directly. So far, so good!