Who created OKRs

OKRs stand for Objectives & Key Results. They are a popular goal-setting framework for organizations, particularly within the startup and tech communities. OKRs were an evolution of MBOs (Management by Objectives), created by Andy Grove in the 1970s, the long-time CEO at Intel. Many decades later, OKRs were popularized again by Google, after venture capitalist John Doerr introduced them to the company.

Google obviously took off, and Doerr very publicly credited their operational excellence to the OKR process. And, he’s become their biggest advocate ever since, even writing a book about OKRs called Measure What Matters and maintaining a website of resources.

When I was building my software company, Classy, we ran the company on OKRs for many years. And during that time we grew from a handful of people, to hundreds, and really stress-tested the system. So, I’ve seen it all with OKRs – the good, the bad and the ugly!

And that’s why I wrote this article. To highlight the top 7 pitfalls I ran into when using the OKR system. These seven things aren’t talked about enough. But now that I’m advising hundreds of founders, I know they’re not unique to just me.

I’m hopeful this article will help other founders and entrepreneurs navigate these 7 costly pitfalls and find huge success on the other side!

Let’s dive in…

What are OKRs

The “O” in OKR stands for “objective” and the “KR” stands for “key results.” The objective is the goal and the key results are measurable milestones underneath the goal. In Andy Grove’s language “the objective is the direction” It’s where you want to go or what you want to achieve. And a key result is “a milestone” that “has to be measurable.”

There is no limit on how many OKRs you can have as a company. There is also no limit to how many Key Results you can have per Objective (although most have 3-5 Key Results, most on this later).

Here’s an example of an OKR from Doerr’s website:

By design, it should be impossible to miss the objective while hitting all of your key results. In other words, your key results define what success looks like for each objective.

In my view, adding the idea of measurable key results to each objective was Andy Grove’s powerful insight. Key results make the goal more tangible by making it measurable. They help the team understand what success looks like, allow for tracking over time, and increase the accountability across the organization. 

Despite the many issues I’ve encountered with the OKR system, which I’ll get into in a sec, I still use this basic two part structure when creating company goals (the objective and the measurable milestones).

The Northstar Operating System (that we teach in my community, Highland) uses a simple goal system called WhyGos. This system is where I landed after using the OKRs for years and running into the issues in this article. Check it out for free if interested.

The Fundamental Flaw of OKRs

I really wanted to love OKRs – I have great respect for the creator, Andy Grove, on so many levels. But Intel was a massive company with thousands of employees when he came up with the idea. So from the very beginning, it wasn’t perfectly suited for startups, small teams and companies under $100M in revenue.

Yes, Google was a famous “startup” that popularized the system. But you could argue that Google, a once in a generation company, could easily ignore their flaws, simply because they were growing so fast.

So what about the other 99.99% of companies? Like the one I founded, Classy. 

The problems we experienced with OKRs didn’t come from the structure of an individual objective and key result, they came from the design of the overarching system itself. 

For example:

• How OKRs are created 
• How OKRs are measured and graded
• How OKRs align to one another (or don’t)

If I were to boil it down to a theme, it would go something like this: 

Too many options leads to hidden, but dangerous complexity

To be clear, staying flexible and adaptable as an organization can be a great thing. Especially in the age of AI. However, too much flexibility at the system level can backfire. When the guardrails aren’t defined, or well-understood, it creates too many options and a lack of focus across the team. 

And the opposite is also true. If a system is too rigid, it will constrict creativity and innovation across the organization. Too many rules leads to a lack of autonomy and an overdependence on the leader. 

Think of it like this: 

So where does the OKR process fall? If ‘extremely flexible’ was a 0 and ‘extremely rigid’ was a 10, I’d give the OKR system a 2. It’s designed for maximum flexibility and choice, which erodes operational focus and efficiency overtime. You’ll end up spending endless hours debating fringe ideas, instead of driving execution against a mutually agreed upon, select set of priorities that every person in the company helps achieve. 

Now that you understand the main theme here, I’m going to address each of the top 7 problems we faced with OKRs, and how to fix each one.

#1 – Too Many OKRs

When the OKR system first came out, there was no limit to the amount of objectives you could have. The average company was creating 5-10 OKRs (including Classy) at the company level alone. I don’t know about you, but I find it tough to “prioritize” 5-10 things at any given time. 

Say you have eight company objectives each with five key results. You are now tracking and updating forty key results at just the company level. Then each team and individual creates their own set of 5-10 OKRs and its not hard to see how this can easily get unhelpful and overwhelming. 

To his credit, Doerr in his 2018 book, did suggest that an organization should try to focus on 3-5 company OKRs. But I’ve found that three company goals is the maximum that any organization can reasonably prioritize at any given time. You’ll start to see diminishing returns after that. Three (or less) drives the optimal amount of focus and alignment. 

But I can hear you now… Scot, we can’t possibly limit ourselves to only three company goals!!! Yes, you absolutely can. In fact, it’s this contraint that forces the tough conversations around your true priorities. If you choose 5, 7, or even 10 objectives, you’re playing it safe. When everything’s a priority, nothing’s really a priority. 

But here’s an important note. Assigning 3 or less company priorities doesn’t mean that these are the only three things the company works on. It simply means, these three things are the company’s top 3 priorities for the year. And everyone should contribute in some way to achieving them. 

#2 – New OKRs Every Quarter 

Doerr writes in his book, “the best OKR cadence is the one that fits the context and culture of your business.”

Huh? I guess this means it’s up to you. 🤪

But from my experience, most companies that use the OKR system reset their OKRs every quarter. And, having lived this process for years, I can tell you that the administrative burden this creates far outweighs any upside from a more frequent cycle. 

Doing OKRs quarterly also leaves you rudderless on strategy. It’s too easy to flip flop all over the place chasing the next shiny object. At Classy, this quarterly OKR approach killed our medium-term (and long-term) alignment around one cohesive strategy. 

Eventually, we shifted to annual OKRs, and created quarterly targets to ensure we were pacing correctly. This annual approach allowed us to put the company’s vision & strategy into action. It gave the team strong conviction at the macro-level, but lots of flexibility at the micro-level. 

Later in Doerr’s book he says: “The best practice may be a parallel, dual cadence, with short-horizon OKRs (for the here and now) supporting annual OKRs.” So it seems that we landed in the general neighborhood of Doerr’s best practice (given the book didn’t exist at the time). 

But there was one big difference. Our quarterly targets were set at the beginning of the year based on the annual key results. They were literally the annual key results broken into four. We didn’t have to reinvent the wheel each quarter and put the team through the misery of goal creation four times a year. Not a single person complained about this change! 

So, instead of getting caught in this resource-sucking quarterly loop, I’d recommend creating annual objectives and key results at the company, team, and individual levels. Then you can create quarterly targets that break your annual key results into four.

#3 – High Chance of Misalignment

The OKR system doesn’t automatically align individual, team, and company goals. Instead companies are told that about half of their goals should be cascaded down from the company goals, while the other half should ladder up from the bottom (or across). 

Doerr says this provides the flexibility to encourage innovation and fast action. While this sounds “cool” in practice, this top down, bottoms up, and even SIDEWAYS approach creates a ton of alignment issues across the company. Follow this advice and I promise you’ll end up with a tangled web of OKRs that (i) are difficult to understand, (ii) create focus and resource imbalances (iii) include too many goals that don’t help the company. 

What I suggest is requiring that team and individual OKRs be tightly linked to one or more key results at the level above. This could mean that a team, for example, inherits a company key result that directly becomes one of its objectives. The key is that the team, or individual, can clearly explain how their objective is driving towards achievement of one or more key results at the level above. This leaves lots of room for how they get there and what their own key results are; but, you at least know that they are pointed in the right direction. You don’t have one team walking left and the other team walking right. 

As a related side note: Doerr also brings up Google’s “20 percent time” policy – where 20% of a person’s time should be dedicated to independent projects that may or may not align with company priorities. This is fine, and can be great for innovation, but I recommend keeping this outside of the OKRs process. In my experience, if the OKRs become a catch all for everything you do throughout the year, you lose the power alignment. 

#4 – Unachievable OKRs

In many ways I find OKRs to be a system that is too clever by half. How to set goals, and how those goals are graded (next section), are perhaps the two areas where this is most true.

When John Doerr brought the OKR system to Google they adopted a scale where hitting 70% of an objective was considered a success. Actually it’s a bit more complicated than that. They first ask everyone to distinguish between two different categories of OKR– “Committed” OKRs and “Aspirational” OKRs. Committed OKRS are ones where you need to hit 100% to be considered a success. “Aspirational” OKRs are stretchy by design so 70% is considered a success. 

So 70% is a success, unless it isn’t. And if it isn’t, then 100% means success. Oh ya, and combining this with section #3, some OKRs should align to the company OKRs, and some don’t need to. Got all that?

Confusing as this may be, the idea is that most goals should be essentially unachievable so you force people to stretch themselves. That’s how businesses make breakthroughs, the thinking goes. As Google puts it, “The “sweet spot” for an OKR grade is 60% – 70%; if someone consistently fully attains their objectives, their OKRs aren’t ambitious enough and they need to think bigger.”

There’s nothing wrong with stretching yourself. However, this goal creation process did not work for my company. Teams were confused and anxious about the whole 70% model of success. There was way too much variability across teams in terms of how they were thinking about what it meant to achieve their goals. It was a mess.

Instead of adding all of this mental gymnastics, just keep it simple. Set aggressive but achievable goals where 100% means you achieved the goal. But how do you safeguard against sandbagging (making the OKRs too easy)? Managers need to sign off on the final OKRs from their team. It doesn’t need to be more complicated than that. 

#5 – Overly Complex Grading System

We just touched on how the OKR approach views 70% as a success (unless, of course, it’s a committed OKR, in which case 100% is a success!). Believe it or not, the grading system is even more confusing than this!

First you need to grade each key result from 0% to 100% attainment. Then you take the average of those scores to arrive at the score for the overall objective. From there, there’s another layer to it, where the individual gives themselves a self assessment score. They add their subjective color commentary on the raw results, which can push the score higher or lower. 

For example, say you had an OKR to sign 20 new customers in the quarter and you actually signed 15. Your raw score would be 75%. But, you know that one of the customers you signed is the largest contract value in company history and you surpassed your dollar quota for the quarter. Because of this context you adjust your score to 100%.

I find all of this overly complicated and convoluted. When we tried to adopt the OKR purist approach at Classy it led to so much wasted energy with people debating the finer points of the difference between a 0.5 score and 0.6 score. It was confusing and distracting.

Instead of dealing with this complexity I encourage founders to adopt a simpler system. 100% is green, it means you are on track if you’re within the period or that you achieved your goal if you’ve finished the period. 80%-100% is yellow and it means slightly behind or a slight miss. Less than 80% is red and it means you are off track or had a significant miss. 

#6 – Disconnected from Compensation

Another part of the OKR philosophy is that OKRs should be disconnected from compensation. As OKR purists will tell you, OKRs shouldn’t tie to compensation because everyone will just sandbag their goals (make them too easy).

But this one was always a serious head scratcher for me. If OKRs are supposed to represent the priority work a person is doing, why would you not reward them for doing a great job? Imagine getting to a performance review after crushing your OKRs. Yet, your new base salary, or bonus, is determined by an entirely different set of criteria. This would drive any person crazy!

This is how Doerr describes it in his book: “…in today’s workplace, OKRs and compensation can still be friends. They’ll never totally lose touch. But they no longer live together, and it’s healthier that way.” I don’t know about you, but again, I find this rather confusing as an operator.

To be fair, he also notes that Google now makes OKRs account for one third, or less, of a person’s performance rating. I guess the idea is that their impact is obscured, so people will still feel comfortable setting audacious goals. But I’m not convinced. I’d much rather just align the incentives of company and individual more directly.

When it comes to individual compensation, my belief is that OKR performance (or whatever goal system you use) should play a material part in compensation decisions. Not only the results, but how seriously they take the system itself. For example, in my companies, I like to split the achievement of bonus compensation between company objectives and individual ones. I’ve found that a fifty-fifty split strikes the right balance between promoting teamwork and rewarding individual achievement (but you can play with the percentages to your own liking). 

#7 – No “Why” Behind the OKRs

My final critique of the OKR approach has to do with the lack of strategic context. When we were using the OKRs at Classy, it was inevitable someone would ask: “why did we prioritize this objective?”. This might be an individual asking their manager, or the executive team. Or, it could be a manager asking an individual. Either way, everyone was thirsty for context on why the objective was created in the first place.

So we started adding a simple ‘why’ statement (sentence or two) that explained why the objective was prioritized for the company, team or individual. And this simple practice proved more powerful than we thought. It forced the creator of the objective to really think about the rationale for elevating it. Their ‘why’ statement would typically bring the company’s strategy and product roadmap forward so it became more obvious how the objective tied into the medium and long-range planning.

Adding in the “why” is a simple but valuable improvement to the OKR. It enhances communication and alignment, provides crucial context, helps the individual think through their goals, and provides a useful device for ensuring you’ve got the right goals in the approval phase. To me it’s the one thing missing from that simple but powerful two-part structure that Andy Grove came up with way back when.

Parting Thoughts

There were aspects of OKRs that we liked – its why we used it for many years! But as a system, it was too malleable to provide the type of clarity and alignment that we wanted at our company. After fixing these 7 pitfalls and more, what was left didn’t really look like OKRs anymore, and didn’t align with Google and Doerr’s guidance. So we started calling our new goal framework by a different name, and scaled with it to great success.

The name of our goal system became the WhyGos (“why-goes”), which stands for: 

• Why – why this is a priority 
• Goal – what we are trying to achieve
• Outcome – how we measure success

I created the WhyGo system to provide a bit more structure, a lot more guidance, and fix many of the OKR flaws in this article. There are hundreds of founders using the WhyGos already as an alternative to OKRs – and I plan to write a full article about them here soon! 

Make sure you’re subscribed to my newsletter and I’ll let you know when it’s out!

How I can help you 

Are you a founder, executive, or manager? I’d love to support your professional growth. 

Here are three ways: 

1. Connect on LinkedIn, X, and Instagram – join 150,000 people where I’m posting simple tips about leadership and startups every day.

2. Subscribe to my free newsletter – every week I do a deep dive on management, operations & startup topics that’s helping tens of thousands become better leaders.
   
3. Join Highland – my private community and leadership program for badass founders and CEOs. Scale into the tens of millions & beyond with us.