The Dark Side Of OKRs (And Why We Should Care)

Written by
- 6 min read

If we could tag one apocalyptic rider for adaptive organizations, it would be "traditional performance management." It is old-fashioned performance management that keeps us in a world of humans as resources, as command-and-control takers, with rigid top-down planning, and solid prevention of curious and exploratively-minded cooperation. Its logic is plan – do – check – act.

article photo

A logic, where individual goals are defined top-down from corporate objectives, any doing is checked against this goal accomplishment in the form of "feedback," and carrots are used to enable "pigeon training."

Most of us will agree that all of this is true for old-school management by objectives, but do we realize that some of the new co-creation tools – such as OKRs (Objectives and Key Results) – are maybe not much better?

At its core, management by objectives fulfills three different management functions where goals are to: a) coordinate and align individual and team efforts with those of the organization, b) channel information on those aspects most important from a corporate perspective, and c) create a motivating push.

Organizations that opt for adaptiveness, agility, and resilience have started to reform this goal process to enable decentral coordination and learning by allowing for much more participation, bottom-up input, and co-creation, thereby relinquishing the information monopoly.

For instance, so-called "FAST goals" are Frequently discussed (quicker-paced and more bottom-up), Ambitious (difficult, but not impossible to reach), Specific (concretized by measurable milestones), and Transparent (shared within and across teams).

Hence, these goals take local knowledge onboard and make sure that coordination is not dependent on hierarchy. The same logic applies to OKRs.

Better? Yes, but not all good!

Because irrespective of whether FAST goals or OKRs are applied – the spirit of pigeon training is still present as both bank on "make your goals specific", "be overambitious", and "show everyone what you have accomplished anytime".

Thus, these goals should come with a warning: "May prevent human ingenuity and co-creation to take place."

To be fair, they have a mighty motivation theory on their side when doing this: the goal-setting approach.

This theory has sparked a lot of what we take for granted in business but also coaching, namely: (1) goals are needed, (2) specific goals are better than "do your best goals" as these provide direction and a clear focus, and (3) ambitious goals exert a stronger motivating push than realistic goals.

Yet, in recent years, several studies have shown that the very success factors of such goals carry downsides, which are particularly problematic for agile organizations and co-creation. These are rooted in largely ignored psychological side-effects, which should be taken into account when working with goals.

Specific goals can make you blind

Specific goals give a focus but at the same time can also act as blinders and thereby prevent out-of-the-box thinking. Do you know the "invisible gorilla experiment"?

In this experiment, participants were exposed to a short video with players passing a basketball. They then got the specific goal to count how many times the ball was passed.

This simple instruction prevented half of the participants from spotting a giant gorilla passing by the ballplayers – it was as if the gorilla was invisible. Why? Because the attention was focused squarely on the task.

Specific goals thus can undermine mindfulness and spotting the unfamiliar, unplanned event. Therefore, it turns out that for agile and creative work that is to respond to changing environments and enable exploration, some amount of goal fuzziness can be helpful.

And by the way – the right amount of fuzziness needs to be teased out by experimentation as there is no handy formula for this.

Overambitiousness creates unwanted overheating

Besides, it is almost impossible to determine the right amount of ambitiousness.

Sure ambitious goals trigger "fire" or, more prosaic, a strong will to succeed. But what has become clear, too, and should not be ignored: overambitious goals – as argued for in FAST and OKRs - carry a lot of risks.

First: if goals are too ambitious and humans never stand a chance to reach these goals, their self-efficacy can deteriorate. However, self-efficacy is a needed resource if we want humans that have a curious, open, and pro-active mindset.

Second: for overconfident humans, on the other hand, ambitious goals create such a will that every means to attain the goals are used. Hence, overambitious goals can also evoke gambling behavior.

Third: overambitious goals – particularly when enacted as "preventing sandbagging" as a standard instrument, can lead to emotional depletion, even to the degree that out of this depleted state, employees start to show unethical behaviors . Hence the manipulation of "ambition" can backfire – enormously.

Transparency meets privacy needs

A further challenge comes with the transparency principle. In its original design, both OKRs and FAST goals bank on individual transparency – everybody knows everybody's goals as well as goal achievements. Yet here, too, psychological research shows some downsides.

First, the balance between transparency and privacy is a delicate one. Too much exposure can backfire as this can feel like a panopticon, a hardly visible but strongly felt control. Consequently, team members start to withhold information or dress-up their image rather than sharing freely and commenting helpfully on the collective work process .

Second, the so-called effect of mere presence triggers either social facilitation or inhibition. Imagine your playing the guitar either alone or in front of an audience. Sometimes, others' "mere" presence leads to a performance never achieved before, but sometimes, felt pressure and anxiety make free and creative play impossible.

The bad news for knowledge and creative work is that the decisive factor, whether performance increases or decreases with others watching is task complexity and difficulty. The more challenging and more complex a task, the more likely fear kicks in, and performance falters .

And now? Unsmart your goals!

Where does this leave us?

Clear goals are still needed to enable thriving and making progress. But these goals are better co-created and defined on the team level: in this way, teams can define their own criteria of success, while for team-members, some room for maneuver is left.

Also, ungoaled time - free time - enables the needed incubation for creativity and innovation.

Besides, moonshot goals create inspiration – but these ideally come in the form of a long-term "Everest goal" – directed toward abundance and the greater good in mind and not toward specific overambitious output targets.

And finally, goal transparency can be an information sharing and co-creation booster, but not without leaving room for experimentation in privacy.

Or, as an alternative summary: Just because Google made it famous does not mean we should embrace an instrument as "best practice." Sometimes unsmarter, less engineered "nonfoci Rs" might work better.

This is a guest post by Antoinette Weibel, professor for HRM (and secretly also for leadership) at the University of St. Gallen, and Meike Wiemann, a senior researcher, project manager and lecturer at the University of St.Gallen. For more information on Antoinette and Meike, check out their rebel pages: Antoinette's rebel page and
Meike's rebel page.

Subscribe to our newsletter


leave a reply

Replies (21)



Thanks for sharing these thoughts! Here is another intresting read around this topic:

| | 0 | Flag


Thanks a lot Antoinette for sharing this great and very helpful "package insert" on OKRs.
I couldn't agree more that it is very dangerous and misleading to see OKRs as the next silver bullet for easy personal and organizational development. To lead people and organizations built on trust is intense work and there is no silver bullet. This is especially important to keep in mind when approaches become trendy and main stream. At the end OKRs are a tool like many others and there is a good chance to hurt yourself with all those tools.
I nevertheless find OKRs helpful to trigger intensive discussions and to balance all those risks that are mentioned in the article like overheating, blind goal setting or violation of privacy. The vertical alignment with the purpose, values, vision can be also helpful to live and not only define the foundation of a team or organization.
Without the necessary exchange, alignment and balance OKRs will become the next bureaucratic overhead and soon after another unused gadget in the agile toolbox. Or to quote Grady Booch who famously said "A fool with a tool is still a fool."


| | 2 | Flag


Thank you all for these great comments. To me you have come up with further learnings here. Most importantly - as someone mentioned here so aptly -
"a fool with a tool is still a fool" - hence you all have found reflected, context-sensitive ways to work with goals. That to me is the essence - we should not adapt management tools as best practices but should try to be critical and reflective, customize, experiment and stay curious whether there are new and better ways to do things. After all we should not only "manage innovation" but also innovate management.

| | 7 | Flag


A very interesting analysis. It’s helpful to read how we can avoid blindly applying current practices like OKRs which have become a corporate norm.
The one thing that concerns me is that all of these shortcomings have been known in social- & organizational psychology for a while, and yet I don’t see dialogue or education which would foster critical thinking when adjusting to new agile - gig - teal - google /tech determined work realities. The creative tension this nurtures is that important principles for organizing human action like goal setting or providing structure are undermined by the pressure to adjust to new expected corporate realities (“structure equals burocracy equals old ways of doing”), while the real conversation should be around the both/and, the right balance and situational adjusting of our activities through critical & systemic thinking, reflection and dialogue).

| | 3 | Flag
Alice Patel

Alice Patel

I greatly enjoyed your article and how you've challenged common assumptions about human behavior in relation to goal-setting. Your last paragraph about privacy and transparency made me recall one of my favorite psychological conundrums...leaders who shared openly were often labeled hypocritical. You can read more here. :)

| | 2 | Flag
Ian Harvey

Ian Harvey

Nice article Antoinette. It is rare to read something so thoughtful on OKR/goal setting. In my experience OKR implementation often fail because of context. They are a small part of a big picture and organisations hope they’ll fix that context. It’s not realistic. Your article is helps me think of the challenges beyond that.

I agree that OKRs can easily fall into the same traps as more traditional objective setting approaches. In my experience OKRs work best where there is a strong empowerment and coaching culture. This helps create a situation where people can discuss, learn and grow with their manager (and others) and be aware where the ‘system’ is inhibiting their growth and thinking.

In terms of space. Also important is that not everything you do is an OKR. I was working with a client last week where they have a pressing goal this quarter. They also need to keep discovering for future quarters. (Think stages of the double or [triple]( diamond). We left next quarter out of the OKRs, but the UX designer and the product manager are still doing discovery work for that.

I still find that getting teams (and organisations) to focus is a bigger challenge, but yes for more mature places you can risk narrowing their perspective too much.

I fully agree that the level of ambition is difficult to get right. I don’t believe there is a set formula to resolve this. Being mindful of the kind of behaviour you mention is a good mitigating action.

Finally, your comment about Google is spot on. OKR is a framework that exists in a context.
1. No two contexts are the same, so it has to be adapted.
2. Thing big, start small, learn fast. You need to iterate to a place where OKR works for you and be mindful of the concern you and others have highlighted.

Love to connect with anybody enthused by the article - @enrvuk

Thanks again!

| | 5 | Flag


Another “dark side” of OKRs is that there is no structured guidance on how to apply them. There is a huge variation in how they are written and how useful they are. The definitions of and differences between “objectives” and “key results” is unclear and inconsistently interpreted and applied. Some treat key results as measures or metrics while others treat them as actions. Another issue is their inability to link and align with the corporate strategy. Despite the thousands of articles, books, blogs, videos and so on there is no instructional process to build good ones.

| | 2 | Flag
Ian Harvey

Ian Harvey

Hi Jenny, I do understand why people struggle with OKRs. There is a lot of bad advice out there and unfortunately the most famous book, Measure What Matters is full of terrible examples.

As someone who has worked with a lot of organisations and teams, I do know what works.

I'd love to hear your views and perhaps write an article that provides the guidance you (and others) seek.

You can connect with me on twitter @enrvuk

| | 7 | Flag


Jenny, I really do take your point. And the lack of clarity and multitude of ideas, is also part of so much of leadership, teams, how to succeed, etc!
My suggestion, is to first understand what is your underlying paradigm of focus. Are you looking at an organisation as a machine, that needs constant adjusting and lubricating? (most of the target based literature)
Or are you looking at an organisation driven by the customer, and operating end to end?
I tend to find it most helpful to focus on individual services, and start with the real needs of customers, and what matters to them. Then take it from there...

| | 1 | Flag
john mortimer

john mortimer

I thought that readers might be interested in an example of new ways of developing measures, that move away from the traditional approach that this article highlights.

| | 1 | Flag
Flags are private, only visible to forum moderators. Be specific: "It's spam/off-topic/inappropriate because..."

Leave a reply

This field is required. Only use letters, digits, spaces and minus characters.
This field is required.
This field is required.
We only use your email for spam detection purposes.