How to Use OKR During the COVID-19 Outbreak
Your OKRs are, of course, a small problem considering the massive potential impact of the COVID-19 (coronavirus) pandemic. But many people have asked me for advice on how to use OKR during the crazy times we are living in, and I felt it was a small thing that I could do to help others, so I decided to write this article to answer the most common concerns.
As I work with companies across the globe, I see segments and geographies that have suffered different levels of impact from the outbreak. Some are seeing a total disruption of their business models, while for others, it’s too early to tell. Most are dealing with unprecedented levels of uncertainty. This article is not supposed to provide a one-size-fits-all approach, but a set of guidelines and tools that you can adapt to your context.
There is a lot to unpack in this post. You don’t have to use all the techniques I mention here. The idea is to show you different tools and give you pointers to learn more about them if you want to.
Now, before talking about OKR itself, we need to talk about people.
Protect your people. Provide as much safety as you can.
As the outbreak spread, most organizations rightly focused on measures to provide health safety for their employees, such as working from home. Now it is time for a more comprehensive approach to safety.
Given all that is going on, even people who are used to working from home won’t be able to perform normally. They have children and family at home. They fear for their loved ones and even their jobs. For people who are new to remote work, it’s even harder as they have to adapt to new tools and routines.
Organizations need to do what they can to make people feel safer and to reduce their cognitive overload. Managers must help employees by providing better clarity, better coaching, and better alignment. Employees may also need help in getting better team discipline around tracking results with distributed teams. Leaders must ask themselves: how can we provide the conditions our employees need to work in this environment?
Companies need to increase communication, balancing the needs of the business with expectation setting and morale building, so employees know that their well-being is top of mind. This video from Arne Sorenson, President and CEO of Marriott International, is an excellent example:
A message to Marriott International associates from President and CEO Arne Sorenson. pic.twitter.com/OwsF14TZgb— Marriott International (@MarriottIntl) March 19, 2020
For many companies, one of the main risks is that managers will fall back to old management styles: barking orders top-down, command-and-control, lots of controls and reports, focused on compliance and fear.
A couple of weeks ago, I got an email from a company that wanted to adopt OKR because they believed their work from home initiatives required a “more robust performance management system.” To me, it sounded like the old carrot and stick approach: “We don’t trust our employees, so we need to punish the ones that don’t behave.” This approach is precisely the opposite of what organizations need to do. OKR should be decoupled from compensation, and the last thing people need right now is for their employer to threaten them with punishment.
Smart organizations are trying different approaches to reduce the uncertainty around variable pay and performance reviews during the pandemic. Google is delaying all of its employees’ performance reviews and promotions for six months because of the coronavirus pandemic, and will instead promote twice as many people in November and backdate raises to August 1. Facebook, on the other hand, decided to give all of its employees positive performance reviews for the first half of 2020, ensuring that all of them receive their biannual bonuses. At the same time, a payments company I worked with is paying full quota achievement for all their sales reps as long as they remain in lockdown.
Distributed work also means organizations have to create more open environments so employees can speak up, ask for help, or offer an idea.
Distributed work also means organizations have to deliberately and proactively create more open environments so employees can speak up, ask for help, or offer an idea. This crucial concept is called psychological safety: Can employees take risks on this team without feeling insecure or embarrassed? Can they overcome fear and reluctance to speak up with potentially controversial ideas or questions?
Google found that psychological safety is the leading dynamic that sets successful teams apart from other teams in normal settings. It is even more important with distributed work since “we can’t just overhear what’s happening in the next cubicle,” says Harvard professor Amy Edmondson, who did the original research on the topic.
Now that we’ve talked about people, let’s talk about a few concepts and techniques that can help you set and adjust your OKRs in times of great uncertainty.
Outcome-based planning and experimentation
If you have read any of my material, you know that OKR is not a tool to manage activities. Instead, OKR is a tool to help companies adopt outcome-based planning, an approach where organizations plan around the outcomes they want to achieve instead of committing to deliverables that may or may not work. We have lots of evidence from different companies that show that most ideas fail and that even experts will misjudge what will work or not. Using OKR as an activity management tool is a huge waste of time and effort, regardless of all the bad OKR advice out there.
If your organization is still using “milestone Key Results” instead of outcomes, that is the very first thing you need to change. Key Results should stand the “So What?” test—they have to deliver value and benefits to your customers, your company, or your employees. Good Key Results make a difference.
For those who believed they could predict the future and tell which ideas would work or not, I truly hope that the current crisis serves as a wake-up call. We live in an unpredictable world, and we need to embrace that.
Outcome-based planning is the stepping stone of a pivotal change, as it enables organizations to create networks of small, empowered teams that are at the heart of Silicon Valley’s success. Instead of giving employees projects to deliver, outcome-based planning gives them problems to solve and the autonomy to experiment with different ideas until they find something that moves the needle.
Outcome-based planning reduces the uncertainty on your OKRs and moves it to the experiment level, where it should be. Adopting experimentation is the primary technique to deal with uncertainty, and the outbreak does not change that.
Outcome-based planning reduces the uncertainty on your OKRs and moves it to the experiment level, where it should be.
Outcome-based planning and experiments go hand in hand, and organizations that want to reduce risk have to teach employees how to run experiments. Luckily, three authors recently published important books on the topic: David Bland, Ron Kohavi, and Stefan Thomke.
Two-way door decisions
To deal with unprecedented levels of uncertainty, organizations need to change their attitude towards failure. As Dr. Michael J Ryan from the WHO explained when talking about virus outbreaks, “If you need to be right before you move, you will never win. The problem with the society we have at the moment is that everyone is afraid of making a mistake, everyone is afraid of the consequence or error. But the greatest error is not to move. The greatest error is to be paralyzed by the fear of failure.”
The key to overcoming that fear is to understand that not all decisions are equal and that sometimes we should not worry as much about being wrong. Amazon divides decisions into two types:
- One-way door decisions: Decisions that are consequential and irreversible or nearly irreversible and should be made methodically, carefully, slowly, with great deliberation and consultation.
- Two-way door decisions: Decisions that are changeable, reversible. If your decision didn’t work out, you don’t have to live with the consequences for that long. You can reopen the door and go back through. Two-way door decisions can and should be made quickly by individuals or small teams.
Large organizations often treat all decisions as being irreversible one-way doors, leading to slow decision-making and risk aversion. Startups and high-performing teams, on the other hand, tend to look for ways to test ideas that are safe to fail, two-way decisions. “Experiments can turn one-way door decisions into two-way door decisions,” explains product discovery coach Teresa Torres.
How can you plan and set OKRs if everything seems to be changing so fast?
Discovery-driven planning is a tool created to help organizations plan high-potential projects whose prospects are uncertain at the start, and it can provide substantial help in setting OKRs in the current context. Discovery-driven planning acknowledges that at the start of a plan in uncertain scenarios, little is known and much is assumed. The tool helps companies make those assumptions explicit and then test them based on real evidence.
When conventional planning is used, assumptions underlying a plan are treated as facts—givens to be baked into the plan—rather than as best-guess estimates to be tested and questioned. Companies then forge ahead on the basis of those buried assumptions. In contrast, discovery-driven planning systematically converts assumptions into knowledge as time unfolds. When new data points are uncovered, they are incorporated into the evolving plan. The real potential of the plan is discovered as it develops—hence the term discovery-driven planning.
Discovery-driven planning influenced the Lean Startup movement and the whole experimentation culture we talked about above. The original version of the tool has five steps, but when setting OKRs, I suggest a slightly different approach:
Document your assumptions. Get everyone together who is working on that problem/opportunity and list all of the assumptions behind your market, your customers, revenue, etc. Don’t overthink it. In the beginning, stick to a few of the most critical assumptions and focus on the conversation.
For example, I recently worked with a company in Europe, and one of their assumptions was that the next quarter would be for damage control due to the pandemic and that the market would start to come back in the quarter after that. They set their OKRs based on that explicit assumption and will test it over time. If that assumption proves to be incorrect, they will adjust their OKRs accordingly.
Iterate your assumptions with your stakeholders. Whatever assumptions you develop, get feedback from your stakeholders, or your board. Let them help you with a reality check.
Set your OKRs based on your assumptions. Use your assumptions to guide your OKRs.
Based on what we learned since the last check-in, is there anything we should change in our strategy, our OKRs, or our overall approach? Should we redefine one of our assumptions?
Check your assumptions. Regularly review your assumptions based on evidence to see if they are holding true or need to be redefined. Start every OKR check-in with two questions: “Based on what we learned since the last check-in, is there anything we should change in our strategy, our OKRs, or our overall approach? Should we redefine one of our assumptions?”
Those questions can trigger important conversations. You may have to adjust a Key Result or even rewrite your OKRs from scratch. For example, as you learn more about the problem space, you may discover that you need a different Key Result, or that you need to measure more than one dimension of the problem by adding an additional Key Result. Or the company may revise its strategy and remove an OKR altogether.
OKRs should be written in pencil, not in stone.
OKRs should be written in pencil, not in stone. Organizations need to maintain lightweight, continuous conversations about their strategy—if the world changes or your understanding of the customer needs changes, your OKRs should reflect that. This does not mean you should always change your OKRs, it just means that you can. Even in this pandemic, if you are changing your OKRs every week you are doing it wrong.
Important: All changes in the OKRs have to be communicated to those involved and to anyone who will be affected by them in a lightweight but proper fashion.
Teams who are new to documenting assumptions should keep it simple. Teams who are more experienced can use tools such as assumptions mapping, business model canvas or value proposition canvas to help them better identify and test assumptions.
You can find the original article on discovery-driven planning here, and updates here and here. Steve Blank recently proposed a model based on an assessment, and although it is focused on helping cash-flow negative companies, it has some useful insights as well.
A useful tool that can be paired with discovery-driven planning to provide clear guidance for the teams when setting OKRs is the use of simple rules. Instead of shaping their strategies as complicated frameworks, companies can use simple rules of thumb that are specific and concrete.
For example, a logistics company created four simple rules for prioritizing capital spending. Any proposal, the rules said, should:
- Remove obstacles to growing revenues,
- Minimize up-front expenditure
- Provide benefits immediately (rather than paying off in the long term), and
- Reuse existing resources.
Similar rules can be used by any company that needs to strengthen its cash position during the pandemic. Another example of a simple rules that could aplly to OKR is “Focus on shared OKRs instead of functional ones.”
Simple rules align key decisions with corporate objectives and translate the broad priorities such as “cut costs” into clear guidelines that managers and employees understand and can act upon. The rules can help people avoid the paralysis that often strikes when they’re confronted with too many alternatives.
How can you set targets?
The next step is to talk about targets. How can you set targets when external factors have an overwhelming impact?
The trends for many metrics over the last few weeks are so wildly different than any other time that it makes it extremely difficult for teams to set targets that will engage them. Either they will set a number so high that it can’t be reached unless the economy fully recovers or so low that it can be achieved with any improvement of market conditions.
There are four techniques that you can use:
First, many of your Key Results won’t be “committed” anymore (what I call roofshots, Key Results where we expect to achieve 100% of the target). Given the current level of uncertainty, it will be hard to get teams to commit to a certain number. Instead, many of your Key Results will be “aspirational,” not in the sense of being very ambitious, but because they will be unpredictable.
Pro-tip: Whenever using aspirational targets, write (Aspirational) at the end of the Key Result so everyone understands it. You can also use the term moonshot if you prefer, although that is more associated with very ambitious targets, which may not be the case now.
Many of your Key Results will be “aspirational,” not in the sense of being very ambitious, but because they will be unpredictable.
Second, companies should evaluate the feasibility of using relative targets, where they compare themselves with peers in their industry, instead of absolute targets, which compare against their own previous performance—which in many cases will no longer be a relevant baseline.
For example, let’s say a product sees a 20% drop in revenue during the next quarter. Terrible, right? What if the competition sees a 40% drop in revenue, meaning that the product gained market share?
The use of relative targets is one of the principles of Beyond Budgeting, and, although they may be less precise than absolute numbers, relative targets can be extremely helpful during times of great uncertainty. Teams can use industry benchmarks and tools such as App Annie to help them set relative targets. This type of data is not always available, but it will be extremely useful if found.
Third, companies can set quarterly targets with monthly reviews. Instead of a full OKR setting cycle, teams can simply review their targets based on what they learned about their OKRs and assumptions.
Finally, teams can also adopt continuous stretching, an alternative technique to help teams set more ambitious targets over time. Instead of setting an ambitious target (a “stretch goal”) at the beginning of the quarter, teams can “raise the bar” and increase the target if they get too close to it. Goals have to take us out of our comfort zones, and if they are too easy, we can review them.
You can combine the techniques if you want to. Just remember to keep it simple.
Selecting your OKR cadence
With the outbreak, many organizations are wondering if they should temporarily focus only on the short term and adopt monthly or even bi-weekly OKRs. By using outcome-based planning, experiments, discovery-driven planning, and the target-setting techniques above, the need for shorter OKR cadences can be drastically reduced.
As you may know, OKR uses nested cadences to balance short-term and long-term goals and to force you to make the crucial distinction between things you want to achieve in the longer term and the things you need to achieve now.
Most companies I worked with set annual high-level OKRs for the company, quarterly OKRs for the teams, and have weekly check-ins for tracking results. Given the current situation, companies should review the model they are using:
First, most companies should evaluate the shift to six-month OKRs for the company (and large business units). Those are high-level OKRs focused on the business fundamentals and generating longer-term impact, and they are crucial to prevent teams from missing the big picture and focusing only on what they can accomplish in the short term.
Next, I recommend that most companies should maintain quarterly OKRs for the teams with monthly target reviews and continuous conversations about the underlying assumptions, as we have seen above. Most teams that want to set monthly OKRs are using OKR as a glorified to-do list. They should adopt outcome-based planning and experimentation instead.
Be careful. Adopting shorter cadences can drastically increase your planning overhead. Unless you have a streamlined process for setting OKRs in place, you will be spending more time setting OKRs than actually working.
Finally, if you are not doing weekly OKR check-ins for tracking your results, I strongly recommend that you do so. While the overall structure of the check-in remains the same, every check-in should start with a conversation about what you learned since the last check-in (see above).
The “four” horizons
How can you avoid being stuck on perpetual firefighting, always putting out fires without preparing for the future? A small tweak in a classic management tool can help provide the framework for prioritizing longer-term impact.
McKinsey’s Three Horizons Model became a foundation of innovation strategy as it allowed senior executives to visualize innovation happening in three time horizons:
- Horizon one represents those core businesses most readily identified with the company name and those that provide the greatest profits and cash flow. Here the focus is on improving performance to maximize the remaining value.
- Horizon two encompasses emerging opportunities, including rising entrepreneurial ventures likely to generate substantial profits in the future, but that could require considerable investment.
- Horizon three contains ideas for profitable growth down the road—for instance, small ventures such as research projects, pilot programs, or minority stakes in new businesses.
Each horizon requires different focus, management, tools, and goals. McKinsey suggested that to remain competitive in the long run, a company should allocate its research and development dollars and resources across all three horizons.
The original version of the model has been criticized as being too slow for today’s fast-paced innovations. However, it still provides an incredibly useful taxonomy to balance short and long-term investments.
By tweaking the model and including a Horizon zero focused on emergency or very short-term work dedicated to the pandemic, organizations can balance the resources devoted to damage control with the ones needed for the other Horizons (thanks to Rodrigo Toledo for the original Horizon zero concept):
Horizon zero can mean ensuring health safety for the employees, supporting customers in different ways, strengthening your cash position, providing the necessary tooling and infrastructure to work from home, or any other short-term needs created by the outbreak.
Different organizations will be able to dedicate more time and resources to longer-term planning than others, depending on their current situation. For example, a company whose business has been disrupted by the pandemic may have to dedicate 90–100% of their time to Horizon zero, while a company with a strong cash position and whose business model wasn’t as deeply affected by the outbreak may be in the opposite situation. In that scenario, the company could devote 10–20% of their time to Horizon zero and 80-90% for Horizons one, two, and three. A similar logic can apply to individual teams that may have to dedicate part of their work to Horizon zero needs.
Over time, organizations should be able to move away from the emergency response represented by Horizon zero and start investing more in the other Horizons.
How to come out stronger than before
In 2004, while referring to the increasing competition that America faces from rapidly rising education levels in other countries, Stanford economist Paul Romer said, “A crisis is a terrible thing to waste.”
As Don Sull, senior lecturer at the MIT Sloan School of Management explains, a crisis marks a break with the past, and people recognize that organizations cannot continue to do what they did in the past. The downturn lowers their resistance to change, cuts through complacency, and provides an external rationale to justify painful decisions that would appear extreme in better times. Finally, an economic crisis enables managers to make decisions that incur short-term financial pain for long-term gain, such as pruning products, “firing” unprofitable customers, or exiting money-losing businesses.
While organizations should not try to profit from the pandemic, there are a few things they can do to emerge stronger than before and create long-term value:
First, a crisis provides the ideal opportunity to force hard choices. For example, rather than spreading the pain of downsizing evenly, demanding an identical percentage cost reduction across all units regardless of their merits, a crisis can force leaders to make hard decisions to ensure the most promising opportunities get the resources they deserve.
While most executives believe they are reasonably good at focus, the truth is that they are often chasing ten times more opportunities than they should. Instead of 2–3 truly important things, they have at least 20–30. Smart leaders will use the crisis as an opportunity to drastically improve their focus.
Sull suggests that managers can ask themselves a series of questions to force hard choices: What initiatives, businesses, products, markets, and so on, have a call on our scarce resources? Can we rank order them in terms of value creation potential? Where should we draw the line that marks the truly critical from the nice to have?
Second, organizations can harness a downturn to renew a sense of urgency around their transformation programs, justifying unpopular decisions and overcoming complacency. For many organizations, now is the time to truly transform and embrace new ways of working.
A simple and powerful technique to help executives challenge the status quo is called the hotshot rule. Ask people to think of the most impressive business leader they know and then imagine that person taking over their company or job. What is the one thing that hotshot leader would do differently their first day on the job?
Viewing their companies from the hotshot’s perspective, flaws immediately become apparent. “They realize somebody who is awesome would find this unacceptable, and they take action on it,” explains Kat Cole the creator of the technique.
Third, companies can use the crisis to develop a continuous improvement culture which continuously reduces costs by identifying and eliminating activities or materials that do not add value for end users. To do that, managers should ask: What processes do we have in place to systematically identify and eliminate waste? Could we improve these procedures? Are there promising best practices in parts of our organization that we could disseminate more widely?
Finally, organizations can use the crisis to identify new opportunities as new customer needs arise, for example. As Teresa Torres recently tweeted, “If you are continuously looking for customer needs, pain points, the world just gave you a lot of new to add to your list. But that doesn’t necessarily mean they are higher priority than what you are working on now. Ask that question. But do so with the short & long view in mind.”
I hope this article can help you and your organization better handle everything that is going on in the world right now. If you have any questions or comments about this article, please send me a message at firstname.lastname@example.org.
This post is also available in: BR