OKRs in China

I finally went to China. Though I’ve had dozens of conversations with Chinese professionals, this was my first trip to China. My work with OKRs is enabling me to see the world. I now have WeChat installed on my iPhone. While China is a very big place, the OKRs community is pretty well connected. I met with the team translating Radical Focus into Chinese and learned that the OKRs book I co-authored with Paul Niven is coming out in Chinese in May! In addition to eating different foods like jellyfish and chicken feet, I learned a bit about how OKRs in China is a big deal right now.

I met with representatives from over 40 companies during my travels to Beijing, Shanghai, and Shenzhen. I sensed BIG potential for OKRs in China. OKRs are especially valuable for rapidly growing companies. Think of how Google used OKRs in its early days. I remember my mentor, Jeff Walker, sharing how he used OKRs to help manage Oracle’s growth from 200 to 12,000 employees in just over 5 years. I met with one Chinese company that had plans to grow from 300 employees to 2,000 by end of year. This is the kind of growth that requires a clear, standardized method for communicating goals throughout the organization.

Several companies I met with in China have used OKRs for years. Several large companies recently completed successful 1-year pilots and are planning to expand their OKRs projects. While I’ve got a limited sample size, it’s clear that the use of OKRs is contributing to a positive impact on employee engagement across organizations of all sizes.

At all four of my 1-day training workshops in China, the same types of questions kept coming up. In this post, I share brief answers to the top 4 questions.

Question 1

What’s the relationship between OKR and performance reviews/appraisal? Please give an example how a company did an appraisal and incentive which failed with OKRs and another example illustrating how best to implement OKRs with performance appraisal.

Analysis

There are two very wrong answers to this question. The first wrong answer is “OKRs are the system we use for performance reviews.” The equally incorrect answer is “OKRs are not at all related to performance reviews and are totally separate.” The correct answer is that OKRs are distinct from your performance review, but are clearly related. Do not replace your performance review system with OKRs.

Example of OKRs gone wrong

First, let’s look at an example illustrating why OKRs should not be the primary system for performance appraisal. Here’s an excerpt from an article detailing how Yahoo used OKRs ineffectively:

Starting in September 2012, Mayer introduced a clone of OKRs to Yahoo. She called them Quarterly Performance Reviews, or QPRs. Employees from Mayer’s direct reports on down would get a score every quarter, from one to five. A one meant the employee consistently “misses” goals, a two meant the employee “occasionally misses,” a three, “achieves,” a four, “exceeds,” and a five, “greatly exceeds.”

Mayer rolled out her cost-cutting, talent-improving plan in stages. First, she introduced the concept of goals to employees.

Then Mayer announced a “target distribution” for the company.

In effect, a target distribution meant Mayer wanted managers to put a certain percentage of the employees they managed in each of the five buckets. Ten per cent would go into “greatly exceeds,” 25% in “exceeds,” 50 percent into “achieves,” 10 percent into “occasionally misses,” and 5% into “misses.”

Then Mayer rolled out new policies wherein employee eligibility for bonuses, promotions, and transfers within the company would be based on their average score for the past three quarters. Employees with low enough scores would be asked to leave the company.

Over her first year, the plan seemed to work — or at least accomplish Mayer’s two goals for it. Mayer felt Yahoo’s talent level was rising. And, without going through any morale-draining layoffs, Mayer was able to move 600 low-performing employees out of the company.

Unfortunately for Mayer, the plan also produced a few unintended consequences.

Click here to see full article and learn about the unintended consequences

Example of OKRs working well with appraisals

Most organizations get this right. They don’t take the path Yahoo took in 2012. Instead, they retain their current Performance Appraisal processes. They introduce a separate, lightweight OKRs Review. The OKRs review is a quarterly process that everyone can participate in. The actual scores of key results do not determine bonuses or measure your performance. Rather, the qualitative data emerging from OKRs reviews are one of the ingredients you bring to your annual performance review discussions as a contributing factor.

Example of an “OKRs Review”

Each team member writes up the answers to 3 questions and meets with their manager for a 30-minute 1:1 discussion each quarter. The 3 questions are:

  1. How did the use of OKRs enable you to achieve better results?
  2. What Key Result did you contribute to the most and how?
  3. What is the one thing you can do to improve your use of OKRs next Quarter?

Question 2

Assuming Key Performance Indicators (KPIs) are used to calculate bonuses, what’s the top differences between KPI and OKRs?

Quite a few organizations with whom I discussed OKRs in China have a system whereby KPIs are used to calculate bonuses. In my experience, the usage of the term “KPIs” is quite broad. I helped several companies identify company-level KPIs. Leadership presented these KPIs on a dashboard to the board or a parent company. In such cases, KPIs were not the basis for calculating bonuses. So, KPIs do not always drive bonus calculations.  In fact, many highly paid engineers and product designers have KPIs for their team, but receive no incentive compensation based on performance. But, in China, this question was referring to “KPI” as a performance measure used to determine incentive compensation. Therefore, I’m calling this type of KPI – “Compensation-KPI” so we know we’re talking about KPIs that are used to calculate compensation. Here’s a table comparing and contrasting Key Results with Compensation-KPIs.

Distinguishing factor Key Result Compensation-KPI
Achievement tied directly to compensation No Yes, by definition
Made public for everyone to see Yes No
Defined in context of an Objective Always Rarely
Focused on status quo / current performance levels Rarely Often
Focused on improvement, innovation Often Sometimes
Intended to increase alignment across teams Yes No
Controllable by your team Somewhat Yes
Originates bottom-up, not by “corporate mandate” Mostly Rarely

For more on how OKRs and KPIs (not compensation-KPIs) work together, check out my quora post.

OKRs and KPIs: Varies across departments

It’s also worth noting that the use of OKRs and KPIs may vary across departments within a single company. For example, sales teams almost always have a form of KPIs for calculating their bonuses. However, very few sales teams embrace OKRs. They feel their KPI system already serves their needs and the OKRs system is redundant, a waste of time. Engineering teams often do not have KPIs for calculating a bonus, but engineers tend to embrace the OKRs framework. Our friends in marketing may have KPIs for calculating bonuses and embrace OKRs as well. I guess marketing teams just like as many metrics as possible! Finally, a legal team may have neither OKRs nor KPIs since they do not feel that planning is viable given the reactive nature of their work.

Question 3

What role should the HR team play when implementing OKRs?

Analysis

Some HR leaders in China expressed their confusion regarding ownership of the OKRs project within their organization. After all, HR teams often own traditional goal setting frameworks. However, OKRs should not be owned by a single team. As noted in our book, when a single department “owns the OKRs project,” your OKRs project is at risk of becoming “just another HR thing” or, a “Finance initiative.”

Recommendation

OKRs should be led by the executive team. In addition, train a distributed set of “champions” or “ambassadors” from various departments from various teams within your organization. These OKRs champions complete a full day of OKRs training. They serve as internal coaches to ensure teams draft and refine OKRs effectively.

Question 4

Based on your past observation, what’s the key factors in the success of implementation of OKR in a company?

I admit my answer to this question is a bit lazy. So, I advise checking out my prior post, Getting Started with OKRs for details. In addition, please look for more on the OKRs adoption framework I’m working on with Paul Niven. At the moment, here are the 9 key factors:

  1. Start at the top and determine why OKRs
  2. Create clearly defined roles for the program
  3. Agree on deployment parameters
  4. Provide OKRs training
  5. Communicate OKRs
  6. Create OKRs with a coach
  7. Own key results, mostly bottom-up
  8. Conduct a mid-quarter review
  9. Document and share what you’ve learned

Conclusion

This post provides short answers to the 4 most commonly asked questions about OKRs in China. While I’d like to answer more in this post, it’s getting long! I believe that if several well-respected companies adopt OKRs in China, it will lead to a rapid OKRs adoption, similar to the adoption trend we saw in 2013 when Rick Klau released the Google Ventures OKRs workshop video in 2013.

As a next step, please add more questions or comments with your reactions to this post. Let’s get a discussion going about OKRs in China!

Related Posts

Comments (2)

In the Yahoo! example, what did you see as the key challenge in management’s decision(s)? Was it to force rank team performance into a targeted distribution, where some portion were winners and others losers, or to use the output of quarterly OKRs in performance reviews, or both these practices? I ask because I’ve seen a number of companies do a flavor of both these things – and though these companies that come to mind haven’t been market-leading, it does seem to happen in a number or organizations. I can’t agree more about the stack – I’ve been asked to do the same thing with my own teams, and in the situation where all delivered on target or exceeded, we still had to deliver results with some members “occasionally missing”… it was incredibly disheartening as a manager who was passionate about staff development. Regarding the alignment of how to leverage OKRs with reviews, you give some nice examples of how organizations should consider tying OKSs into quarterly performance reviews, thank you. I’d like to see these adopted at companies I’ve worked with in the past – I plan to reach out and share these examples with them.

Great comment!

The Yahoo example reflects the need to use OKRs as a way of identifying poor performers in order to fire them. And it worked the first iteration. The problem was the continued use of OKRs to rank employees after that first wave of layoffs.

For some reason your comment makes me want to share a learning that may be pretty obvious from my experience at the recent 2017 Goal Summit in San Francisco: we should be asking for feedback, not giving. it. Does that make sense at all? let me know if want more details on this…

Leave a comment