I've been pondering a question recently: how to do everything well, how to unify goals and indicators?
In response to this question, I would like to give two examples:
Case 1:
At the end of the 18th century, the British government decided to distribute criminals to Australia, and hired private boat owners to transport the prisoners. According to the actual number of prisoners on the ship, the government paid the owner a fee.
The British government soon found that the average death rate of prisoners was as high as 12%, and the worst was as high as 37%. The government has tried many methods, such as sending supervisors, sending doctors, regulating the living standards of prisoners, and educating ship owners, but none of them can solve the problem of mortality.
In the end, the British government was forced to change a set of payment standards-paying according to the number of prisoners ashore in Australia, and the prisoner mortality rate soon dropped to below 5%.
Conclusion: The goal is of course right, but the metrics are wrong and things will screw up.
Case two:
Since the resumption of the college entrance examination in 1977, thousands of provincial, municipal, and county-level college entrance examination champions have appeared in China, but now it seems that among the "college entrance examination champions" who have truly entered the social elite are rare, and most of them have become mediocre. generation.
A provincial college entrance examination champion in the late 1980s once said, "I suddenly found out that all my efforts in my life were all about accepting passive examination instructions and assessment instructions, all for the sake of the people around me. I'm satisfied. I'm trembling to fit myself into a pattern of behavior set by others..."
Conclusion: The indicator may be fine, but the target is wrong, and the direction will go wrong.
The government, enterprises and even life are all the same, and there must be good goal management - indicators and goals should be coordinated and unified.
The concept of "management by objectives" was invented by Peter Drucker. There are two methods in the specific operation of performance management: one is to convert goals into various performance indicators through KPI (key performance indicators); and key results), turn the goal into the accumulation of each small goal.
What is the difference between KPIs and OKRs?
In my opinion, KPI performance management is like "programmers write code", how the system works depends on what you write; OKR performance management is like "AI (artificial intelligence) engineers design algorithms", the system can be based on data, Scenes learn and evolve automatically.
At present, it seems that Google, Intel, Facebook, Pinduoduo, and Toutiao all rely on the OKR algorithm to think invincible.
So, are KPIs and OKRs really competing? Here, I would like to systematically elaborate my views based on three aspects:
Why do you say that the OKR management method is to solve "high efficiency ineffectiveness"?
What are the most important principles of goal design?
How to comprehensively use KPI and OKR management methods to comprehensively improve the effect of work?
Peter Drucker famously said: "Some people have limited mobility and have very little space for movement, but instead do a lot of things. Most people are mobile and travel around the world, but they actually do very little in a year."
Therefore, I believe that "effect is king" and oppose "high efficiency without effect".
1. Why do you say that the OKR management method is to solve the "ineffectiveness of high efficiency"?
If only one dimension of "efficiency" is considered, no one can beat a computer. So, how is the computer's "super work efficiency" achieved?
Professor Shi Yuanchun of Tsinghua University has been doing computer research for 30 years, and the most profound part is "Addressing".
For example: You send an instruction Latest Mailing Database to the AI robot: Help me get the white candy box on the coffee table in the living room. The computer will quickly find the room position corresponding to the living room, the furniture position corresponding to the coffee table, and the space position corresponding to the white candy box, and then determine a shortest path to achieve the goal according to "action priority sorting" and "unified rules (algorithms)" .
Note: There are three key elements - target location, action prioritization, and uniform rules.
The negative factors that affect work efficiency are basically also derived from these elements.
When doing the same thing, different people in the team have different estimates of the target position, and different people’s resources, abilities and selfish reasons also have very different priorities for actions, and it is difficult to implement unified action rules and principles.
Therefore, the company's goal management often emphasizes "digital quantification", that is, KPI, which has three advantages:
If you just define the goal, for example, your goal is to "improve customer satisfaction", there is too much room for ambiguity, and without using numbers, it is difficult to know how far you are from the goal;
"Digital quantification" is to give an objective evaluation standard. For example, if you say that the company's business structure is optimized and its business model evolves, what does "optimization" and "evolution" mean? Different people may understand it completely differently;
The biggest role of "quantitative goals" is to build consensus. Only with a certain number will there be consensus in everyone's understanding, and then a consensus on the goal will be determined.
But the worst thing about management by objectives is "highly efficient ineffectiveness" or even "highly efficient negative energy".
There is a second-rate mobile phone company in China, and the target management of product design is very "clear". For example: To surpass Apple's iPhone, the phone needs to be thinner, faster, and the material needs to be cooler. However, the thinner the mobile phone, the more difficult it is to dissipate heat, the faster the speed and the more power consumption, and the choice of appearance materials is subject to functional devices. As a result, what you think is an excellent product is often not as effective