How to cultivate the data thinking favored by big companies

Data thinking is not simply a data analysis skill, but a way of thinking based on data. This article uses a vivid little test to explain the four core principles of data thinking in an easy-to-understand way for everyone to learn.

Data thinking is a key point in many large companies' interviews and is also something that many company leaders talk about. But what is data thinking? Since it is called "thinking", it refers to a way of thinking, rather than mechanically reciting a piece of code or formula.

There are many questions that can test your data thinking.

Today, I will take you through a simple topic to experience:

Data thinking test

For example, one day, an acquaintance of yours asks you with a sad face: "I lost 1 million in stock trading, what should I do?" How would you respond?

think

Test

one

point

bell

Data thinking first

Note! "Loss" is not a data indicator, but a qualitative description.

According to different stockholders, there are at least six situations in which "loss" occurs:

• Case 1: I saw a good stock, hesitated, and didn’t buy it, and ended up losing $1 million
• Situation 2: I bought a good stock and sold it too early! As a result, I lost $1 million
• Situation 3: I bought a good stock, but forgot to sell it at a high point. Now it has fallen, and I lost 1 million yuan.
• Situation 4: I bought a bad stock, but I kept silent while others made money, and I lost $1 million
• Situation 5: I bought a good stock, but it started to fall right after I bought it, and I lost 1 million yuan

This is the first rule of data thinking: use data to quantify and describe the problem.

This seems simple, but not everyone can do it.

For example, when faced with this problem, many people’s first reaction is:

1. Which stock did you buy? Let me see if this stock is okay.
2. How much money do you have on hand? Is there room to increase or decrease your position?
3. How much assets do you have in total? If you lose 1 million, can you lose your pants?

Interestingly, these are also the three typical thinking patterns:

• The discussion of stocks is a typical product mindset, focusing on product quality/performance/pros and cons
• Those who discuss adding positions are typical salesmen. The important thing is to do it! Do it! Do it!
• The discussion of operation is a typical operation thinking. Let's see the operation method and try it again.

If you put these four types of people on a table, you will find that they have very distinct characteristics (as shown below):

Data Thinking Rule 2

The second rule of data thinking: base your judgment criteria on data comparison. For example, “I lost 1 million yuan in stock trading”. What does this 1 million yuan mean to “me”? There are many situations: (as shown below)

In addition, it may not be enough to just look at the total assets, because a large part of the total assets may be real estate/cars/fixed deposits/precious jewelry that cannot be quickly converted into cash. Many people do not have that much liquid funds on hand, and if they lose 1 million, they may have lost all their living expenses, so we also have to look at the asset structure (as shown below):

This one seems simple, but it is actually very difficult to do.

People always instinctively bring their own situation into play and ignore the situation of the person asking the question.

This is the greatest use of data thinking: through quantitative and detailed data, thinking can be based on facts, so as to find more effective solutions to problems.

People often say: Tailor your clothes according to your body shape and eat according to the food you eat. This is actually what it means.

Data Thinking Rule 3

The third rule of data thinking: find solutions based on data differences.

For example, we have determined that this guy has a net loss of 1 million, and we can also ask:

• Is it because the current environment is not good, and whoever speculates will lose money?
• Is it that this guy just doesn't know how to trade stocks? Nine out of ten times he loses money?
• Is it just a stock that is overweight? Is there still a chance of a comeback?

Attention! Different data will lead to different judgments:

• This guy will lose money nine times out of ten, so there is a high probability that he is a rookie, so don't hire him in the future
• This guy just missed this time, so he might not perform well, and there is still a chance for a comeback.

Data thinking cannot guarantee 100% correct judgment, but it can avoid wrong thinking direction with a high probability.

Data Thinking Rule 4

The fourth rule of data thinking: come from reality and go to reality. Data thinking is not "data-only theory". On the contrary, people who master data thinking will discover more facts and truths through abnormal data performance.

For example, this guy is obviously worth a lot of money and he only lost a little bit, so why is he still frowning?

It is very likely that what he is sad about is not the property, but:

• Aroused self-doubt
• Afraid of being looked down upon by friends
• Afraid of being scolded by my wife!

At this point, don’t dwell on the numbers themselves, and quickly communicate with him to dig out deeper issues.

In short, data thinking is not a spell like "Avatar Krafla" that works as soon as you chant it. Instead, it is through careful and meticulous combing, drawing conclusions from clues one by one, and finally piecing together a complete picture.

For example, if you sort out all the clues for the seemingly simple question at the beginning, the logic may be as follows:

Tips for practicing data thinking

The best way to exercise data thinking is not to read books like "Underlying Logic" or "Core Thinking", but to try to ask more questions and work harder to find data in your daily work and life.

• When you encounter data, ask more questions: How was the data collected? What was the collection time range? What was the calculation formula? Instead of ignoring the source of the data and saying: I just feel that this data is wrong because I didn’t see it…
• When you are faced with judgments such as size, quantity, speed, quality, etc., ask more questions: What is the indicator? What is the standard? Instead of ignoring the judgment criteria and just saying: I think it is right! / No!
• When faced with a decision, ask more questions: What do I base my decision on? How do I measure the result? To what extent do I want to achieve it? If the size of the conditional data changes, will I change my decision? Instead of just saying: I have been in this industry for ten years, what I say is right, let's do it this way.

Although you may not find the answer, with more practice, you will develop a good habit of thinking about data when encountering problems.

When we have data to support our decisions, we can make better judgments.

Of course, the quiz in this article is only suitable for use as a joke.