Just a few days ago, I talked about big data in the WeChat group. What is big data? What business models can big data produce? Actually, I'm an amateur, and I have only a little knowledge of this problem. The purpose of my article today is not to tell you what big data is. What business models can big data produce? The purpose of this article is to share with you how to deduce ideas and disassemble thinking through the deduction of big data and business models!
Speaking of thinking, please allow me to pretend here. Although it is empty, I hope everyone can understand the intention!
Everything must be connected as long as it rises to the level of Tao. We often hear that there are many routines in operation. It's true. If there is no routine, it is the same to do other things, which means that everything has rules to follow. The deepest routine is "Tao" at best, and then "philosophy" (please note that I am talking about the deepest routine). Just as we can look at the problem from the four levels of Tao, Dharma, Technique and Apparatus, in fact, this is the routine of thinking. For another example, we often say that we should look at the essence through phenomena. The essence here is actually what everyone calls "routines."
Philosophy is the deepest routine, such as matter determines consciousness, consciousness reacts on matter, the dialectical relationship between internal and external causes, the dialectical relationship between major contradictions and minor contradictions, and the concept of development mentioned in philosophy. In fact, it is what we often call "routine", but this routine is very powerful, because it can guide human production practice and affect social development. When we analyze things, do operations, or do anything, we must establish layers of routines, that is, routines at the philosophical level, routines at the methodological level, routines at the technical level, and even routines at the equipment level.
Ok, after loading, let's talk about big data. What do you think of big data? What business models can be derived from big data?
We use this to interpret and disassemble our thinking. I think the first thing to do when looking at anything is to understand the definition or basic logic of this matter.
To answer this question, we must first understand what big data is! There is no specific definition here, just asking questions and giving examples! Does big data have to be presented in digital form? Is the collection of all telephone numbers in the world called big data? Is the data collection of operating financial reports of all companies in the world called big data? All economic data sets are called big data? If we want to understand big data in this way, it will be a bit one-sided. In fact, big data contains a collection of all elements, and we will give an example later.
Did big data only appear in the Internet era? The answer is also no. In fact, big data existed before the emergence of the Internet, and the Internet was just an accelerator for the formation of big data. Here we can give an example of big data:
All species in the world are a set of big data.
The collection of different temperatures, wind power, air humidity and negative oxygen ion content of all provinces, cities and counties in China at a certain point in time is a set of big data.
For another example, the collection of physiological characteristics of China/KLOC-0.4 billion people is a set of big data.
All user search behaviors included since Baidu search went online are a set of big data.
The taxi behavior record of all users of Didi taxi is a set of big data.
……
There are many such examples. Of course, the collection of all the different big data examples in the universe is also a set of big data.
After understanding the concept of big data, let's take a look at how big data works. The application of big data is not to pursue causality, but to play its relevance. For example, there is no causal relationship between the search behavior of netizens in a specific area and the flu in a specific time period, but the relationship between this search behavior and the flu can be established through digitization. Be more specific. Let's assume that the search volume of China and Guangdong netizens for keywords such as fever, cough and dizziness has suddenly increased in the last week. For example, these keywords have been searched for millions or even tens of millions of times, so it can be inferred that some kind of * * * phenomenon has occurred in this respect. According to the key words such as fever, cough and dizziness, we can calm down. There may be some kind of flu or epidemic in Guangdong Province.
How can you be so sure? Because everyone's search is purposeful or intentional, the keywords mentioned above appear, and it is likely that netizens are searching for such things as old cough. Or what is the reason for dizziness and fever? How to treat these symptoms? It may be nothing for two or hundreds of thousands of people to search for this at the same time, but if millions or tens of millions of people search for these questions with high-frequency words at the same time, there must be some kind of sexual problem, and it can be concluded that there is some popularity for the above problems!
This is the expression of the relationship between big data, that is, the analysis and prediction of big data. When we find this abnormal search behavior, we can quickly start the investigation and investigation of relevant departments, and generally we can basically conclude that there is an epidemic situation. At this time, the emergency mechanism can be quickly started, and the role of big data will come into play. This is an example of big data correlation, analysis and prediction. Now this analysis method has been widely used. For example, when you are driving, you often hear the broadcast of real-time road conditions or the display of real-time road conditions on the map. This is the result of big data. Then you can avoid these congested areas.
We can also give examples by assuming or imagining. For example, if we can digitize climate data and people's work efficiency and find out the correlation between them, we can accurately judge what level a person's work efficiency can reach under what climate conditions (temperature, wind, air humidity, negative oxygen ion content, etc.). ) when? In this way, we may not need a fixed working system of eight hours a day in the future, and we can even think that the intervention of climate data can best match people's mood and working state. Will this improve everyone's work efficiency by ten times or even hundreds of times?
Of course, if we are bolder, we can also predict people's next walk according to all the data of people's behavior characteristics. Of course, this is another story.
Let's go back to the original question, what is big data? What business models can big data produce? The previous question has been answered above. Let's continue to think about the latter question: What business model can big data produce?
Let's do a gradual deduction of ideas!
The big data we mentioned earlier must be massive, diverse and changing, so we can think that big data must not be owned by everyone and every institution, which means that the precipitation and output of big data must be scarce. Then companies or institutions with big data precipitation have a scarcity advantage. First of all, there is no doubt that big data is useful. Useful and scarce things will certainly produce commercial value. So the first business model of big data appeared: selling big data resources.
We went on to think that although some institutions have big data resources, they do not have the analytical ability or application ability of big data. For example, the government may lack this kind of thinking or what we call big data thinking, and do not have such professionalism or professional talents. If big data can't be analyzed, it can't generate value, so the second business model of big data appears: providing professional data analysis or data solutions and charging service fees.
Don't stop, let's continue! There are also professional data analysis and solutions. How to implement these solutions? For example, through analysis, we found that people's physiological characteristics, such as height, height, weight and butt contact area, are actually related to the strength and squeezed shape of the car seat, so we can make an intelligent car seat anti-theft system by solving the data! Then this data analysis and data solution company specializing in anti-theft system can't do it, and it can't land! So the third business model of big data has emerged: it is to make big data solutions into commercial products and make money through product sales or services.
We can summarize three mainstream big data business models:
1, selling big data resources
2. Provide big data analysis and solutions.
3. Implement big data solutions to form commercial products.
Of course, these three mainstream big data business models can derive many business models, and a wave of business model changes can be set off through big data thinking.
Of course, there are also companies or institutions that have the ability to apply the above three mainstream big data business models, such as Google, Ali, Facebook and Baidu.
The above views are not novel and may not be accurate. It's important to look at the specific demonstration ideas. In the process of commercial operation of big data, there will definitely be problems that are inevitable in any era, that is, fraud and flicker. In fact, this is easy for us to understand. Because in the practice of business model, it is inevitable to achieve the maximum output of benefits at the minimum cost, that is, to pursue the maximization of benefits. This is the essence of business operation, but it is only a phenomenon in the market and will be optimized in the process of market development! The commercialization behavior that truly meets the needs of users can persist until the end.
Maybe someone will ask, what are the criteria for judging these commercialization achievements?
Everything has standards, and standards are the reference for us to do things. Of course, the standard is not static. In fact, we can use what everyone thinks is big and empty here: the so-called standard, in line with the law, must be the biggest standard. Standards can be formulated by pioneers or pioneers, but the standards must be reasonable and high, that is, they must conform to the law. Unreasonable standards often appear in the industry, but these standards will eventually be replaced or optimized.
The standards mentioned above are all about the level of Tao, which is the biggest routine mentioned above. Of course, we need to concretize and refine the standards around this routine for reference and implementation.
Let's take agency operation as an example. Isn't the ultimate goal of agency operation to maximize profits for customers? Isn't the reality to maximize the agency's own interests on the basis of bringing visible effects to customers? Then we can disassemble it around this idea, and what kind of standards can be used to measure the effect of specific work representative operations? The essence of enterprise management is profit, so the measure of effect should be sales and cost. After specific distribution, this is a great measure. Then we split it. How can we increase sales and control costs? This involves specific means and skills, and the effect measurement of these specific means and skills involves specific standards. When we break this idea down to the minimum granularity, we can form a very detailed operating standard.
Similarly, the idea of determining the above standards can also be analogized to the standard of establishing community operation, such as the evaluation standard of community essence posts, the evaluation standard of high-popularity posts, hot posts, editors' recommendations and high-quality posts can be disassembled and further refined with the above ideas.