What Does Consumer Crazily Buy Mean For Marketers?

If you've spent the whole weekend watching frenzy, or other videos, you can enjoy it, because you're not alone.

Nowadays, all people are consuming electronic products and services in a "centralized" way. This way remains unchanged in the long run, and consumers will buy products from time to time within a short period of time.

Walton marketing professor Eric Blatlaw (Eric Bradlow) said that once marketers are aware of this phenomenon, and they have the data to follow up this phenomenon, they will be able to discover a new development of electronic content with rich content.

He pointed out that for decades, a basic measure of customer value has always been the RFM model, namely the latest consumption Recency (Frequency) and consumption Monetary (value).

"However, my research shows that this system is not comprehensive, and a" C "should also be added to represent Clumpiness," he added.

Brett Law explained the concept in an interview with Walton knowledge online, which came from his research paper, Dylan Small, a partner of Zhang Yao, a partner of Credit Suisse Group, and a new New Measures of Clumpiness for Incidence Data.

The following is an edited interview excerpt:

What is "centralization" in customer data and why is it important?

At present, the most popular way of marketing and customer value is to generalize the consumption behavior of customers through the RFM model, that is, the latest consumption, consumption frequency and consumption amount. That is to say, using all the information about customers in the hand to calculate the data of customers in the following three aspects, namely, when, when and how much is the most recent consumption?

This is a basic indicator for most companies to measure valuable customers and other customers.

My research found that this is not all the characteristics of customers. The RFM model should also add the letter "C", which represents the degree of concentration, indicating that some customers have certain rules of purchase behavior.

For a long time, customers will follow certain rules when buying items like orange juice and diapers.

But concentration refers to the fact that people rush to buy goods.

This period of centralized purchase can reflect some distinctive characteristics of customers, who may be extremely valuable customers.

Important conclusions:

The important conclusions drawn from my research are easy to understand.

Imagine assuming that you want to predict who will be valuable customers in the future.

You have four indicators that you can use to predict, which I mentioned earlier: the latest consumption, consumption frequency, consumption amount and marketing on customers.

This is a scoring model designed by many companies.

But I think you need to add one more indicator, that is, concentration -- how high the concentration of customers is.

This index is as easy to calculate as R, F and M indicators. It can be completed in Excel, and it can be calculated quickly, and it can calculate one hundred million customers' data in one second.

My research also found that in the experiment, customers with high concentration of purchase have higher value in the future. Even after using the RFM model and marketing cost analysis, the same conclusion can be reached.

This shows that we have found another variable about customers. Companies should follow up this variable and use it to predict the value of customers in the future.

The most surprising conclusion:

Two of the conclusions of the study surprised me.

First, I discovered that the RFM based model has existed for many years. At present, many companies are using it, and in this sense, it has been recognized as the simplest way to help explain customer value.

You can search all kinds of websites, and you can also set up many other variables, but concentrating this variable is the easiest.

So first of all, to my surprise, concentration has never been noticed. In other words, hot selling and cold spell can reflect some characteristics of customers.

Another surprise to me is that from the data I analyze, this concentration is applicable to electronic and online consumer goods, and is not suitable for regular mass consumer goods.

In other words, I can understand why the traditional model is suitable for conventional mass consumer goods, because consumer behavior of buying toilet paper and orange juice is regular, but consumption on website Hulu is irregular.

There are no rules for consumers to auction on eBay or buy books on Amazon.

Concentration is not applicable to traditional commodities, but for emerging products and emerging economies, it is a very convincing data of all data sets I have analyzed.

Pragmatic value:

I think this is probably the most practical research I have done in my 20 years of career.

I prefer to refer to my work as complex and complex statistical modeling.

This work is not only about statistical modeling, but also related to data concentration. Today, companies can do practical calculations without collecting additional data. They can be calculated using data similar to those calculated for R, F, M and customer lifetime values.

Using this model, we can determine the value of customers' prediction, and the ranking of customers' value will change. What changes do you have in recognizing customers' redemption value?

I have come to the conclusion that although highly concentrated customers are not stable, they are still the most worthy of recovery.

If you win these customers again, they will become highly concentrated and will spend a lot of money in the future.

So I think this model is of great practical value.

The charm of this model can be seen from my website. There is a Excel form listing some successful cases.

You can download this form and use the concentration degree to analyze it from today.

What new laws, procedural changes or strategies can you draw from your research?

Nowadays, many people are talking about big data.

I like big data, but I want to say that there is something more satisfying than big data, that is, data concentration.

Data enrichment means that you can now collect thousands of variables from people and find out the source of these variables, what items they buy, and what web pages to browse.

But this can not be called science, it is simply collecting data.

The question is which information is really useful for solving immediate business problems? This is what I call data concentration.

Therefore, I regard concentration as a supplement to traditional variables such as RFM, marketing activities and similar activities, and a form of data concentration.

I want to emphasize that you need to master more data, not just concentrate things to three variables, but to concentrate into four.

I have done a lot of research on concentration, and I know that it exists in various industries and has predictive value.

But I don't know what brought about concentration.

I associate marketing activities with concentration.

The company can make you spend your money by sending you an email, sending a picture album, and setting you a target customer.

Although I know the existence of concentration, there has been no research on the best way for the company to locate its target customers.

For example, I have never paid attention to whether a consumer will have a higher concentration in the purchase of a series of products.

Just imagine what it would be like to watch the "poisoned master", "mad man" or a series of similar videos.

For example, when a company tries to sell facial care series, moisturizing series and other similar products, will it package all products into a suit so that it seems to have the meaning of people struggling to achieve a goal?

I know how to perform mathematical operations on concentration, and I know that this is insignificant for a company, but it is foreseeable.

But I don't know how to explain it in psychology, which is what I have studied with other colleagues in consumer psychology.

We will conduct a lot of behavioral experiments in the laboratory to try to get the psychology behind people's concentrated behavior.

"Clustered" vs. "Crazy"

I like the word "concentration", others like "Crazy".

The reason why I like this word is that it is relative to "non concentration", just like a regular entry into a certain place or regular shopping.

We haven't heard the story of concentration, but when you see stories about people doing something crazy or consuming something, such as "a student staying up late watching a video for 18 hours", you understand the meaning of "Crazy".

The concept of madness is also very persuasive: every time I talk about this concept with managers, students or college students, everyone believes it exists.

Dispel misunderstanding:

My research should eliminate this misunderstanding: to some extent, the classification of customers is only a series of figures.

We need to consider more factors, not just what I call simple human behavior theory.

If we study the three indicators of consumption, consumption frequency and consumption volume, we will find that the basis of traditional consumer behavior ignores the interval time I proposed.

It shows that, basically, I can collect all the data, such as the consumption interval between consumers at this window is two days, then four days, three days or six days, but I can ignore all of them, just remember the last time consumers spend time and the number of times they spend.

In addition, we must dispel a misunderstanding:

Lack of information on shop rules.

In fact, the information of consumers is very detailed.

People flooded into the store, left, poured in and left.

These people are very different.

I think people can be divided into two types: centralization and non centralization.

Our results also show that the degree of concentration varies with product type.

For example, we found that female concentration was higher than that of men, and young people were higher than the elderly.

Therefore, I think what we need to clarify is that people are not only born equal, but also have many simple ways to divide people into several types.

How can this study be different?

Although a series of mathematical models have existed for 50 years, they are still very popular. In the past 10, they are called "hidden Markov model".

Suppose there are two countries in the world, one hot country and one cold country.

You take turns in the cold and hot countries.

This mathematical model is centralization. In hot countries, you will do many things. In cold countries, the opposite is true. This is how to alternate between hot and cold.

What I need to do is to provide professionals with a way to calculate simple numbers, namely statistics.

This is not a statistical paper, but what we call a number and statistics.

All you need to do is to figure out the numbers and use them.

Do your own research, you can use it to predict.

To determine whether men are more centralized than women, or to divide people into different types.

This is the typical and distinctive feature of this research. It is a metric based approach rather than a multi model approach, but these two approaches are designed to solve the same problem.

Proving that "concentration degree" may be a useful example:

We have studied the centralized consumers in order to prove that e-mail, publicity brochures and other different types of marketing channels are more effective.

We found that e-mails had no accident to become a better publicity tool in the short term, but the long-term effect was better.

What we need to study is whether certain words in e-mail, publicity brochures or video marketing are related to people's concentration or make people more centralized. Are some topics centralized? Are some product types related to concentration degree? What we have done before is to prove the existence of this phenomenon. And I know that many industries have this phenomenon. Some types of people are more likely to become concentrated people.

At present, my unfinished task is to discuss the marketing implications of concentration. As a marketing professor, I have failed to do that. It is indeed surprising.

I also need larger and updated datasets to relate information about marketing activities to people's centralized consumption behavior.

I know how to do it, but I still need more abundant and better data.

Follow up work?

I am thinking about following up three different aspects of the study.

For the first time, if I can analyze more data sets and prove how convincing the measurement of concentration is, I will be very excited.

I analyzed data from Amazon, CDNow, eBay, Hulu, YouTube and other traditional mass consumer company.

At present, I just need to apply these data to the new data set.

Secondly, I want to understand people's psychological process.

Why people's consumption is centralized? Third aspects and the last aspect. I want to link marketing activities with concentration.

This not only needs to understand people's behavior, such as what they do, what websites to browse, what items to buy, etc., but also needs to know the information about the marketing activities itself, perhaps the propaganda leaflet of marketing activities, and the channels through which these leaflets are distributed.

All these help me to find out the best solution to provide the company with optimized marketing activities so as to activate the scheme of people's consumption concentration.