This is effective user segmentation, not just high, medium and low

This article will systematically explain how to conduct effective user segmentation, including the operational steps of segmentation, how to choose classification dimensions, and the difference between user segmentation and recommendation algorithms, to help companies achieve refined operations.

We have said that user segmentation is a special form of user segmentation: segmentation by value. So how to do general user segmentation? Why are many students criticized as "useless" after completing segmentation? Today, let's give a systematic answer.

1. Intuitive experience of user segmentation

User segmentation itself is very simple. For example, the user stratification we talked about in the previous section is actually a simple user segmentation based on a classification dimension, such as:

• Segmentation by user spending in the past year: High-level (10,000+) Intermediate (5,000-10,000) Low-level (1-5,000)
• Segmented by user activity behavior: Active (logged in more than 15 days in the past 30 days) Inactive (logged in ≤ 15 days)
• Even simpler, segment by basic attributes: male/female, old/middle-aged/young

It is easy to do user segmentation, but it is difficult to do effective user segmentation. The so-called effective means that it can help operations, products, marketing, and sales.

For example, we have distinguished high, medium and low-level customers. We know that high-level customers are very wealthy, but how should we serve them? When, in what scenario, and what activities should be done? It is still unclear. Therefore, it is not enough to stratify by one dimension alone. We need more classification dimensions to make more detailed divisions.

Let’s look at a small example:

Let's see how this example can be analyzed:

Let’s first look at consumption habits. In terms of the income they contribute to the company, the three categories A, B, and C are at the same level.

But in fact, the three categories ABC represent three different consumption habits:

A. Centralized purchasing (most likely to buy the cheapest on Double Eleven)

B. Seasonal purchases (it is very likely that new products will be introduced every quarter)

C. Frequent purchases (high daily activity, most favored by operations)

Different user consumption habits will directly affect the means of operation:

A. Centralized procurement: one big event will be launched!

B. Seasonal purchases: new product promotions every season

C. Frequent purchases: check-in + points + weekly activities

To decide which one to use, you can refer to the proportion of the three categories of ABC in the entire user structure and choose a main tactic. The effects are as follows:

Note that the existing ones do not necessarily mean they are reasonable. It is also possible that the leader said: Although we currently have 60% of Group A, we hope that Group C will account for 60% in the future and we need to change the status quo. In this way, when choosing tactics, we have to consider the characteristics of Group C more and find activities, products, and discounts that better meet the needs of Group C users. In short, a more detailed understanding of user characteristics can help operations.

This is the intuitive effect of user segmentation: through segmentation, more detailed data guidance is provided for operations. Of course, for the sake of teaching convenience, the data in this example is very extreme. In actual operation, as long as you can find a classification dimension with high enough differentiation, you will have a similar effect. The core question is: how to find it. This is the key to user segmentation.

2. User segmentation steps

1. Step 1: Define what is “effective”

This step is very, very important. There are endless ways to segment users. If you don’t define what is an “effective” segmentation in advance, you will be stuck in a sea of ​​details.

Many novices tend to overlook this point. When it comes to user segmentation, they rush to cram a bunch of user feature variables into the clustering model. After the clustering is completed, they are at a loss and ask everywhere: "Is there a scientific and authoritative standard for user classification, and anyone who challenges it will be dragged out and severely punished?"

In the end, the operation department criticized him for what he was doing! This was because he was divorced from the actual business and only focused on addition, subtraction, multiplication and division.

Effective classification standards are, of course, based on operational needs. We can break down the corresponding data indicators from the operational goals, KPIs, and tasks. For example, the operational task is to increase revenue. We follow the steps below to transform business problems into analytical problems.

With the classification criteria, we can check whether the segmentation is effective. For example, the goal is to find a user group with high cumulative consumption. Then in the end, we need to see whether the consumption difference of the segmented groups we found is high enough and whether we have really locked in the high-spending group. The specific effect is shown in the figure below:

2. Step 2: Find the classification dimension from the perspective of operational means

After finding the classification criteria, we can see from which dimension to segment users to make the differences between user groups more obvious. This is another big pitfall, because it seems that there are a lot of optional dimensions.

Many students are confused about how to choose. Or after finally choosing, the operation asks: Why is it divided like this? He answers: There is a big difference! Then he is criticized for not understanding the business and doing it blindly. So depressed...

In fact, there are certain standards for filtering classification dimensions, and there is no need to run around everywhere:

(1) How to choose the classification dimension

①Choose dimensions with reliable data sources

For example, many companies do not have strict data collection processes for basic dimensions such as gender and age, so there are many gaps in the data and it is difficult to guarantee the authenticity. Don’t use these. Try to use reliable data such as consumption, activity, and registration source.

②Select dimensions that operations can influence

For example, the development team may be very concerned about the device model, but the operation team can't do anything even if they know it, so don't choose it at this time; there are some indicators that the operation team pays special attention to. For example, if the operation team wants to issue coupons, then the user's coupon collection rate and usage rate are particularly good indicators.

③Select indicators with obvious differences in their own stratification

Some indicators have little difference in themselves and the data distribution is very concentrated. In this case, they are not preferred. Indicators with larger distribution differences are preferred.

Based on the above three standards, you can avoid blindly conducting experiments to find a needle in a haystack, and also avoid being criticized by operations as "what's the use of this".

Some students may think that this process is very similar to finding features when building a risk control model. It is indeed very similar, but there are differences. The business actions corresponding to the risk control model are only "pass/reject", so there is no need to consider so much.

When doing user segmentation for operations, you need to consider: activity theme, time, product, selling point, communication channels... a lot of things, so you must consider which dimensions are useful for operations.

3. Step 3: Try segmentation and observe the results

With the classification dimension, we can try to divide the classification criteria:

There are three very thorny issues here:

• What are the different categories and how many segments are there?
• How many classification dimensions should be added?
• How many categories are appropriate?

Let’s start with the results: In principle, the final number of categories should not be too many, and each group should have operational significance.

To operate an event, you need to design posters, prepare goods, develop systems, and prepare to deploy resources. Therefore, if the group size is too small, it is not suitable to do an event alone. Therefore, when doing user segmentation, it is customary to limit the group to a maximum of 8 categories (each group has a share greater than 10%). As for the specific size, it can be designed according to the project goals and operation conditions.

Under this general principle, it means that there cannot be too many classification dimensions and divisions of each dimension. Try to select key dimensions and key division points.

If there are too many dimensions, you can consider using a dimensionality reduction algorithm to compress them. When segmenting in each dimension, you need to pay attention to the following issues: If you segment in a single dimension and find that some segments are special, you cannot merge them at will (as shown in the figure below)

In short, the classification process requires repeated attempts in many steps until the ideal result is finally output.

3. Special Note: The Difference Between User Segmentation and Recommendation Algorithms

Many articles on the Internet confuse user segmentation with personalized recommendations for different people. Although many people will say verbally: we do user segmentation to understand user needs and achieve personalized recommendations for different people, in business terms these are two different meanings.

For a segmented group, operations can take many leading and innovative actions. For example, if we want to expand the high-end user group, we can launch a new product series, a new reward policy, and a new service to attract high-end users. As long as I understand their preferences and behavioral habits, I can do it very accurately.

However, the premise of the new design is that the user has a certain size and is worth doing. Therefore, when doing segmentation, we cannot consider too many dimensions, cut too fine, and make the promotion extremely complicated. I want to let everyone know that we are doing this, so that we can form a herd effect and achieve greater results.

The recommendation system is not subject to this restriction. The recommendation system completely closes the information channel. Everyone sees something different. As long as the user response rate can be increased a little, it will be fine. Therefore, the recommendations are all existing, stock products, trying to achieve the match between users and products.

The recommendation system cannot generate new ideas and new effects, nor can it design new products. So there is no need to worry about whether the split is detailed or not, as long as the business goals can be achieved.

4. Summary: The real difficulty of user segmentation

After reading the whole process, you will find that user segmentation is a simple concept with complex operation. The complexity of operation is not the modeling process at all, but the grasp of the target, the selection of dimensions, and the grasp of the segmentation size, all of which must take into account business needs.

Although data and statistics provide us with many tools (classification tools, dimensionality reduction tools), we still need to consider specific business scenarios to put them into practice. We have never lacked students who can recite textbooks, but we lack analysts who can consider actual scenarios.

Many newcomers don’t understand this. If you ask them:

• What goals does user segmentation serve?
• Do the “core users” mentioned by the operations refer to those with high spending power, high activity, and referral behavior?
• What can you do after knowing the “male/female” operation?
• What are the operational means to achieve the goal?
• If there is only a 200 yuan difference in consumption, how much room is there for operations to operate?

Their answer, of course, was: We don’t know.

Then he stubbornly asked: Why do you care about this!!! I just want to know, isn’t there an authoritative standard number of categories for Kmean clustering in the e-commerce industry!!! Is it 5 or 8!!!

A special reminder: living in the books of the school library will not solve the actual problems of the enterprise.