How to build user portraits to achieve brand marketing

User portraits, also known as crowd portraits, are labeled portraits abstracted based on customer demographic information, social relationships, preference habits, consumption behaviors and other information.

The core work of building a customer portrait is to put a "label" on the customer (just like the personality of a star in the entertainment industry)

The label consists of two parts:

< p>1. Directly obtained based on customer behavior data

For example: data that users actively fill in on the website or APP. Strictly some platforms will require customers to upload ID cards, student ID cards, driver's licenses, etc., so The data accuracy is higher.

2. Obtained through a series of algorithms or rules mining

For example: a user has recently started to purchase maternal and infant products, milk powder diapers, etc., then based on the frequency and quantity of the customer's purchases, Infer whether the customer is a new mother/father based on the customer's age and gender.

User profiling actually hopes to identify users through certain means, divide them into groups or individuals that are the same or different from each other, and then provide differentiated services for observation and analysis.

The value of user portraits

1. Precision marketing

Precision marketing is the most direct and valuable application of user portraits or tags. This part is also the work content that the advertising department pays most attention to. After we put various "tags" on each user, advertisers (stores, merchants) can use the tags to delineate the customers they want and carry out precise advertising.

2. Support products

For a product to be widely used, audience analysis is essential. Product managers need to understand users, so user portraits can help product managers see through the surface of customer behavior to see the deep motivations and psychology of customers.

3. Industry reports

Through the analysis of customer portraits, we can understand industry trends, such as analysis of consumption preference trends of the post-90s generation, analysis of brands favored by high-end customers, and consumption differences in different geographical categories. Analysis and more.

After talking about so much "dry stuff", are you a little confused... Let's look at a simple case to help you understand better.

Scenario case

There is currently a Douban rating data of ten movies by more than 200 users. We use this data to depict several groups of user portraits.

The ten movies are as follows:

Action: The Bourne Supremacy, Operation Mekong, Hacksaw Ridge, London Has Fallen;

Youth Love: Beijing meets Seattle, July and Ansheng, Six Lane Cafe;

Animation: Zootopia, Kung Fu Panda 3, Big Fish and Begonia.

The following begins with user segmentation and characterization:

1. User segmentation

(A rating of 7 or greater is defined as "loving this type of movie")

The three types of audiences are similar in number, which shows that people have different tastes and hobbies and are relatively average.

2. Comparatively depict user portraits of action movies and romance movies

(1) Gender ratio

As expected, more girls like youth romance movies. Girls have more delicate emotions and are sentimental. For example, "Love in Manhattan" can be accurately recommended to such girls when it is released; action movies are indeed dominated by boys, and boys like movies with action and cool scene effects, which can arouse their "hero" "Feelings", such as "Thor 3" can be recommended when it is released~

(2) Are you single?

Dear single friends, will men be silent after watching it, and women will cry after watching it? ah! Combining the above data, when we recommend it to single young men, can we write something like "Watch more romantic movies, maybe you will meet your future girlfriend" in the copy.

(3) Social preference

The number of comments on the movie is used to divide the degree of preference. Those with 25 or more comments belong to social preferences. Focus marketing can be carried out for this group of people. , they can bring secondary promotion effects to the movie.

(4) Brand preference

Although Apple is expensive, it is still recognized by most people. Can we boldly speculate that Apple users have certain spending power? Recommend some high-end theaters or 3D giant screen movies.

(5) Job Distribution

Among those who love romance movies, where there are more girls, there are indeed more female occupations such as nurses and teachers; on the other hand, there are more boys who love action movies. IT and engineers account for the majority, but the most obvious one is the student party team. Students still have more free time, so film promoters can promote more in major universities and hold meetings, etc.

(6) Geographical distribution

It can be seen that people in some big cities like to watch movies to relax and entertain themselves after busy work. The movie staff are It is not possible to arrange more studios in big cities to promote box office growth.

From the simple case above, we can see that user portraits make product service targets more focused and focused, can better meet customer needs, achieve precision marketing, and improve the company's operating efficiency.