I'd like to see the avatar of XXX user or if I can see the avatar of XXXX user. Or other products will ask: how do you make user portraits?
But in the process of communication, I found that different people's understanding of user portraits is still very different. Some people think that user portraits contain detailed information about users, some people think that user portraits can reflect the statistical characteristics of a group, and some people think that user portraits can be used for user research ... These ideas are more or less one-sided. This paper expounds the basic knowledge of user portrait, analyzes some DMP products, and explains the application of user portrait in K 12 products.
1. What are we talking about when we talk about DMP and user portraits?
This part combines the common DMP &；; The definition of user portrait and my understanding of DMP&; Explains the definition of user portrait.
User portrait is a very important link in DMP, so DMP and user portrait are explained separately.
1. 1DMP
What is 1. 1. 1 DMP?
Dmp is a data management system and data management platform. From the name alone, this definition is still very broad, so many domestic enterprises or individuals will misunderstand the core functions of DMP.
According to my understanding, DMP is actually a comprehensive platform for data collection, processing and integration. It absorbs data from various data sources, cleans up and arranges structured data tables based on users, and calculates user tags to accurately describe various users.
Pure DMP platform refers to a small DMP technology service provider with strong customization ability and good neutrality. The US DMP market is extremely subdivided and the China market is highly integrated. The demand of DMP is often closely linked with DSP and SSP, so it is difficult to have a pure DMP platform at present.
1.1.2 What can DMP do?
Precision marketing, advertising, personalized recommendation and other applications.
Infrastructure and data processing flow of 1. 1.3 DMP
Infrastructure of DMP:
DMP data processing flow:
Practical application of 1. 1.5 DMP (products that can be seen in the market)
DMP advertising platform: Tencent Guangdiantong, Ali Mama Dharma Disk;
Independent third party DMP:talkingdata, Ce Shen Data;
Personalized recommendation: today's headlines, a little information, Taobao, JD.COM, etc.
Note: We can feel the application of personalized recommendation, but we can't see the logic behind it.
Other applications
1.2 user portrait
What is the user portrait of 1.2. 1?
There are two definitions of user portrait: user role and user profile.
User roles: typical users abstracted from user groups by product designers and operators. For example, in the user research stage, the product manager understands the similarities and differences of users through questionnaires and customer interviews, and concludes them into different virtual users;
Classic case-user portrait mentioned in user experience elements
UserProfile: according to the user behavior data of everyone in the product, a set of tags describing users is generated. For example, guess whether this user is male or female, where he lives and works, which star he likes, what to buy and so on.
The user portrait mentioned in this article refers to the user profile.
1.2.2 How to make a user portrait?
Basic data collection: collect static data and dynamic data of users inside and outside the website;
Behavior modeling: based on the basic data of users, behavior modeling is carried out through technical means;
Architectural Portrait: Through behavior modeling, a series of user tags can be output, and each user's tag can form a collection, which can represent the characteristics of this user.
1.2.3 Common applications of user portraits
Personalized recommendation (e-commerce, information products), risk control, forecasting, etc.
1.2.4 Relationship between user portrait and DMP
DMP is a data management platform, which can be simply understood as providing data to DMP platform, and DMP platform outputs a series of labels or other desired results.
User portrait is to input user data into DMP, and DMP outputs user tags.
DMP can output not only user tags, but also other tags, such as inputting articles and outputting article tags.
Therefore, user portrait is an application direction of DMP.
2 related product introduction
Because user portrait is mainly used as the underlying application, many of its applications are perceptible, but invisible. For example, the personalized recommendation page of the e-commerce platform and the personalized recommendation of the homepage of the information App are all behind the user portrait, and the user tag and the content tag/product tag are intelligently combined.
Because I am engaged in K 12 education industry, I only choose two competing products: open DMP platform and educational products, among which the open DMP platform is the main one.
3 competitive product analysis
3. 1 Open DMP platform
3. 1. 1 product description
3. 1.2 function comparison
Dharma board
Guangdiantong
Ce Shen data
Note: Ce Shen data looks more like a data analysis tool, but its underlying structure and data management are similar to DMP, and the part we can see, which is similar to BI function, can be regarded as the performance of DMP in the application layer, so it is also listed as a competitive product.
3. 1.3 interface comparison
Note: Because all three products need to pay to experience all product functions, the interface mainly comes from description documents, which may be different from the real version.
Dharma sign-label
Dharma plate-new label
Dharma plate-crowd report
Dharma Plate-Crowd Details
Dharma plate-overall report
Radio and TV communication operation interface
Guangdiantong-making advertisements
Guangdiantong-Creating Advertisement 2
Ce Shen Data-User Analysis-Event Analysis
Ce Shen Data-User Analysis-User Attributes
3. 1.4 Thinking and comparison of the underlying technical architecture of products
Description:
1) This part is to think and abstract the possible underlying architecture through product experience and reading documents, which is incorrect;
2) The underlying technical architecture of the product: The functional modules we see are all realized through the cooperation of different technical modules. The underlying technical architecture of the product describes the underlying technical modules corresponding to the product and the relationship between the modules.
Dharma board
Ce Shen data
3. 1.5 Summary
To sum up, Dharma Disk and Guangdiantong are typical applications of DMP in Internet advertising. DMP is the core of targeted advertising, and the user portrait provided by DMP is the core and key step of targeted advertising. The most important goal of Dharma Disk and Guangdiantong is to show the right advertisements to the right people at the right time.
As for Ce Shen data, as a data analysis tool, DMP also plays a very important role in the process of data analysis and data visualization. What kind of user behavior is more prominent, and what kind of users will there be in the future.
3.2 Educational products
Application of user portrait DMP in educational products-
( 1)? Onion data personalized course development (similar to adaptive learning);
Before learning, users should be tested according to the test conditions and make personalized course packages, as shown in the following figure1;
Figure 1
After the user finishes learning, a visual learning analysis report is displayed according to the user's learning test results, as shown in Figure 2 below.
Figure 2
4 how to use the user portrait?
User portrait is a new underlying product, which is usually difficult to see in applications. So, how to use user portraits? Combined with the analysis of DMP products and educational products, the application of user portrait is summarized as follows:
(1) User Label Visualization-
Related product functions: Guangdiantong and Dharma Disk directly display user labels, and users can directly select labels, combine labels, select target users, and then place advertisements.
For example, I want to send a push message to these students who have just signed up for courses this year, have high spending power and study hard, hoping that they can participate in a live broadcast to promote their mastery of knowledge. Then you can directly choose the label: new users+high spending power+study hard, and then send text messages to these users.
Benefits: refined operation and improved ROI；; Simplify operation
(2) User tag association analysis:
Related product functions-Guangdiantong Looklike: (1) provides seed users; (2) screening the characteristics of seed users; (3) Match seed users with Tencent users to expand the crowd.
Scenario for reference: A group of users bought product A, and I want to find out the users with high similarity to the product. Then you can import the information of these users into DMP, calculate the labels of these users, and then find users with high similarity to these users through correlation analysis.
(3) Personalized recommendation:
Related product functions: personalized curriculum development
Scenes for reference (taking K 12 education as an example):
(1) Recommend personalized course pages for unregistered users according to their visit behavior, so as to improve the purchase conversion rate; If users come in through internet advertisements, they can make personalized landing pages for them to improve registration and purchase conversion;
(2) Recommend personalized course pages for registered courses that have not been purchased according to their browsing behavior, so as to improve the purchase rate;
(3) According to the user's learning behavior (mainly solving problems), make personalized test questions and learning suggestions for them.