"Peel" the User Role creation role model method

Recently, I have been doing some active page and mobile game, I gradually realized the importance of the role model. Role model, is the design of products, refers to the street lamp, product managers and interactive Designer design reference.

Creating a role model is skinning (like peeling an onion, although it will shed tears, but does the onion taste good)? Yes, we need to peel out the souls of the users and then give them flesh and clothing (demographics). In this way, we will feel the user is around us, vivid image, impressive. Just skinning is not enough, we also need to summarize the classification, understand the user's goals, ideas and behavior, find out the differences between users and common ground.

According to the different types of user research and analysis methods, there are three ways to create role models: qualitative personas, quantitative qualitative personas and quantitative personas. Based on the Alibaba Chinese station transaction line user role model project, the following creation methods are analyzed:

Research methods are also many, commonly used methods are: questionnaires, user interviews, on-site observation, usability testing, data analysis, Web site traffic/log analysis. In the transaction line project, interviews, questionnaires and data analysis are helpful to discover the users ' goals and viewpoints; Site observation, website traffic/log analysis is helpful to understand user's behavior.

During the process of creating a role model, you often encounter the following issues:

1. How to use the data to subdivide? How do you look at the rules of data?

Find the latitude variance from the data and find all the relevant factors that contribute to the difference.

2. How to design a questionnaire? What latitude?

The entire process order by transaction-management-payment-Logistics and Product dimensions (consider the user's actual operating process).

3. How to write a deep interview outline?

Understand the user's information, refer to the user Role partition dimension questionnaire.

4. How to conduct CRM analysis? See Related Topics

5. How do I make a crosstab analysis? See Related Topics

6. How to subdivide users?

In general, breakdown by user goals, by the use of the cycle of segmentation, with the combination of behavior and ideas to subdivide. In the transaction line personas, the breakdown of roles is a key factor in driving user goals, behaviors, and viewpoints, such as: different sources of goods and different motives for shopping.

7. How to preliminary test subdivision latitude?

The subdivision group can explain the key differences known, such as: the purchase target (second-hand housing user and new house user) different, can explain the difference of keyword search use; the subdivisions should play a decisive role in determining functional design, interaction design, and sketch.

6. What are the methods for quantitative verification?

Data Cross tab analysis (CRM analysis, quantitative questionnaire, website traffic/log analysis), statistical analysis.

7. What characteristics do personas need?

Reference role model parameters, personas are driven by goals, behaviors, and viewpoints, not some simple demographic features.

8. Use of Persona models?

Develop new features and functional improvements (understand user needs), interact with design details (understand user habits).

Related topics to learn when creating a role model:

1. CRM Data Analysis

Bind a user's history and values to his questionnaire, looking for an intrinsic connection to better define or describe personas. It includes three types of data: Transaction record, financial data and demographic information.

transactions, which show what products or services the user has purchased, how often they are purchased, will strongly affect the site's objectives and behavior and can be used as a basis for user segmentation. Financial data, using numbers to measure the financial value of different personas, can also help determine the priority level of individual personas. Financial data can be associated with a user survey questionnaire. Demographic information is not a significant determinant of character creation, and personas are driven by goals, actions, and perspectives.

2. Website Traffic Analysis

Two ways: A. Find the behavioral patterns that determine their role, analyze the data, and try to connect the data results with the segmentation of the group behavior. B. The individual user's click Stream and his response to the questionnaire are bound together for further detailed analysis. Explore the various behaviors of the user, includes the entry page (where to enter the site), the reference page (from where, such as the Search Keyword page), the Export page, the common path, the functional use (most visited content and function), search conditions, conversion rates, duration and frequency of visits, and so on.

3. Segment Users

Target segmentation: User goals such as: Want to buy a house, want to understand the market dynamics and want to sell goods; use cycle segmentation: The same person uses the site as a different persona at different times; behavior and Perspective segmentation: If the site has too many functions or a long product category, it is not easy to segment by Target, consider using this method to subdivide. Through the behavior and the point of view matrix, the quadrant becomes the main basis of the subdivision user.

4. Quantitative Verification

Data Cross tab analysis to analyze whether subdivision latitude can explain or affect other key differences, verify that the subdivision latitude is correct (you can use Excel PivotTable reports). Statistical analysis, calculates the ANOVAs (variance analysis) of these subdivision data to measure whether these differences are statistically significant.

Article Source: Tencent Isux