Product managers build user portraits persona: it can be unprofessional, but it must be authentic

In the face of building user portraits, a hundred product managers, ninety-nine have inexhaustible bitter water. Should we focus on the user or the product itself, and be responsible for the project and business? This problem is probably a pain point that many product managers can't understand. The author of this article believes that product people do not have to pay too much attention to building user portraits, fall into the evasion of single point work, and elaborate on the specific methodology of product people facing the construction of user portraits. Recommend people interested in the product to read learning ~

• "Is there any template for a character model that can be set?"
• "How many people are going to be investigated to get this portrait?" How to make it? ”
• "I haven't had a user role, what books can I read?"
• "I did a questionnaire, this data is enough to make a portrait, right?"
• ……..

This is a question raised by many new product people, and I have always stressed to them that as a product person, there is no need to pay too much attention to building user portraits.

First, the "user portrait" is true and false

1. Don't get obsessed with points

Doesn't it mean that product managers should pay attention to users and grasp user needs? Why not pay too much attention?

First of all, from the role responsibilities of the product manager, the product manager should take the balance of the business, participate in various links but can not go deep into it, otherwise the vision will focus on a single point, this is not the product manager, but the function planning, especially in the second half of the current mobile Internet, user needs have long been tapped, product managers should pay more attention to the ever-changing market.

Secondly, interlaced as a mountain, do not underestimate the professionalism of the user model, just build the need for sociology, anthropology, psychology and other professional knowledge, if the product manager is running to focus on the user portrait, this is not contempt for the user experience design team, including interaction, vision, research, professional things should be handed over to professional people.

Even if the product manager really wants to do the user portrait, what is the actual action?

Most of them just think of a few painless questions, do an online questionnaire, and send a red envelope to some groups to ask grandpa and grandma to fill it out, in fact, this becomes formalism.

Early user research was not done properly, resulting in a lack of first-hand data and no authenticity.

And falsehood will definitely affect the product manager's judgment of the business, once the direction is wrong, followed by a wrong mess, which is a fatal professional mistake for the product manager.

2. "There is not enough data, experience to make up"

How did most user portraits end up?

Although there are many professional theories for guidance, but after all, it is professional and academic, many product people feel that the scalp is numb as soon as they see it, and the time-consuming and labor-intensive output is not obvious, so in the end it is more than fooling around:

Export some of the data in the questionnaire you put in before, and then find the user reports compiled by others in the industry, combine the two, fill the content into the template, and complete!

In the end, the user portrait really becomes a "portrait", just to complete a necessary part of the PRD, and after the end of the project, the portrait is thrown aside, and it is not iterative and no longer refers to it, and the result is taken away by the user's "real" feedback.

It is said to be user-centered, but in fact it is still a subjective self-reference design, which moves itself.

Second, the professionalism of user portraits

I believe that everyone has seen the "Essence of Interaction Design", this book can be described as a classic textbook of the industry, the theory in it is also very professional, but this theory is summed up by the author's countless times of industry experience, if the operator does not have enough field knowledge and project practice, it is Dong Shi Gong.

Let's take a look at the relevant concepts and theoretical processes of user portraits, after all, they are foreign works, and I have made some of the changes described according to my own experience.

1. User portrait concept

First, let's be clear: the model represents complex phenomena through effective abstraction, emphasizes the significant characteristic relationships of the structure it represents, weakens unimportant details, and visualizes them.

Therefore, the user model/user portrait/persona: refers to the use of user research results to build a descriptive paravirtual model of the user.

It is semi-virtual because it is not a real person, but comes from the behavioral data and motivations of many real users in user research, based on real and actual user behavior patterns.

The person model, which clearly explains user goals and helps researchers understand user goals in specific situations, is widely used in the field of interaction design, not as an accurate metric, but as a visual communication tool for decision design communication.

2. Classic user portrait professional process

According to the contents of the book, after doing user research and preliminarily collating the data:

1. Interviewees are grouped according to roles; respondents are categorized according to different roles.
2. Identify behavioral variables; divide notable behaviors into several different sets of behavioral variables, such as attitudes, abilities, and other dimensions.
3. Map the interview subject to the behavioral variables; determine the relative position of the respondents on the behavioral variables.
4. Find important behavioral models; look for the aggregation of subjects, find significant behavioral patterns, and find logical connections.
5. Synthesize various characteristics to clarify the goal; get close to the observed behavior, infer and define the goal behind the behavior
6. Check for integrity and redundancy; close loopholes and remove duplicates.
7. Specify character model types; prioritize, identify primary design goals, and identify primary and secondary character models.
8. Further describe the characteristics and behavior; add virtual user situations to make the character model more realistic.

It can be seen that there are not many steps, and the content is very clear and easy to understand, but the last one is changed to steps 1, 2, and 8, and the most important behavioral data classification and focus in the middle are directly sloppy.

It's like "it takes a few steps to put an elephant in the refrigerator", and a professional process does not mean that it is very convenient to have a meticulous output in the end.

In other words, there is no diamond diamond, no porcelain work.

Third, the method suitable for the product people

"Why not eat minced meat?" Who doesn't know that the profession is very important, but my company does not use research and interactive posts, ah, I do it alone. ”

Now back to the question I just said:

Since the external environment cannot provide this, but it is necessary, then we can only change ourselves, we cannot follow the inherent process, we must make adaptive changes.

I have always opposed the hard way to move the routine without following the actual situation, which is actually a manifestation of laziness and non-thinking, and excellent product people must use limited resources to find a suitable way out of the actual problem.

In the following, I will combine past experience to give you a new product a feasible character model building method.

1. User Research

This is the first step, which includes questionnaires, personal interviews, scene observations, personal interviews, etc. Telephone interviews are not recommended because we cannot see the user's emotional changes and body movements, of course, during the epidemic, it is also a secondary auxiliary method.

It's hard to get started, and if you don't even want to take the initiative to grasp the opportunity of direct contact with users like interviews, it's hard to believe that he will produce a subjective and convincing portrait.

This link also does not need professional operation, detailed operation can click on my homepage, see my other two articles, one is to introduce the theory, one is to introduce practical operation cases.

2. Cluster data

As the name suggests, the qualitative answers and quantitative data in the interview topics in the previous study are extracted and processed according to the method of cluster analysis.

Here is an approach: the KJ method.

Also known as the associative diagram method/card method/z-diagram method, it is a qualitative method of consensus-building based on the similarity of the collected data and information for categorical and comprehensive analysis, with the purpose of layering individual ideas and insights to show common structures and themes. Can help groups organize complex ideas and information, can be used to identify problems and needs, and can be used to find solutions to problems.

Operation: Classify the cards according to the strength and weakness of the content connection, group the cards with common points together, and add an appropriate title, write on a card with a color pen, called "group title card". Cards that cannot be categorized, each in its own group. All the categorical cards, with their affiliation, are pasted in the appropriate spatial position, and the lines are used to connect the related ones.

Because we get the data under a number of topics, at this time, we need to group the data under a certain topic again, and extract the aggregated views, popular points, that is, to find out who answers is the same.

These topic types include, but are not limited to, pain points, needs, scenarios, attitudes, ideas, and so on.

The following figure is a simple illustration, similar colors (answers) can be grouped together, and several sub-categories can be extracted.

In fact, this step is to go deeper into the summary behavior variables in the previous classic process, but it is simplified here.

3. Convergence combination

This step requires comparing the clustered data with previously unstructured first-hand data, because the subject behavior variables cannot all be accurate to a person.

At this time, it is to classify the user groups of the research and extract the most typical main users and secondary users.

For example, if 20 users are surveyed and 15 of them behave in a generally consistent manner, the same behavior of these 15 bits is the behavior of the main user, and so on, where another type of behavior of 13 users is consistent and one of the behaviors of the main user.

4. Improve the summary

When the user data is integrated into the main persona, the content can be populated, as shown in the figure:

1. Say the most painful point of the user in one sentence, such as "Buying a mobile phone always does not know when the lowest price is!" ”；
2. User avatar + demographic data, such as age, occupation, region, etc., this can use the statistics in the questionnaire;
3. Write about certain personality traits, don't write about what is cheerful and optimistic, but what you observe, such as "learning is easy to move yourself";
4. Explain the user's Internet usage habits, such as TikTok short video heavy users;
5. Character preferences, which are also content for writing observations or summaries, such as "listening to music while studying";
6. It is to set the problem scene or product use scenario according to the behavior pattern obtained by the focus, that is, what problem the user encounters in what context, what time and where, what he does, what the results and impacts are, what the feelings are, and so on. You can write this piece as a movie script;
7. It is the content of the outline, and what kind of pain points the user encounters in order to achieve what goal.

Note that the above content is a general classification, we should make appropriate adjustments according to the actual work content, do not use a template to the end, the template is to provide ideas and references, rather than a rigid set.

4. Conclusion

Product managers can not replace the professional work of UXD positions, can only participate in it, to learn their thinking, rather than some fancy template hard set content, it is better to summarize some real data.

Many people say that to be a product manager, you must be professional, and you must be specialized and refined, and you must always consider the user.

But in my opinion, the professional of the product manager is always responsible for the project and the business, to be a business person who can make money, after all, what is done is a commercial product, and to make money, it is necessary to put more energy on thinking, to observe the market, coordinate resources, track data, and pursue user value, rather than falling into the escape of a single point of work, hidden in the excuse of "user experience".

This article was originally published by @loosaumin everyone is a product manager, and reproduction without permission is prohibited.

The title image is from Unsplash and is based on the CC0 protocol.