User operation: How to build high-quality user portraits?

If there is any core ability that a product manager needs to have, then the answer must be "understanding users." As for how to demonstrate the extent of a product manager's understanding of users, it must be the ability to output a qualified user portrait.

1. Why do we need user portraits?

As a product manager, we always hope that the products we make can meet the needs of users well. However, everyone is limited by their own perception of things, which may lead to deviations in the understanding of users.

Especially when everyone in the team regards themselves as users and defines requirements based on their own understanding of the product as the user's understanding of the product, the so-called "elastic users" are likely to appear. Everyone says that it is for the sake of user experience, but the users defined in this way are obviously not the real users of the product.

So the user portrait tool came into being. It is a prototype built by real users to help product designers develop product functions, service strategies, and sales strategies in a targeted manner. Fundamentally speaking, user portraits are to help product managers understand users.

But can the user portraits we build really help us? Let’s look at an example of a common user portrait:

In this user portrait, we can see some basic information about the user, such as age, gender, education, marital status, and work information, as well as some of the user's characteristics and goals. It feels very complete.

Now this user has neither a house nor a car, so the question is, assuming I am a car seller, what kind of car should I recommend to her?

Maybe you will say that she is more literary, so you should recommend her a car with a literary atmosphere.

Maybe you will say that she focuses on enjoyment, so you should recommend a luxuriously equipped car to her.

Maybe you will also say that she has neither a car nor a house, so you can recommend an RV to her, which solves the problems of both the house and the car.

So even with user portraits, "elastic users" still appear. So what is the problem? Why can't this user portrait help us make correct decisions?

2. Population structure ≠ user profile

User portraits were first proposed by Alan Cooper. In his classic book "About Face", there is a special chapter on user portraits. In the book, he mentioned that the core of user portraits is to observe users, list some unique aspects of the observed behaviors, and form a set of behavioral variables .

Although demographic variables (such as age, gender, education, geographic location) and other factors also have a certain impact on behavior, this influence cannot constitute the core of differentiation between users. What really forms the core of differentiation is user behavior, or more deeply, the motivation behind user behavior.

So this is why the user portrait above cannot help us make a strategy to sell cars, because it does not tell us what the main considerations are when users buy cars, whether it is price, brand, or other factors.

Another common user portrait mistake is describing a day in the user's life. This is because by observing the user's behavior for a day, you can only observe what he did, but not what motivated him to do it. Especially for something like buying a car, which has a long decision cycle, observing the user's life on a certain day is actually not very meaningful.

Here we need to explain a concept called consideration . What is consideration? It refers to the degree of thinking that users need to do before making a decision. It can be divided into three levels of consideration: high, medium, and low.

For example, buying a house is a big deal for most Chinese people. It requires careful consideration rather than making a decision casually. Before actually meeting the salesperson, the user may have already conducted many investigations, such as location, price, supporting facilities and other factors. At least 60% of decisions are made before seeing the actual product. This kind of product, after users have conducted in-depth self-analysis and can clearly know or even state the basis for their decision, is called the college entrance examination measurement product.

Low-consideration products, on the other hand, are decisions made by users unconsciously. For example, when scrolling through the news feed, users may decide whether to read this piece of news or that piece of news, or order this dish or that dish when ordering takeout. It may be difficult for users to explain the basis for this decision. We call this type of product a low-consideration product.

Zhongkaoliangdu products are somewhere in between the two. Before seeing the product, users may not know what they want specifically, and only have a rough range of choices. But after seeing the product, there will be a product feature that touches the user and makes the user make a purchasing decision. Afterwards, the user can also say what factors led to the purchase behavior. We call this kind of product a Zhongkaoliangdu product.

After we have clearly identified the type of product based on its own attributes, we can then know how to build our user portrait. For products that measure the college entrance examination and high school entrance examination, because users can clearly state the decisions that prompted them to buy, it is appropriate to use qualitative interviews to understand users' decision-making motivations.

For low-consideration products, user preferences are difficult to grasp and may change at any time. This type of product is not suitable for interviews, but should be modeled and analyzed through big data label statistics, analyzing user preferences in real time, and then using AI algorithms to intelligently recommend products that users like.

3. How to build a user portrait?

The user portrait of low-consideration measurement involves AI feature engineering, which I will explain in another article. This article mainly talks about how products that measure high school entrance examinations and college entrance examinations build user portraits.

We have just learned that for this type of product, the best way is to determine the user's decision-making motivation through qualitative interviews. The specific interview questions can be designed from the following five perspectives. These five questions are as follows:

1. Priority motivation: what causes users to need a product?
2. Success factors: what results do users want to get after using the product?
3. Known obstacles: what are the reasons why users do not choose your product?
4. The buyer’s journey, which is the decision-making process that users go through from idea to outcome;
5. Decision criteria: which factor do users value most when choosing a product.

Let’s look at a practical example. Qidian Academy hopes to build a user profile for its trainees, so it interviewed the trainees. The trainees described themselves as follows:

I graduated with a degree in chemical engineering and have two years of experience in traditional manufacturing. I hope to become a product manager in the hope of entering a sunrise industry, achieving higher career heights in the future, and becoming the person who designs solutions to user pain points.

After determining the goal, the next question is how to become a product manager. For students who want to change careers, it means a pay cut, zero experience, and having to work harder than others.

As someone who wants to change careers, how can I quickly become a product manager? First, you need to understand the abilities a product manager needs to have, and secondly, you need to know how to develop the corresponding abilities. You can find the abilities a product manager needs to have through recruitment websites, and then test yourself based on these abilities to see what abilities you are currently lacking. What I lack most at the moment is how to improve my abilities and produce a work that can be used for job applications.

Generally speaking, there are two ways: self-study and professional training. I tried self-study first but couldn’t find the direction. The learning cycle was long, and it was difficult to get guidance from experts. The most important thing was that it was easy to give up.

After comprehensive consideration, I decided to choose Qidian Academy's Product Manager Training Camp. First, there are product gurus from BAT who will personally guide me, so I can quickly and systematically build a product knowledge system. Second, the course focuses on cultivating thinking, and the product practical assignments also have teachers to personally guide and correct errors, which will eventually produce a portfolio that can be used for job hunting.

Through this interview, we can extract some key descriptions, such as the user's priority motivation is to achieve higher career heights, the known obstacles are concerns about lack of practical experience, no portfolio, and insufficiently systematic learning, and the decision-making criterion is the desire to quickly build a systematic knowledge system.

We matched the user's words with the five questions one by one, and summarized the user's description in general terms, thus forming a user portrait:

Then the subsequent product design and operation can be carried out around this user portrait.

The target user group can be identified as those who lack breakthroughs in their careers and hope to achieve higher career heights. When marketing to this user group, you can consider using the JD advertising space for product manager positions on recruitment websites as the delivery channel.

In the design of the advertisement, the emphasis is placed on two aspects. First, to eliminate the perceived barriers of users. Changing careers does not necessarily mean a pay cut. And through systematic training, a lot of self-study time can be saved. The second is to highlight one's own advantages based on the user's decision-making criteria: guidance from experts, focus on cultivating thinking, guidance from teachers of practical assignments, and the ability to produce a portfolio.

In the design of the course, we focus on the cultivation of product thinking and strengthen one-on-one guidance for students, so that students can produce works that can be used to find jobs, realize users' expectations for career changes, and truly create products that meet user needs.

4. How many user portraits do we need?

We produced a user portrait through the above interview case, but due to the different nature of the products, there will be many differences between users of some products, and the points of consideration are also different. For example, enterprise SaaS service products may involve many roles, and buyers, decision makers, managers, and actual users all have their own considerations.

Even if the user groups are similar in demographic structure, there are still big differences in the considerations of price, usage, etc. between the user groups. For example, the portrait classification of car buyers by Autohome, in which the parts marked in red are the key considerations:

Therefore, there is no absolute definition for the number of user portraits. Instead, they need to be classified according to the user's considerations. If it is found that the user's considerations are completely different, they can be classified into different user portraits. If most of the user's considerations are the same and only a small part of them are different, they can be merged into the same user portrait.

After constructing a user portrait through qualitative interviews, you can conduct quantitative research through questionnaires to verify the authenticity of the user portrait. Use the elements of the five dimensions of the user portrait as questionnaire questions and the key words of the portrait's consideration points as options for the questions. A quantitative survey questionnaire is designed.

By counting the number of returned questionnaires, we can tell which portraits are our key user groups, which portraits are secondary users, and which portraits are non-users.

5. What kind of users should be interviewed?

After we figured out how to make user portraits, the next step is to find the corresponding users for interviews. But what kind of users do we want to interview, what kind of users are the focus of our interviews, and what kind of users are not suitable for interviews? Here we make 5 summaries of user types, which are:

1. People who have considered your product and chosen you;
2. People who considered you but chose a competitor;
3. Those who considered you but decided to maintain the status quo;
4. People who didn’t consider you and chose a competitor’s product;
5. Someone who is considering you but hasn't made a decision yet.

For the first type of users, since they have chosen our products, we can call them customers. We can interview this group of people. Through interviewing them, we can know what we did right to make users choose us and strengthen our advantages. They can also help us find problems with the products and make us do better. However, for a product that measures the college entrance examination and the high school entrance examination, the decision-making cycle is usually longer and the repurchase rate is lower. They are not the focus of our interviews.

For the second category, people who considered us but chose our competitors, these users are the most valuable to interview. They can tell us what we did wrong that led to our competitors winning in the end even though we were clearly in the user's choice category. Only by facing failure can you know where your weaknesses are, and constantly iterate and upgrade from failure to win in the fierce market competition.

The third type of users considered us but decided to maintain the status quo and did not make any decisions. In actual work, this type of users is difficult to find and is not the focus of our interviews. Because at this stage, users’ needs may not be a pain point, but just an itch point. There is a lack of exploration process for their inner demands, and more factors are needed to pique users’ interest.

However, this only means that this type of user is not suitable for interviews and building user portraits at present. It does not mean that product managers can do nothing when users are at this stage. We can use the classic consumer behavior AIDMA model to identify the user's current decision-making stage, and then do some targeted stimulation to guide the user to the next stage. As for the content that does not belong to the user portrait, it will be omitted here.

The fourth type of users are those who choose competing products without considering us. There are two possible reasons for this type of users:

One possibility is that the channels where the users are located are not covered. We can discover these channels by interviewing users and asking them how they know about the competitors, so as to increase our publicity efforts. Another possibility is that the competitors may have formed some core competitiveness that we do not currently possess. When this phenomenon occurs, we must evaluate our own resources and decide whether to follow the practices of competitors to improve our own competitiveness, or to adjust direction and compete with competitors in a differentiated manner.

The fifth type of user is someone who is considering us but has not made a decision yet. We should not interview this type of people because conducting an interview at this time is likely to interfere with the user's decision and affect the final sales.

In addition to these five types of users, for products that measure the college entrance examination and high school entrance examination, the final decision usually involves multiple stakeholders. For example, in a typical product like children's education where the buyers and users are separated, the actual users are the children and the buyers are usually the parents. However, the children's grandparents, uncles and aunts may all become stakeholders in the decision. These people who do not directly purchase and use the products but have a great influence on the users' decisions are also the subjects we need to consider in our interviews.

6. User Interview Notes

1. Know your interviewee well and don’t fight unprepared.

Before the interview, it is best to understand the basic information of the interviewee so that you can ask targeted questions. In 1939, when the world-famous movie "Gone with the Wind" was being screened in the United States, the heroine Vivien Leigh flew to the United States from the United Kingdom. As a result, a new reporter asked: "What role do you play in "Gone with the Wind"?" The answer was: "I have no intention of talking to an ignorant person like you." This is a counterexample of not being prepared before the interview.

2. Prepare interview tools

The best interview tool is recording, because interviews usually last a long time. Recording the interview process makes it easier to organize it later. During the interview, you don’t have to be distracted by taking notes, but can focus on communicating with the user.

Of course, you need to obtain the user's consent before recording. If the user disagrees, it is recommended that you find a partner to participate in the interview together, with one person talking and the other recording, so as to divide the work and cooperate.

3. How to improve your interview skills

Imitating excellent interviews is the fastest way to improve interview skills. Here I recommend that you watch the interviews of famous speakers such as Lu Yu, Yang Lan, and Bai Yansong to learn how they open up topics and how to find the words that can further deepen the topic from the other party's answers.

In addition, you can also learn some cold reading conversation skills, which can help you quickly gain the trust of the interviewee and resonate with him/her.

Summarize

User portraits actually help us design better products by understanding the criteria by which users make choices, just as Nobel Prize winner in Literature Albert Camus said: Life is the sum of all your choices.

You cannot convince your users; only the choices they make can convince themselves.

author: CDA Data Analyst

Source: CDA Data Analyst