How to analyze user portraits in product operation and promotion!

User portraits are like labeling users, and each user is analyzed accordingly.

For a novice operator , user portraits may become one of the many headaches. This phenomenon is common among operators around 0-3 years old, but it is not unsolvable. When it comes to flirting with girls/men, I think everyone seems to be very good at it.

Is it possible to make user portrait as easy as flirting with girls? In fact, many things have their core logic. As long as you understand it, the rest are just some specific methods.

1. User portraits and flirting with girls/men

1.1 The process of flirting

Let me give you an example that I saw online before, which I think is very vivid. Suppose one day you are walking on the street and see a young lady with a great figure, good skin and good looks walking towards you, would you fantasize in your mind: "Wow, this girl is so beautiful"! Don't say you can't, unless you are not a normal man.

Suppose you are a single guy and think you are in good condition in all aspects, would you like to have a good talk with her about life and ideals? What would you do at this time? Stop fantasizing. You don’t even know her, how could she be won over by you the first time we meet? (The wording is inappropriate, but it’s the truth). The girl also fell in love with you at first sight, just ignore what I said...

So the best approach should be to start with mutual understanding. You may rack your brains to get the girl's contact information, then chat with her, and after gradually getting familiar with her, you can ask her out for dinner, movies , and a tryst in a small park. In this process, you will slowly get to know her, including but not limited to her age, occupation, family situation (single or not), hobbies, etc.

After you have a basic understanding of all of this, you may choose to confess (appropriate) or give up (inappropriate).

1.2 User portraits during flirting

In fact, if you think about it carefully, isn’t everything you do before making a choice equivalent to a user portrait? And without realizing it, you have already created user portraits from two major directions.

Appearance: She is a beautiful woman who meets your aesthetic standards;

Judging basis: Very good figure, fair skin, beautiful face, long legs...

Inner qualities: Gentle and virtuous, good at managing the household, positive and optimistic, the perfect choice for a wife;

Judging basis: The person has a pleasant and gentle voice, does not spend money on useless things, works hard and does not fall behind in study, has positive values, can do housework, cooks first-class dishes, etc.

Of course, it is basically the same when a young lady flirts with a man, but the basis for judgment is different. This is the user portrait, the one that gives you a headache. In fact, you have already mastered it. You may even use it better than I do. Can you still say that you don’t know how to use it?

2. User portraits in operations

2.1 User portrait, what is the difference between flirting with girls and work?

The user portrait in the process of flirting that we just talked about is actually logically the same as the portrait in our operations work. The only difference is that when you are in a relationship, you are only doing it for one person, but when you are in a product, you need to do it for a group of people.

2.2 The Essence of User Portraits

In my opinion, the essence of user portraits is to label users accordingly , and then analyze or judge user attributes through label classification. In this way, you can come up with different solutions for users with different attributes in your future work and implement them.

From the above description of dating, we can see that the girl’s age, height, interests and hobbies are all her labels, and we can use them to judge the girl’s preferences, personality and needs, and then make targeted moves or choices.

Let’s take an example of tagging which is common in daily life. I think everyone does it in their circle of friends , right? Think about it, do you also tag someone through his or her Moments? If someone always posts sad things on his Moments, would you label him as "a very depressed person"? You might even block or delete him in order to avoid being influenced by him.

After reading this, I believe you have a general understanding of user portraits. Now let’s talk about some specific ideas, processes and implementation methods of user portraits.

3. What is the user portrait actually depicting?

3.1 Explicit User Profile

User portraits can actually be divided into two dimensions. The first is the explicit user portrait, which is the appearance that is easily seen, such as age, region, interests and hobbies, etc.

3.2 Invisible User Portrait

Invisible user portraits are relatively more complicated and require in-depth analysis and understanding of users to clearly know, for example, in what scenarios the user uses your product, his usage preferences, etc.

4. Standard process for building user portraits

4.1 Data Collection

Data is the basis for building user portraits. Any user portrait without objective data to back it up is nonsense. I personally cannot imagine how your user portraits are made without data, and what problems can they solve?

In my work, my data sources can be roughly divided into two categories. One category comes from various industry analysis report materials, and the second category is what I get from users through various methods.

4.2 Data Analysis

After getting these data, the next step is to analyze and organize them. After processing, a data report is formed. Finally, I suggest that you try to visualize the data report. Although it may not be beautiful at first, just practice more and it will be fine.

Strictly speaking, at this step you still need to build a data model for data analysis, but considering that this is not something you will encounter in normal operations, I won’t go into details. Friends who are interested can look up what a data model is on the Internet.

5. Issues that need attention in the data collection process

In the process of data collection, the most direct method that best reflects user characteristics is to ask users directly, no doubt about it. But when asking questions, we must also learn to ask questions skillfully, otherwise not only will we fail to get the user’s true thoughts, but we may also lead you astray.

5.1 Clarify the topic

Before collecting data, it is important to have a clear theme around which to develop research.

5.2 Problem Setting

I often say that we must learn to ask questions, and yes, it is important to ask the right questions. Asking the right question is not just about asking accurate questions, but also about whether the user is telling the truth.

Some questions naturally have no real answers, so you need to use some strategies to go around them, such as a joke I saw recently.

An operator reported a product bug to the company's programmer . He did not say it directly, but asked the programmer if there was something wrong with his phone and why the product function was different from the design? Do you think the programmer will get angry at this time? Or would you think, damn, this is not a problem with the phone, it’s clearly a BUG.

Therefore, whether you are asking questions to your users or communicating in life and work, some questions really cannot be raised without any embellishment, otherwise it would be inhumane.

5.3 Questionnaire Collection Channel Selection

I have seen some young people who have just entered the industry. Their leader may ask them to do a questionnaire survey, so they just make a random link and throw it into a garbage group. The people in the group may have nothing to do with their own products. They just go in and send a red envelope and ask people to help fill it out. Worse still, some don't even say a word, just drop a link and run away. You don't know how tiring it makes me feel for them to see this.

First, try to find some users who are related to the product positioning to collect data;

Second, don’t just drop a link and disappear without a trace.

It is best to have in-depth communication with users before using methods such as questionnaires or user interviews. The results you get will definitely be beyond your expectations.

5.4 Learn to understand the user’s brain from the user’s perspective

Users also often tell the most truthful "lies". Many times they don't actually know what they really want to express. They may be describing 1, but what they actually need is 2. It depends on the people who collect the data to distinguish.

This is indeed somewhat difficult. In the words of Steve Jobs and Zhang Xiaolong , it means turning yourself into an idiot and thinking like a newbie.

VI. Conclusion

User portraits are actually the same as the process of choosing a boyfriend or girlfriend. We need to understand them from the outside to the inside. You can’t just look at the surface. Of course, that doesn’t mean you can’t do that. If you just want to find a vase, then just ignore what I said.

A complete user portrait cannot be hastily judged based on external information alone. You need to obtain external data through some methods, and then conduct in-depth mining and analysis, so that you can finally find a user portrait that is better or closer to the facts.

This is roughly the process of user profiling, at least that’s my current understanding. Summarize the final results and a fairly complete user portrait is ready.

Finally, I hope you have gained something. If you have anything to add, you are welcome to communicate with me!

Source: Operations Growth Notes