User surveys are essential skills for operations, with two forms and five methods

A newbie who has no basic knowledge and switches to Internet operations often feels overwhelmed by so many knowledge points to learn and doesn't know where to start.

Copywriting , channel placement, data analysis , adjustment and optimization... We need to be proficient in each of these operational links, but if we want to apply this knowledge scientifically, there is also a very important prerequisite - understanding your users, which is what we often call user surveys.

Only by truly understanding user needs can we create copywriting, select channels, etc. in a targeted manner, thereby completing the entire operational closed loop scientifically and effectively.

However, in actual work, most people still have these four questions when conducting user surveys:

1. Thinking that user research is just sending out questionnaires

2. Not knowing the specific workflow of user surveys

3. Apply the template directly

4. Not knowing how to iterate and optimize the questionnaire

This article will solve the above problems in a targeted manner. First, the directory:

1. Specific forms of user surveys

2. A systematic approach to user research

3. Summary of points to note in user surveys

Let’s get started:

1. Specific forms of user surveys

There are two main forms of user surveys: 1v1 surveys and questionnaires. These two forms need to be used in combination according to different survey scenarios.

Let's take a look at the differences, advantages and disadvantages of these two methods:

Let's look at some specific application scenarios:

The one-to-one user survey method is more suitable for scenarios where you need to discuss some issues with users and explore some problems.

For example, we ask paying users or old users questions about writing structure, writing methods, and layout. You don’t have to ask closed questions. You can just discuss with users how the layout can be better. Do you have any suggestions? This requires discussion and communication, not simply options.

For example, if you are operating a public account , you can leave your personal contact information in the public account. When you get in touch with your readers, you can discuss with them what kind of content they would like to see. This is actually collecting user feedback to update the content.

The above are application scenarios of 1v1 user survey.

Now let's look at the questionnaire. Standardized questions are more suitable for the questionnaire format.

For example, if you want to get some user feedback from paying users or old users to guide you in the direction of adjustment, you can use some closed questions, such as, which of the following theme activities do you like best?

Such questions can provide you with guidance on the direction of your work.

For another example, for unfamiliar users, we can use the options in the questionnaire to find out how these users usually find certain content. Which channels do you usually browse? At this time, you can give users some multiple-choice questions through standardized questions in the questionnaire.

The above are two specific forms of user surveys, and these two forms should be used in combination.

2. A systematic approach to user research

User research is divided into five steps:

Step 1: Clarify the purpose of the survey Step 2: Design questions Step 3: Attract target users to participate in the survey Step 4: Organize and interpret data Step 5: Optimize the questionnaire

These are the five steps to scientifically conducting user research. Let’s take a look at them in detail:

01

The first step is to clarify the purpose of the investigation

The first thing you need to do to solve the problem is to clarify the purpose of the investigation. Only after the purpose is clear can you design questions around it.

First of all, we need to think about what problem we want to solve. Second, what kind of results can solve this problem.

The purpose of the investigation can be clarified by clarifying the above two points:

→For example, I am wondering what the identity of a paying user is?

→For example, a product has a function, but you don’t know whether users like this function.

→For example, what new content do you hope to see in the product?

…

This is a specific problem to be solved. Design relevant questions based on the purpose, then get feedback from users, adjust and optimize the product based on the feedback, and then verify whether such adjustments can solve the problem. This is a way to clarify the purpose of the investigation.

Now that the purpose of the survey is clear, the next question to consider is who should be surveyed. A specific purpose corresponds to a specific group of people.

→For example, if you want to know the identity characteristics of paying users instead of all users, this would be a targeted survey.

Therefore, think clearly about the purpose of the survey and also think clearly about to whom the questionnaire should be sent.

Here are five common survey purposes:

1. Determine the user’s identity

Regardless of whether the product is in the embryonic stage or the development stage, users are changing, and the identity of users is also changing. Therefore, determining the identity of users is an investigation that always needs to be done.

2. Determine user needs

Because user identities are changing, user needs are also changing. For example, a user used to be a college student, but now he is a working person and may become an entrepreneur in the future. The demand for products is different at each stage and for each identity, so I have to continue to investigate how user needs have changed.

3. Product functions

Whether it is the operating experience or whether the product can solve user problems, it needs to be continuously improved.

4. About the market

This is about traffic , what content do users prefer? From which channels do users learn about the product?

5. About Operations

For example, when organizing an event, you can collect information about the forms and themes that users like.

02

Step 2: Design the problem

A survey form can actually accommodate multiple survey purposes. Here, let’s assume that I want to solve two survey purposes:

1. Want to know the user profile of paying users?

2. What are the pain points of paying users? Why pay?

What information do you need to know to solve these two problems?

To understand the identity portrait of paying users, we need to collect information such as age, occupation, current salary, region, education level, etc.

Please note that user portraits should be obtained in a targeted manner based on the characteristics of the product. For example, for maternal and infant products, it is very important to know how many children the user has.

Therefore, even when investigating user portraits, you should adjust the information you want to obtain based on the characteristics of the product.

Want to understand what the pain points of paying users are? Why pay? We need to know what problems users hope to solve after purchasing the product? Which product feature do you value most? Which features do you hope the product will improve?

After understanding the problems that users want to solve, we can put the message that our product can solve this problem in a prominent position during promotion .

After understanding the product features that users value most, we can analyze and explain in more detail how this feature provides the service.

This is an example of using survey results to actually solve a problem. By breaking down the problem to be solved, you can find the specific information you need to know in order to solve the problem.

Of course, the questions listed above are by no means complete. You can improve them when you optimize the form at the end. You can add new information that you want to know during the optimization process. This is the iteration and optimization of the questionnaire.

Let's look at a style issue that you should pay attention to when designing a questionnaire:

• Header

What you need to pay attention to in the header part is: you need to clearly write in the header how many questions there are in the questionnaire and how much time it will take the user.

If you are conducting a one-on-one survey, you can tell the user that the survey will take about 15 or 20 minutes, so that the user has an idea of ​​the survey. This way, the user will be more receptive to the survey. The significance of doing this is that it will be more user-friendly.

Some points we should pay attention to when designing:

▶ 1. The number of options depends on your needs.

▶ 2. If you can answer multiple-choice questions, try to answer them.

▶ 3. Set every answer to be mandatory.

▶ 4. Add "Other" option to give users more choices.

▶ 5. Provide targeted prompts for problems:

(▲Question prompt)

▶ 6. Every question must have a serial number.

▶ 7. Ask closed questions:

(▲ Closed questions)

▶ 8. Establish connections with users and leave communication opportunities:

(▲Leave an opportunity for communication)

▶ 9. Note: Set the form to "limit each person to fill in once"

▶ 10. Inform the user that the form has been filled out successfully:

(▲Successful filling prompt)

03

Step 3: Attract target users to do surveys

What is the prerequisite for attracting target users to take the survey in the first place?

The prerequisite is to gain the trust of users, and gaining the trust of users is something that requires a lot of time and energy to operate.

The main way to build user trust is to create professional and helpful content to achieve the effect of becoming friends with users:

▶ If you can become a knowledge expert in a certain field, you can obtain this part of vertical traffic. If users recognize your content, they will take the initiative to establish contact with you.

You can embed contact information in this content. After you contact the user, you can exchange some relevant knowledge with them. You can also conduct surveys with them (one-on-one surveys or questionnaires).

At this time, your relationship with the user is mutually beneficial because you are useful to the user. It is an equal relationship.

▶ If a survey form is suddenly thrown out in a group, and this form is actually of no use to the user, why would the user help you fill it out?

Note: Don’t be too utilitarian when gaining user trust. Really understand the users and become friends with them.

04

Step 4: Organize and interpret data

On the one hand, interpreting and organizing data is to solve problems. On the other hand, these data can allow us to draw some objective conclusions about user behavior, user identity, and user cognition.

For example, conducting a user survey has two purposes: obtaining the identity portrait of paying users and obtaining the pain points of paying users.

Let’s look at the data we collected:

1. Identity portrait of paying users and its extension

Based on the above 5 points, many questions can be raised, such as:

age:

67% are between 23 and 26 years old. What is the identity and living status of users in this age range? What kind of life do they live? Are they satisfied that they are satisfied with themselves? If you are not satisfied, what is it that you are not satisfied with? These dissatisfactions will be the real pain points for users.

Profession:

Traditional industries account for 80%. Which positions have the most users? As we gain deeper insights into the issues, we can create more targeted content.

Salary:

How much does it mean to earn 4,000 per month if you earn less than 50,000? Or 3000? Or 2000? This question is related to the user's living conditions.

area:

First-tier cities account for 90%, so what is the proportion of the four first-tier cities respectively? What is the proportion of Guangzhou? What is the proportion of Shanghai? These specific data will affect which province or city we should focus our money on when placing advertisements in the future.

Education:

Bachelor degree and above account for 70%, so are they specifically ordinary third-tier colleges? Or a 985 or 211 university?

2. User pain points and their extensions

Purchasing a course to solve a problem:

If you hope to reach a career-changing rate of 80%, then you should increase the amount of career-changing content, including promotional content and product display content, all of which should be closer to the needs of target users.

→The extended question is: In addition to changing careers, what other problems do users want to solve? You can sort these questions and know what the product should focus on.

Most important features:

68% of people think it is practical operation, so the product page needs to emphasize practical services.

→The extended part is: which function is the least important to users? If this function that users do not value consumes a lot of manpower and material resources, it should be cut.

The ability you most want to improve:

42% of the users hope to improve their copywriting skills, so we should enrich the copywriting courses so that we can understand specifically where users are confused in the existing copywriting courses.

→The extended question is: What other abilities do you hope to improve besides copywriting ability? Sort the abilities that users want to improve so that relevant courses can be created in a targeted manner.

The above means that interpreting the survey data is to apply the data results in hand to actual work, and organizing the survey data is to raise some new questions to help product iteration.

05

Step 5: Optimize the questionnaire

It is impossible for a questionnaire to be perfect at the beginning, and we certainly need to iterate it. Iterating the questionnaire means making the questions more detailed and more suitable for application in work.

Optimizing the questionnaire is divided into two parts: enriching existing survey questions and developing new survey questions.

1. Enrich existing survey questions

Here is an example of a paid user profile for the purpose of the survey:

• About age

The original questionnaire did not have an option for over 35 years old in the age column, but a considerable proportion of the feedback we received was 42, 45, and 38 years old. Based on this feedback, we added the option for over 35 years old. This is a detail we are optimizing.

• About occupation and region

We can turn fill-in-the-blank questions into multiple-choice questions. Based on the first round of data, we know which occupations and regions have a high proportion. When we optimize, we can directly let users answer multiple-choice questions because the threshold for answering multiple-choice questions will be relatively low.

• About salary

In the first round of data, the option "below 50,000" accounts for a large proportion. When optimizing the questionnaire, the salary options can be further subdivided. For example, below 40,000 and below 30,000 can be added. We can better understand the user's specific salary status. Such optimization makes the data clearer.

• About academic qualifications

We have added some new questions, such as what type of university. This optimization can let us know the approximate level of most users. We can also know which type of school students are more likely to become our paying users.

2. Develop new survey questions

Expanding new survey questions means generating new thoughts, new questions, and new survey objectives through the first round of data, and then creating new questionnaires.

Example:

→Are you satisfied with your life now? What are the reasons for being satisfied, dissatisfied, or making do with your life?

This is to analyze the users' real living conditions and their psychological states, so that we can make targeted adjustments to products, content, wording, etc. This is a new survey purpose.

→For example: Which school did you graduate from/attend? What is the monthly salary when you become a paying customer? These are new questions with new investigative purposes.

Surveying users’ schools may be trying to find new promotion channels , and surveying users’ monthly salaries may be trying to find a reasonable price range for new paid products.

As mentioned above, a questionnaire is not a static thing. First, you need to think clearly about the purpose of the survey, then do the survey and let the survey results solve practical problems in your work.

3. Summary of User Survey Points

Earlier we talked about the specific forms and methods of user surveys. Now let’s summarize the points that should be paid attention to in user surveys:

1. Always design questions based on the purpose of the survey

This is a problem that many people encounter. Without thinking clearly about what problem they want to solve, they start to design the problem directly. This is actually putting the cart before the horse.

You should think clearly about what problem you want to solve, and then design related questions based on this purpose. This is the correct way to do things.

2. Think clearly about the target audience of user research

If the questionnaire is designed to investigate the behavior of paying users, then paying users are a very clear survey subject.

If you just find some random people to fill out the questionnaire, the data is actually not directly helpful for your work. First, the groups of people do not match, and second, it may increase your work costs.

At the beginning of the user survey, think clearly about the subjects of the survey, and then find these people in a targeted manner.

3. Try to ask closed questions

What are closed questions? For example, for multiple-choice questions, users only need to choose one option from three options, and the operation cost is very low.

If the questions in the questionnaire are very open, for example, how to do WeChat marketing ? This topic could actually fill a whole book. It will confuse users because the question is too broad and difficult to answer, which will affect your survey results.

4. The larger the number of survey samples, the better

The purpose of user research is to solve problems, not to collect as much data as possible. Don't put the cart before the horse.

Trend data is enough to illustrate the problem. When 18 out of 20 users give the same answer, there is no need to survey the 21st user because the answer is already very obvious. This kind of data with obvious bias is enough to solve the problem.

5. The prerequisite for obtaining reliable survey results is trust

After finding the target users, we gain their trust by creating helpful content, identifying them, and adding them as friends, and truly provide value to them to achieve a mutually beneficial relationship.

6. Survey questions need to be constantly optimized

The details of the questionnaire need to be adjusted based on the collected data, whether it is adding new questions or raising new ones, the questionnaire should always be iterated and optimized.

For a specific product, the target users are changing, and their needs are also changing, so user surveys need to be conducted continuously.

7. The two survey forms need to be used together

As mentioned in the first part: One-on-one surveys are more suitable for situations where divergent thinking and problem discussion are required.

Questionnaires are more suitable for collecting information and can be used in scenarios where standardized questions can be asked.

These two survey methods actually meet different purposes and can be used together to make the survey results more real and objective.

As mentioned above, mastering user surveys can not only help you pass the interview smoothly, but also be used in daily work to solve some practical problems.

We don’t want you to apply a so-called questionnaire and show your interviewer or superior a seemingly impressive survey result to pretend that you are proficient in user surveys.

Instead, I hope that students can truly master the mindset of user surveys and be able to apply them skillfully in their work. Then, user surveys will become a skill that you can rely on to join Internet operations and even get a raise or promotion!

Author: Nuanshi Dry Goods Group, authorized to publish by Qinggua Media.

Source: Nuanshi.com (ID: nuanshi100)