How to use questionnaires in app operations?

Note that it is “use”, not “do”. This article talks about my own understanding of App questionnaire surveys, which is suitable for those who are just starting out in App product operations . There is no specific explanation on how to design specific questions or how to write the analysis report. So, if you are a professional researcher and you see this, please leave a message and give me some advice. Just be gentle.

1. What is the use of questionnaires for App operations ?

1. It is used to assist in decision-making when an App project is established or before a major functional iteration of the product. For example, the boss wants to do payment and feels that the existing payment system has a low security factor. Then other decision-makers raise objections. At this time, it is necessary to conduct a market demand survey. Do existing consumers really think that existing payment channels are not secure enough?
2. It is used for grayscale feedback before the App is officially launched. For example, if the App is going to be launched next month, then if the team lacks confidence and for safety reasons, they will choose a small-scale grayscale release. At this time, the feedback from Grayscale users is particularly important.
3. After the app is launched, it is used to collect formal feedback.
4. It is used for verification after the product has major functional iterations.

2. Is the operations questionnaire not as professional as the user research questionnaire? Is it of great reference value?

Theoretically speaking, questionnaire survey is one of the means of user research and is used to assist quantitative analysis. Some companies will be equipped with user research teams, but such professional configurations are hard to come by, and professional user research has its limitations in actual combat, so in many cases the questionnaire survey work will still be undertaken by operations.

Both questionnaires have their pros and cons.

1. User research is more professional in questionnaire design and analysis, there is no doubt about that

Due to my major in college, I also took a professional course on questionnaire survey. I still vaguely remember that the teacher taught us that privacy questions should be placed at the end, verification questions should be set before and after the questionnaire to determine the validity of the questionnaire answers, and SPSS should be used to do correlation analysis and cross analysis, etc.

Speaking of SPSS, it is an amazing data analysis software. Really, anyone who uses it knows it.

2. Professional = long-term study + a small group of people are good at it

What I didn’t finish saying about the previous point is that, however, in the past few years that I have been working in operations, I haven’t used anything my teacher taught me (I feel so ashamed of my teacher here/(ㄒoㄒ)/~~), and I almost failed the SPSS course.

The reason I reveal my stupid past is to say that operators don't need to force themselves to become professional researchers. This requires long-term investment and talent. (Acknowledging that people are born with unequal talents will benefit you in your future life...)

3. Professionalism ≠ effectiveness. Professional research has its limitations in actual combat.

Even for a company capable of providing professional research and development, it is still a flat department that provides basic support. Just like the BI teams in some companies, they connect to all the vertical businesses of the company.

Then there is a fatal flaw: not having a deep enough understanding of the business often leads to some meaningless conclusions or even nonsense. For example, after a half-month investigation, the user research found that white-collar workers have higher purchasing power than students. I'm just wondering how I know this. If you live long enough, you'll see all kinds of research. . . Of course, it may also be that my level is low and I have never come across any useful research.

As for the questionnaire, operations have a unique advantage in this regard;

1. Operations understands the product and is more likely to hit the mark when asking questions.
2. Operations understand users, so the wording of copywriting is more likely to speak in human language. Being too professional can easily lead to being too official, and cold robotic language is not uncommon.
3. Operations understands the survey needs better and knows what points of analysis will be more helpful for the future of the product.
4. There is also an ugly but real function, just like data reporting, the same data can be analyzed into completely different conclusions, it depends on where you start, from which angle you analyze, and which point you report. You as an executive need not say more, you understand it. . .

3. Steps for App Operations to Conduct Questionnaire Surveys

In terms of steps, the questionnaire survey on the app does not have any special steps and is similar to other questionnaire surveys.

1. Needs understanding

What should operations investigate? Does the Product Manager have any questions? What does your boss want to know?

2. Questionnaire Design

(1) Question design:

• Don’t make your mobile questionnaire too long, and limit the number of questions to 10 or less.
• The title should be concise and clear, and do not use words that mislead users;
• Try to use multiple-choice questions instead of asking users to fill in long paragraphs of text;
• If there are interview needs or lucky draw settings, remember to ask users to fill in their contact information;
• The questionnaire platform I often use is WJX .

(2) Questionnaire distribution channels:

• App’s own banner, text links, push notifications and other online advertising spaces ;
• Some questionnaire platforms even support embedding questionnaires into other products, which is to use other people's headers and templates;
• Offline QR code channels such as coffee shop counter cards, campus roll-up banners, and roadside advertising stands;
• Group notification message channels such as mass emails and mass text messages are not very effective. A 5% return rate is considered good. The highest I have seen is 20%.

(3) Questionnaire collection:

• Never use paper questionnaires, because collecting and inputting large quantities of questionnaires is simply a living hell. I was tortured by 300 paper questionnaires for 2 days, and I just sorted them into electronic documents.
• If you use a questionnaire platform, you will generally see the number of responses, each person's questionnaire data, and a data chart for each question in the questionnaire creation background. The current questionnaire platform experience is getting better and better, and some even provide complex cross-analysis. If you have the conditions, you can buy an enterprise account and spend some money.
• Download the original questionnaire data and filter out some invalid questionnaires, such as those with many questions not filled in or those filled in randomly;
• I think the bottom line for the number of questionnaires is 50. Any lower will have little reference value. Be careful not to be misled.
• In order to increase the recovery rate of completed questionnaires, you can set up a lottery. Some questionnaire platforms support a roulette wheel or something after filling out the questionnaire. Wenjuxing has it.

(4) Questionnaire analysis:

• It is best to make a report. The so-called report should include the ins and outs, why the survey is conducted, how many responses were received, the time period, the problems to be solved, the existing data, data analysis, product recommendations, etc. This will make it easier for people who don't know the overall picture to read the analysis report, and it will also make it easier for you to archive and search it later.
• If the report can be made by combining user feedback and user interviews, it will be more valuable for reference if both quantitative and qualitative data are viewed together.
• Give the conclusion first, then display the data. Use less absolute values ​​and more percentages and nice charts. You are not a backend data porter of the survey platform. Show your ability to summarize and refine!

Let’s talk about the basic data analysis of operations next time, it’s quite troublesome. I want to use an analysis I wrote to demonstrate this, but I can't post it due to data confidentiality. Students who are interested in discussing it can send me a private message~~O(∩_∩)O

4. Do it and analyze, but don’t be superstitious

Like all auxiliary means, the results of the questionnaire are for reference only and should not be relied upon superstitiously. Just like BI data can lie and user feedback can be inaccurate, they all need to be identified and thought about. (Do you think this is nonsense? No! No! No! It’s really like this!!!)

The author of this article is @啤酒泡. It is compiled and published by (Qinggua Media). Please indicate the author information and source when reprinting!

Product promotion services: APP promotion services, advertising platform, Longyou Games