1. Collect the feedback from background users and sort it out to understand the frequency and severity of problems.
2. Watch the video of the user's class, interview relevant personnel, and restore the problem site; At the same time, user portraits are established and subdivided to select the most representative users.
3. Establish the user trip chart and analyze the demand points by combining the above data mining.
4. Convert user requirements into product requirements, prioritize and put forward iterative suggestions.
Background user rating+evaluation content, course video, interviews with relevant personnel.
Judging from the original comment data collected this time, after filtering out pure praise, pure emotional bad comments, brush comments and meaningless comments, there are still 2262 effective user comments.
From reviewing the statistics, we can draw the following conclusions:
All user comments are divided into five categories: system BUG, system optimization, courseware optimization, teaching optimization and new requirements. As can be seen from the figure below, more than half of the feedback occurs in the teaching process, which is a problem that most users are very concerned about. In order to restore the scene of the problem, the author tracked the teaching videos corresponding to the relevant comments, interviewed the teachers and customer service staff who were teaching at that time, and analyzed some representative problems with this information.
In the process of watching the course video, the author also noticed that even children in the same class (the same age) have many attribute differences. These differences will directly or indirectly affect their classroom experience. For example:
Obviously, it will be very tedious and ineffective to build a user portrait for each different attribute, and we will eventually design t a based on a typical user. Therefore, the author's idea is that we can list the more important dimensions of user characteristics, and then build a user portrait with the most coverage as a typical user according to these dimensions.
According to my observation, try to list the user characteristics in the following dimensions:
In fact, after careful observation, we can find that the two poles of these dimensions can correspond to "users who are the easiest to complete learning tasks" and "users who are the most difficult to complete learning tasks" respectively. This speculation can also be verified in the course video.
At this point, we can construct a user, Jerry, who will be a user who is difficult to complete the learning task smoothly. If even he can finish the whole teaching process efficiently and smoothly, then others will be easily satisfied.
We try to integrate the typical user portrait with the previous user feedback questions in the form of user journey map, so as to understand the behavior and feelings of users when using products more intuitively, which is conducive to demand mining.
As can be seen from the above figure, in the whole product experience process, the overall emotional experience of users is basically progressive. However, in the course of class, due to the long time-consuming and complicated process, the actual users become children, and they may encounter more obstacles when interacting with the equipment, especially when they can't express their difficulties well. When seeking help, they are prone to great frustration, which directly affects the effect of class, and also makes parents and users dissatisfied with our products and services, thus affecting the conversion rate.
Therefore, it is necessary to focus on the needs of users during class. In addition, through the above analysis, we can find that the pain points in many classes come from the lack of evaluation of students' actual mathematical ability before class, and only recommend courses according to their age, which leads some students to match courses that are not suitable for their current level, thus causing many follow-up problems.
After all the above links, we have gained a lot of user demand. When faced with numerous demands, setting up tasks immediately without thinking will only lead to the waste of manpower, material resources and time costs, and will also greatly dampen the enthusiasm of development partners. Therefore, before putting forward action plans and suggestions, it is necessary to prioritize the requirements, and it is imperative to find the most urgent problems to be solved at present.
Generally speaking, requirements can be sorted by the number and frequency of users, development difficulty and effect.
Large number of users+high frequency is the most important and urgent problem, which needs to be solved first.
Large number of users+low frequency is a concern of users and should be ranked second.
Small number of users+high frequency of use is a concern of a small number of users, and key issues can be solved first before consideration.
Small number of users+low frequency is the last requirement, that is, it is not urgent or very important at present. You can record the shelving first and then observe whether it needs to be adjusted.
Quick effect+low development difficulty as the function of early priority and rapid iteration;
Quick effect+high development difficulty as a function of growth iteration;
As a function of moderate innovation, low effectiveness+low development difficulty;
Low efficiency+high development difficulty as a function that may be an opportunity.
The product is currently in the growth stage. After accumulating a large number of stable users, in order to retain these users, we should constantly optimize the core user experience and gradually reduce the user turnover rate. That is to say, the demand of quick effect+high development difficulty and quick effect+low development difficulty is the main demand, and then the demand of low efficiency+low development difficulty and low efficiency+high development difficulty is concerned.
Priority is divided into five levels: high, high, general, low and low.
When faced with complicated user feedback, we need to have a set of methods to simplify the complex, eliminate false requirements and extract real requirements as the focus and direction of product iteration. In this process, we will get to know users better and better, and at the same time, our product positioning will become clearer and clearer. In the early stage of the development of online education products, the user's class experience is the core focus of optimization. Only by providing enough user value in this link can the product develop continuously and stably.