Who-who is the user, when-when, where-where, why-what is the attraction and what has been done.
For different product types and different variable complexity, the connotation and extension of the scene are different. Among them, the scene of e-commerce products is more complicated, which is manifested in many variables, many user types, many demand types and latitudes, and many use paths. At the same time, as an experience design method, scenario analysis is often mentioned and used in e-commerce products. Therefore, taking Meituan hotel reservation business as an example, this paper introduces the steps and methods of scene analysis and experience optimization in products.
Experience optimization driven by scenario analysis can be divided into the following five steps:
The last link belongs to the category of design stage, so I won't expand it in detail. Here I mainly introduce the first five parts closely related to scene analysis.
According to the variables contained in the scene, firstly, the variables are structurally decomposed through the 4W dimension:
Taking hotel reservation as an example, these four elements can be broken down as follows:
In the scenario of hotel reservation, the dimensions of users include two categories: subscribers and peers.
1) Booker: Users have many dimensions, including gender, age and occupation. For commercial products, the consumption power of users is the core. Around this angle for the time being, according to the portrait tags in the background, the two core groups are extracted: students and white-collar workers. This dimension can help to judge people's consumption habits and preferences.
2) Peer: Peer mainly affects the predetermined room type and appeal, and the relationship between peers has the greatest influence on appeal, so it can be divided into several categories according to the relationship: loners, couples/spouses, relatives/classmates, parents, elders and colleagues.
The dimension of time also includes two categories: scheduled time and actual check-in time.
1) Scheduled time: the dimension of time can be divided into years, months, days, hours, minutes and seconds according to granularity, and working days, weekends and festivals according to particularity. The dimension here is: the scheduled time in advance. Because users decide in advance or on the same day when booking, it determines their demands for efficiency.
2) Check-in time: For tourism products, the core influencing factors of staying at different times are daily life, seasons or festivals. Traveling at different times usually has different goals. However, what we are talking about here is the improvement of the basic experience, and the check-in time usually affects the category and price of the hotel itself, so the travel time is not considered here for the time being.
According to the relationship between the scheduled location and the actual location, it can be divided into local and off-site locations.
The core element is the purpose of travel, including dating, traveling, vacation, business trip, temporary residence and other scenes. The objective variables can be obtained according to the conclusions of industry reports and historical user surveys.
To sum up, the following scene elements can be obtained:
Tips:
1. The disassembly of components shall conform to the MECE principle-"no weight and no leakage".
2. According to the purpose of the analysis and the stage of the product, it is decided which variables should be adopted, and not all variables need to be considered.
For example, in the above case, the variable of check-in time is more to guide the operation of the hotel, such as summer vacation, National Day and other festivals, and corresponding activities can be planned around the demands of the scene. The analysis here is mainly aimed at the basic experience, so this variable is not included.
From the product development stage, the variables used in the initial stage and exploration stage of the product can be more focused, such as user groups, travel purposes, etc., and the core key scenarios can be analyzed first to help the demand mining in the vertical market. For products in a relatively mature stage, the demand gradually covers more comprehensive and subdivided scenarios, covering more people and more diverse needs, so the scenario analysis at this stage can cover the variables as comprehensively as possible.
Hundreds of subdivided scenes can be obtained by cross-combining the variables of the scenes, but it is obviously impossible to completely cover all the scenes when building the experience. So these scenes need to be focused on, focusing on high-frequency scenes.
In this link, it needs to be clarified by combining quantitative means. The usual practice is to define high-frequency scenarios through order distribution. Specifically, you can select the latest order, check the background data, and clarify the source of the order, including background user tags, local or off-site, hotel type, scheduled time and other information. For variables such as travel purpose and peers, it is difficult to see them through data, so it is necessary to combine questionnaires to clarify the scenes covered by these orders.
Through the above data analysis, we can list the most core scenarios.
For example, the six core scenarios in hotel reservation include:
After defining the core high-frequency scenes, we can further clarify the user's demands in each scene. There are two methods, one is interview and the other is questionnaire.
1. Interview-Summarize the types of appeals
Interviews can help us sort out the specific cases that users booked in various scenarios, and dig deep into users' demands and the reasons behind them. Through interviews, we can summarize and refine the types of users' needs and make clear the typical types of consumption habits.
For example, the six core scenarios of the above accommodation correspond to the following specific situations:
Through interviews, we can understand the factors that users pay attention to when making consumption decisions. According to the interview conclusion, users' consumption habits can be summarized as follows:
1) Convenient and fast: the location is convenient, the price is within the budget, and it is clean and hygienic.
Corresponding scenes: business trip, friend visit, local date.
2) Cost-effective type: close to scenic spots/business districts, overall comfortable and cost-effective.
Corresponding scene: friends travel
3) Ambient atmosphere type: It is expected that the ambient atmosphere will be good and distinctive.
Corresponding scenes: couples dating, couples vacationing.
4) Service facilities: catering and entertainment facilities are expected to be comprehensive and have a sense of quality.
Corresponding scenes: parent-child vacation, couple vacation
Around these typical requirements, we can deeply analyze the function and experience of products and explore the possibility of optimization.
2. Questionnaire-quantitative decision-making habits
Through the questionnaire, we can more accurately and comprehensively understand the factors that users pay attention to when making consumption decisions, and the degree of attention of different factors. Before doing research, the first thing to do is to split the decision-making elements. For commercial products, the dimension of splitting can be expanded according to Maslow's demand model:
The core elements of hotel decision-making include:
1. Basic elements: price and location
2. Preference factors: environment, surroundings, catering, facilities, services and brands.
Each element will contain specific variables:
According to the price, from low to high, it can be divided into cheap, cost-effective and high-grade.
For example, according to the different scenes, the location can be divided into the vicinity of the location, the vicinity of landmarks and the vicinity of scenic spots.
Through the above investigation and analysis, we can finally get such a general scenario analysis table:
With user scenarios and requirements, we need to analyze the internal usage path of the product, so that we can walk around the path and locate the experience problem.
The behavior path can also adopt a combination of quantitative and qualitative methods. First, from the data dimension, we can see the click distribution of each page entry of the core process, so as to summarize and sort out the core path within the product. The core path of Meituan Hotel is as follows:
Finally, we can make different scenarios and requirements correspond to the corresponding product paths, conduct experience traversal and walkthrough, and locate and summarize experience problems.
It should be noted here that a scenario may have multiple requirements, and multiple requirements flow back to multiple paths. Therefore, in the process of walking through, we need one-to-one correspondence to be more comprehensive.
Finally, through the scene walkthrough, the problem of insufficient matching of user needs in the product process is discovered, and then the opportunity point of experience optimization is discovered.
1. Before experience optimization, it is the most basic work to comprehensively sort out and analyze user scenarios, which can be continuously applied to the long-term work of product optimization, so building a basic user scenario library is the core work of the designer.
2. The scene here is relatively coarse-grained, which can guide the whole process, large-scale function optimization and opportunity point mining. However, if it involves specific functions, such as search optimization, it is necessary to further subdivide the scene of this function, such as analyzing the brand and location of users' search.
3. The scene analysis here, although combined with interviews, questionnaires and other methods, is essentially a qualitative way based on user subjectivity. In the application process, it is necessary to combine more behavior data and traffic data about functions or pages to make comprehensive design decisions.
As a matter of fact, this analysis method can't guarantee to be comprehensive. Although the product development is more mature, the dimensions such as user groups, scenes and paths will be more and more subdivided. In this case, the big data algorithm should be more accurate and efficient.
The current way can help designers find problems and locate the lack of product functions or interaction layers at this stage. For example, through this exercise, we may find that couples need to screen environmental facilities when booking hotels to help them find romantic hotels. So we can provide such a filter functionally.
However, if we look at finer particles, we need to know what kind of environment they need, whether it is pink romance, pure white simplicity or cool technology. Different people have different tastes, and the displayed content level labels should also be different.
Therefore, when the function and structure of the product are perfect enough, such pre-scheduling can bring us limited help, and in essence, we still have to return to intelligent recommendation and thousands of algorithms. For example, the waterfall of Taobao, even if we can traverse all user scenarios, the final content strategy still needs to follow a systematic and perfect data algorithm model to truly meet the demands of each segmentation scenario.