Hello, I'm Benny, the founder of Wings Class. Personally, my understanding of user analysis mainly includes the following three points:

1. Customer classification

For large-scale production enterprises, the products sold are definitely not single, but have certain product lines and corresponding price bands, so as to meet the needs of different customers and improve their competitiveness and thus increase their market share. If you only make high-end products, then you will lose the share of the low-end market; Enterprises generally have a clear product matrix, positioning of different products, target groups, prices, selling points, marketing strategies, etc. For example, some enterprises mainly sell a high-end product, but at the same time, they also sell related low-end products, just to compete against the standard and prevent losing market share in this market segment.

For customers who buy products with different quantity and frequency, if these raw data are available, we can subdivide customers through cluster analysis or RFM analysis, so as to adopt different marketing strategies for different user groups; For example, personalized service can be provided to customers with high purchase amount and frequency to improve the user experience; For customers with high consumption but low frequency, optimize content marketing and enhance user loyalty; For customers with low consumption, adopt appropriate marketing activities to increase the purchase amount and frequency of customers.

if an enterprise can obtain other basic information of customers besides sales data, such as gender, age, income, region, purchase time, etc., it can carry out more detailed customer classification and build a model to divide customers into different groups. For example, enterprise users are mainly concentrated in that age group, high-value customers are used to buying products at night, some customers are used to buying products on weekdays, and B-end customers may purchase products periodically, etc. According to these basic information,

Identifying the target users in the target market and classifying the target users will greatly reduce the marketing cost. For example, if you sell things that only the elderly will buy in a circle where young people gather, it will result in twice the effort, wasting money and effort and not pleasing.

2. Analysis of customer sources

In order to promote products and increase product exposure, enterprises will choose to advertise, and the advertising channels are not single. Offline there are traditional TV media, ladder media, railway light boxes, etc. Online there are Tik Tok, Aauto Quicker, Xiaohongshu, other apps, websites and e-commerce platforms for resource replacement. There are various channels for advertising, and if enterprises invest the same cost and energy in each channel, the cost will be very huge. Therefore, some enterprises will focus their resources on one or several platforms according to their core target customers. "Marketing pioneer" John Wanamaker said: "I know that half of the advertising expenses are invalid, but I don't know which half is invalid." If we know the source channel of customers, we can know which half of our advertising expenses are effective.

Take a simple example. For example, a store sold XX yuan this month, half of its customers saw Xiaohongshu planting grass, and one third bought it from Tik Tok. It will be easier if it is a catering business, such as a restaurant, the target customers are young people under the age of 35, and the product positioning is parity. It bought group purchases in Tik Tok and Meituan respectively, and distributed coupons on Little Red Book. Most young people choose group purchases when eating. According to different write-off platforms, merchants will get which platform has better drainage effect (considering the cost of delivery), and at the same time focus on this platform. However, most enterprises, especially large enterprises, will advertise online and offline channels at the same time, and customers will choose stores or online flagship stores to buy products according to their own information. For example, a customer first saw the advertisement of the subway, then opened the Tik Tok, and the opening screen was also the advertisement of this product. Then she went to Xiaohongshu, found this product, and then issued a coupon. Then he chose to go to the offline store to experience and buy the product. In this way, we can't get the specific customer source channel information.

In fact, the most difficult thing about customer source analysis is data collection. Personally, I think the best way is to do questionnaire survey, but it also means that the cost is relatively high. I remember one time when I went to an offline experience store of an electronic product, the salesperson asked me to fill out a questionnaire and give me an umbrella. One of the core questions was: "Where did you learn about this product?" Including when we were looking for a job after graduation, did the enterprise questionnaire also ask related questions such as: "Where did you learn about this recruitment information?" And so on.

3. Specific customer analysis

After classifying customers and customers' sources, we can conduct detailed customer analysis and research based on these two dimensions, including but not limited to: new customer analysis, customer acquisition cost analysis, customer conversion rate analysis, customer retention analysis, customer satisfaction and loyalty analysis, TOP N customer contribution rate, customer life cycle analysis and so on. Funnel analysis models (such as AARRR model, AIPL model, etc.) can also be considered in this part of the analysis. For example, through the specific analysis of customers, we know that there are more new customers on a certain platform, and the resources are less (the cost of obtaining customers is low), but the general purchase amount of customers is not particularly high (there are not many high-value customers), and there is no secondary purchase after using coupons to buy once (the customer loyalty is low). According to the characteristics of platform customers, corresponding marketing strategies can be formulated.

I hope my answer is helpful to you. Thank you for your praise and attention! If you have any other questions, please pay attention to "scyyhd" in WeChat official account, and there will be a special person to answer your questions as soon as possible ~ ~