1. Introduction
As we all know, sending coupons to users is an important means for merchants to increase their business volume, and it has been tried and done. The psychological appeal of users who like coupons is also very simple, that is, they like to be greedy for small gains. So, if the boss gives you a question and asks you to build a preferential system from 0, how would you start? My idea is to determine what the target is for issuing coupons, which customers will issue coupons, in which scenarios will issue coupons, and what coupons will issue. If you answer these four questions clearly, the idea of the preferential system will be clear.
The following picture can quickly understand the logic of building a preferential system:
Next, the logic of this system will be explained according to the user's life cycle, identifying marketing sensitive customer groups, and designing reward system.
2. Dividing user life cycle
The first step in building a preferential system is to divide user life cycle. The standards for dividing life cycles of each company and each product may be different, so I will only explain the general standards here.
introduction period: market customer acquisition stage, have not registered yet or have registered yet to experience the main process. For this customer group, the goal is to increase the registration rate and increase the conversion rate of the main process
Growth period: the main process has been experienced and belongs to the activated user. For this customer base, the goal is to increase the repurchase rate
mature stage: frequent use of products/repurchase products. For this customer group, the goal is to increase the repurchase rate and unit price of
dormant period: there was a period of time when the product was not used. For this customer base, the goal is to reduce the churn rate
churn period: no product is used for a long time. For this customer group, the goal is to reduce the churn rate
or above, which is the overall idea of dividing customer groups, but should I send coupons to all users? Surely not. We need to identify the marketing-sensitive customer group that will be lost at at each stage, and this is the real target user to issue coupons. is about to lose , which is because the user does not follow the preset path and leaves the process or product at some nodes; marketing sensitive is because the user is willing to "buy", but the user believes that the price is too high and will pay as long as the price drops. So, how to identify these users?
3. Identify marketing sensitive customer groups that are about to be lost
Method 1: Users use real-time operations to judge
How to recognize : Generally, based on the behavior of clicking back/clicking cancellation of each process node. This judgment method is very common and is suitable for products with fewer early users, because at this time, the data accumulation is relatively small and it is difficult to do intelligent identification.
Life cycle stage : Generally for users in the introduction and growth stage
coupon method : When the main process exits, pop-up window will pop up and issue coupons
Method 2: judge according to time/frequency:
0 How to identify : Users who have not operated for a long time/low operation frequencyLife cycle stage : Mainly for users who are dormant and churn period
coupon method : Users who have not used the APP for a period of time will send coupons to the user silently and send text messages/push notifications. The disadvantage of this method of issuing coupons is that it is a bit lagging and the user response rate is low, because it is difficult for the user to wake up
Method 3: Users actively receive coupons
How to identify : Give all users the opportunity to receive coupons. Users who receive coupons can be judged to be marketing sensitive. This method is not very controllable, so the threshold for using coupons is generally set relatively high and the discount is relatively small. The following two pictures are the coupon module of Taobao product details page. You can feel it:
Method 4: Intelligent model based on big data recognition
How to identify : Establish an intelligent recommendation model based on user's long-term behavioral operations, order data, user portraits and other data.The threshold for using this method is very high. First, the company requires an algorithm team, and second, the product user data must reach a very high level, otherwise artificial intelligence may become an artificial intelligence retarded.
Life cycle stage : Generally suitable for growth stage, maturity stage, dormant period, loss stage
coupon issuance method : There are real-time coupons based on scenarios, and there are also background silent coupons
Summary : In addition to the intelligent model, the other three methods are generally used in combination. Users return to the issuance of coupons in real time for users in the introduction and growth periods; users are not allowed to issue coupons for a long time for users in the dormant and loss periods; users actively receive coupons for all users, which is suitable for national promotion activities or activities that increase customer unit price. The customer groups and scenarios targeted by the three methods are different, and combined use can maximize the value.
4. Design a discount system What prizes are given to users by
1. Coupons related to products: common e-commerce products include full discount coupons, deduction coupons, exchange, purchase and gift; common loan products include interest-free coupons, discount coupons, repayment bonuses, deferred repayment coupons, etc.;
2. Rewards that the public generally likes: cash red envelopes, phone bills, merchant coupons, physical goods (such as iPhone), etc.
Author | Xiaohu Does not brag about product concepts, only talk about product dry goods
