Practice: How to implement member repurchase in the 618 event?

This article mainly analyzes how to conduct member repurchase based on Tmall’s 618 event? With the retail domestic market segment cosmetics companies as the background, let’s take a look at the article~

This article seems to be a little late, but in any case the biggest e-commerce festival in the first half of the year has ended. The author is responsible for the brand's omni -channel CRM work. There are three major e-commerce platforms: Taobao , JD.com , and the official self-operated mall, as well as some scattered small platforms.

This year, the entire promotion rhythm of 618 on Tmall was lengthened. Compared with the previous 3 days, this year it lasted from 601 to 620, basically the whole month of June. Both merchants and customers felt physically and mentally exhausted.

From the perspective of member operations , big sales are very good wake-up nodes. Let’s take this event as an example to talk about the specific big sales wake-up strategy and execution. Of course, there is also daily maintenance, which we will talk about later.

Background: A cosmetics company that retails domestically produced cosmetics in the niche market.

Number of members: 2.4 million. The brand is in a rapid growth stage and has no money for high-profile advertising. The rapid development in the past two years has been achieved through meticulous operations.

Here we are only talking about member repurchases. If you want to hear more about attracting new members , please speak up. Welcome to discuss. Although the industries and models may be different, the logic is the same. You are not to be argued.

Generally, when we do activities, the complete membership work is divided into the following modules:

Identify your goals

First of all, the sales targets for each month and each major promotional event are usually determined at the beginning of the year. Of course, adjustments may be made during the actual implementation due to various circumstances. After getting the sales target for the overall activity, start breaking it down.

The main indicator of the membership side is to get the number of members who repurchase:

The performance share of new and old customers is generally based on the three-year historical data of previous promotional activities, and some variable factors will be taken into consideration, such as: this year's external investment, resource status, etc. Work out a more reasonable estimated contribution ratio of new and old customers.

The same goes for the average order value: whether to use a low average order value to attract consumers or to maintain the average order value, these are issues that require multiple departments to sit down and communicate and confirm repeatedly in the early stages. This directly affects the formulation of the main product structure, as well as the division of promotional positions and activities. There are quite a lot of details involved.

After the contribution ratio and the number of new and old customer orders are determined, the promotion side and the membership side have basically received their own goals, the number of new customers and the number of repurchasing members. Of course, the whole process is a collaboration among everyone, because in fact the main products and level divisions for old customers also need to be worked backwards, and the same applies to new customers.

Decomposing the current situation

After getting the target, of course, it is not just a matter of the number of repurchases. It needs to be broken down and subdivided. Here is a table to share:

• The first column is the member life cycle segmentation: new members, active old customers, high-recovery and pre-lost customers, low-recovery and pre-lost customers, churned members, trial members, and potential members;
• The second column is the total number of people in each stratum;
• The third column is the membership level in each life cycle: ordinary member, senior member, VIP member, supreme VIP member;
• The fourth column is the specific number of people in each level, followed by the percentage of the number, the number of repurchasers, the repurchase rate, the average repurchase price, the contribution amount, and the percentage of the contribution amount.

There are three years of data in total, 2016, 2017, and 2018, because we need to compare and see the changes.

Finally, we will estimate the repurchase rate, average order value, and overall contribution of all segments in 2018.

What we have concluded here is the current situation. For example, according to our estimate, the performance of old customers is to repurchase 200,000 people and generate 80 million yuan in performance, but our goal is to get 100 million yuan. How do we get the remaining 20 million yuan?

Set improvement goals and strategic initiatives

Ps: Let me explain, it is not an estimated 80 million. It does not mean that old customers will come back on their own. The original estimated value is also the repurchase rate calculated after taking many actions. It’s just that it is still not enough and needs to be improved.

In addition, regarding the user segmentation model, you can also use other combinations such as sku, user source, etc. instead of life cycle + level, as long as it meets the brand attributes and can be used.

Okay, now let’s talk about the remaining 2000w.

Adjustments can be made to the original tiered repurchase rate and orderability, but this adjustment cannot be made casually, because changing numbers is just like doing math problems. It is quite easy to add 20 million, but each number you adjust is a real indicator and you have to make a plan to complete it.

Let’s take a practical example: when we see that the repurchase rate of active old customers is already 64%, the possibility of adjustment is basically low. Can we consider it from the perspective of average order value?

These customers have a strong intention to buy, so you can push large targeted coupons to them, or make customized packages for each customer, showing them some high-order packages that they like.

Let’s look at other stratifications. For example, new customers with low average order value still have a relatively large room for repurchase. So for this group of people, can we make related recommendations to increase repurchase rates? In short, every target adjustment must be matched with actions, otherwise it will be just empty talk.

Another table is used here:

The table is too big, so I haven’t cut it in its entirety. Anyway, the horizontal part is a timetable, which is divided into the accumulation period, the outbreak period, the stable period, and the cleanup period according to the execution time of the activity. Each period has its own goals and corresponding population.

The vertical dimensions include: target population, corresponding strategies, gameplay, gameplay details, publicity and promotion, and required resources.

This is basically the entire strategy table. After it is confirmed, we need to make a specific work progress table, but I won’t go into details about this.

A few words about strategy. You must have an understanding of the users you are segmenting and be able to match them with measures. We often talk about member portraits and member segmentation. No matter how detailed the portrait is, it will be useless if it is not relevant to your business and you cannot give differentiated treatment.

Let me talk about it from the perspective of the life cycle, hoping to give you something to think about:

• New members: They have a fresh impression of the brand, but are in a skeptical stage about both the product and the brand. They can increase the frequency of contact, do some interaction or knowledge education, popular lectures, courses, activities, etc.
• Active old customers: They are loyal to the brand and directly inform customers of product-related activities. It is even better if they are given special interest points.
• Highly recover members who are expected to churn: Start with the product or demand, provide interest points, and do interactive awakening first;
• Low-recovery and expected-loss members: Start with the product or demand, provide benefit points, and do interactive awakening first. Flash sales, points redemption, lucky draws, etc.
• Lost members: ROI is too low. Choose high-quality groups for marketing . You can circle out customers with high orders, high frequency, and recent behavior, such as recent browsing, collection, and added to cart. After circle out the group, directly stimulate benefits.
• Execution and Optimization

By the execution period, the most difficult parts have basically passed. At this time, we generally carry out work in several modules according to the work progress table, establish crowd packages, produce content, and execute key event nodes. For example: launching a page, pushing text messages, publishing on WeChat , holding events, launching Diamond Exhibitions, etc.

A few points to note:

• It is necessary to follow up and record the data of each execution node, including content clicks, page browsing, add-to-cart, favorites and other data, in order to leave good reference materials for later review and future activities.
• A testing mechanism can be set up. For example, we can set several interest points for a group of people to conduct testing, and then set different content to reach each interest point. Finally, we will choose the one with the highest response to reach the whole group.
• Responding quickly to changes does not mean that what is decided will remain unchanged. We must respond quickly to changes. When a certain reach is not ideal, we must have an improvement plan.
• Review

I won’t say much about this, it is very important. Compare the estimated value, target value, and actual difference to explore the good and bad performance in execution and the points that can be optimized.

Finally, let me complain about the 618 rhythm that was disrupted by Tmall this year. The battle line was unusually long, users were confused, and merchants were extremely tired.

Author: kitty, authorized to be published by Qinggua Media.

Source : Everyone is a Product Manager