-
The third level: customer cultivation
Retaining one old customer is equivalent to developing five new customers. The customer life cycle mentioned above is more of a marketing mechanism, but how to truly cultivate customers from raw to mature is actually far more than a set of marketing mechanisms. . The two levels mentioned above are techniques, easy to teach and easy to learn, but data marketing is actually a theory, not a skill, and can be used in all aspects of marketing. So far I can only give examples.
Let’s look at how data marketing can cultivate customers in several stages, using the example of an online supermarket.
Non-customers - potential customers - browsing customers - purchasing customers - secondary purchasing customers - loyal customers - high ARPU value customers.
Stage 1: Non-customers – potential customers
From "non-customers" to "potential customers" the core is about the differentiation and positioning of user groups. What kind of people are likely to become No. 1 customers. Of course, in the era of horse racing, this is not the most important problem. Traffic is king, and conversion rate is the second issue. Any channel that can obtain large traffic is a good channel, which is why we can see it now So many B2C companies have taken up almost all the homepage advertising space on the portal website. But I believe that when B2C matures and traffic costs further increase, blind homepage placement will not be able to bring reasonable ROI. At that time, B2C will start to consider who my target customers are and how to use more accurate way to interact with customers. Take a regional B2C like Yihaodian as an example (I don’t know much about Yihaodian’s promotion strategy, I’m just giving an example from YY). After 2 years of data accumulation, I believe they have been able to draw portraits of some typical customer groups through cluster analysis from the existing data: that is, from the data of 1 million customers, find some with similar characteristics. Characteristics of customer groups. This specific data mining process is relatively complicated. To put it simply, it is to find some groups with specific characteristics from the user's wide table (in the CRM) through probability distribution. For example, we found that 5% of the people who come to buy in Beijing are 15 -25-year-old Beijing women, we can classify them as the "Beijing female students" group. In this way, it may be effective to provide some dormitory supplies in subsequent promotions targeting this group of people. (China Mobile has done a good job in this area. Their packages are summarized through cluster analysis and business judgment, and are designed for specific groups of people to meet their specific needs. For example, through data analysis, they found that many users The phone bill is 0 from January to February and July to August in the middle of the year, and many people will lose their phone numbers during these times and will never use their mobile numbers again. After research and analysis, it was found that students go home for vacation every year. Numbers would be changed, resulting in vacant numbers. When they returned to school, many people could no longer find their original numbers, resulting in the loss of customers. So they launched a long-distance package for student cards so that students do not need to change their numbers after returning home. You can also make phone calls at similar rates. This package became very popular as soon as it was launched.)
Stage 2: Potential customers—browsing customers
The problem here is the promotion conversion rate, or click-through rate. We usually call this value CTR (click through rate), which means that after an email is opened by 100 people, X people click on any action on the page. button, then CTR=X/100. This is an indicator used to measure the effectiveness of promotion. The main indicator that affects CTR is whether the target customer group is accurate and whether the marketing content is attractive to this group. Following the example in the previous paragraph, No. 1 store wants to promote in Beijing and cooperates with NetEase Mailbox. At this time, how well is the data accumulation of NetEase Mailbox? Can it provide accurate segmentation of customer groups? Can No. 1 store be able to The use of such segmentation has become the key to its marketing conversion rate. Judging from the example I cited in the first layer of data marketing, it seems that everyone receives the same email. For example, on March 8th, a girl aged 15-25 years old received the message "Send a good mop to mom", and a woman aged 25-35 received the message "Comfort yourself and have a snack feast" , what men aged 25-35 received was "care about your woman and protect her skin". How harmonious and beautiful this is. Here, the data accumulation ability of promotion channels has given advertisers great constraints. Nowadays, general portal website homepage advertising can only be accurate to IP addresses at most, and some even IP cannot be distinguished. Under such circumstances, precision marketing is There is no way to talk about it. In this case, all we big B2C investors can do is to delineate the target customer groups and then find the places where these target customer groups gather. Some vertical websites and forums provide good resources in this regard. If No. 1 Store finds that the gaming crowd in Internet cafes is a major customer base for online supermarkets, then find game channels such as pplive to promote instant noodles and ham sausages. , it should be a good match with Coke and the like.
Stage 3: Browsing customers – purchasing customers
From this stage on, from the outside of the website to the inside of the website, the controllability is greatly enhanced, and what we can do is greatly increased. If we divide the core competitiveness of B2C into "good goods, good sales, and good delivery", from here on, in one website detail after another, the core competitiveness of B2C's front-end (good sales) will be shaped. Comparison and selection ultimately become a major factor in customer selection. (Next time I will share with you a Swiss online supermarket with an ARPU value of more than 20,000 Swiss francs per year. The customer experience is very good)
Let’s start with the customer entry. In the most ideal situation, when a customer comes to a website, we should be able to know his information (such as gender, age, preferences) based on the cookie he left, and then show him a message specially designed for him based on his characteristics. His customized homepage. Of course, due to limitations in data accumulation and cost considerations, this is still impossible to achieve under current conditions, but we can still think along this line of thinking:
1. Is this his first time here?
2. Has he ever bought anything?
3. Is he a frequent buyer?
Through the above three questions, we can classify the people who come to the homepage into several categories: new customers, old customers, and loyal customers. A good homepage needs to take into account the needs of the above three types of people at the same time.
For new customers, the biggest confusion is usually not knowing what to do here. Take Yihaodian as an example. If I were a new customer, I would first want to know a few things: What does this website mainly sell? How is it different from other b2c websites? Why don't I buy it on Taobao? At this time, it is a big challenge for the website to make customers feel that this is a supermarket through products and visuals. In this regard, Yihaodian can actually do better. Through the products and visuals, people will know at first glance that this is an online supermarket selling daily necessities. In this way, Yihaodian can be compared with full-category supermarkets such as JD.com and Dangdang. The shopping mall differentiates itself, and then emphasizes its advantages over the online supermarket purchasing experience (cheap, door-to-door delivery). "Buying daily necessities online" is a very large market, and customer stickiness is also very high, and the Internet currently belongs to the blue ocean. Such positioning is not a pity. Then when the customer comes, he feels interested in this online supermarket and is willing to try it. The next question he thinks about is: What should he buy? I have tried my best to collect the 100 yuan for postage at Yihaodian many times. I believe I am not the only one who has encountered this problem. The reason is: consumer goods are an area with quite dispersed demand, and there are dozens of main brands in food and beverages alone. For a new customer, when he comes to No. 1 Store, he has no idea what he wants to buy, and if he sees something that happens to interest him, he will not buy it. It is really difficult to let so many different customers find something to buy in such a limited page space. For new customers, data marketing is of little help, and business sense is more important than data. According to the positioning of the entire supermarket, find the mainstream products that customers like (such as toothpaste, shampoo, rolling papers, etc.) and give them a reason to try (such as cheaper than supermarkets, such as not having to move it by themselves, etc.) , may be a good way to attract customers to experience it for the first time. In addition, showing them what others are buying and having various rankings can also be helpful for those who are undecided.
Let’s take a look at the old customers who see the home page. These people have already had purchase records. If they can come back for the second time, it means that they were satisfied with the first experience. At this time, some people already have a clear intention to buy certain goods (for example, they have finished eating the biscuits at home), and some people are here to "shopping" (to see what is available). These two types of people are very different and require very different experiences. For those with clear intentions, search is the first choice, category is the second choice, and product display on the homepage is the icing on the cake. For people who "shopping", it's complicated. Some people like to see various activities, some like to start shopping from the category, some like to see what others are buying, etc. At this time, through the tracks left by him while browsing, we can help from time to time: if we find that a person has changed multiple keywords and still has not added anything to the shopping cart, we can pop up the advanced search and help him update it. Find accurately. If we find that a person has browsed 6 pages in a certain category and has not clicked on a product, we can pop up a search box or even pop up a customer service dialog box: "What product are you looking for? Do you want help?" Here We can come up with a lot of scenarios. The core logic is to judge the customer's purpose based on his behavior, help him when he needs help, so that he can find it quickly and have a happy shopping.
For loyal customers, they have already made many purchases and basically do not need much help from us. Let them fly for a while~~ We will talk about how to increase their ARPU value and purchase frequency later.
Stage Four: Purchasing Customers – Second Purchasing Customers
If a customer can come for the second time, the possibility of him coming for the third time will be very high, and the churn rate between the first purchase and the second purchase is the highest. The main reason is definitely dissatisfaction with the shopping experience, slow delivery, broken items, wrong goods, etc. These supply chain problems cannot be solved through marketing. What digital marketing can do is to promote customers’ secondary purchases when they are basically satisfied. Let’s look for touch points first. Without touch points, there is no marketing. After the customer pays, the contact points may include: delivery notification, receiving the goods in person, unboxing the product, customer complaints, etc. Data-based marketing treats every touch point as an opportunity to market to customers, understand customer needs through data, and give customers the best experience. Part of the content here has already been discussed in the second level: customer life cycle management, so I won’t repeat it. It’s more about grasping the details. For example: if a customer who bought two bottles of Rejoice sees a coupon for Safeguard Shower Gel when he opens the package, will he be more likely to buy it a second time? If a customer calls to complain about slow express delivery, we can give him the right to free shipping for an unlimited amount of time within one month. Will he be more likely to come back next time? These are endless. How to connect data and business sense to create the optimal customer experience is the core skill tested here. It is also the core competency of front-end competition in today’s increasingly serious b2c homogeneity. If you are interested in this topic, we can discuss it with specific examples, but we won’t go into details here.
Stage 5: Second-time purchasing customers—loyal customers
When a customer comes twice, how to cultivate him to become a loyal customer is the ultimate challenge facing the B2C industry. The single customer introduction cost of 100 yuan+ can only be recovered by turning customers into loyal customers. This topic cannot be discussed apart from the industry. Let’s take No. 1 store as an example. Supermarkets are in the fast-moving consumer goods industry. In English, they are called "fast moving consumer goods", which are consumer goods that move quickly. Going fast means a lot of repeat purchases, and for B2C, it means a high return rate. Really? I don’t think so. The key depends on whether consumers’ habits have been cultivated. Usually when consumers like me buy things from Yihaodian, it is very random. For example, I found out that I had run out of biscuits today. My first choice would definitely be to go to the supermarket in front of my home to buy them, otherwise I would not have anything to eat tomorrow morning. In traditional supermarkets, different groups of people have formed relatively fixed purchasing habits. Housewives may go there once every two days, and students and office workers may go once a week or two weeks. However, on the Internet, this habit undoubtedly does not exist. . How can you develop this habit? Habit = repetition of behavior. Data-based marketing is to conduct targeted marketing and guide them to behave by understanding people's needs. For example, suppose we find through data that 70% of customers who bought 500ml shampoo and made a second purchase made their second purchase within 30-45 days, then we can set a rule that within 30 days Sometimes it triggers primary marketing and gives customers some promotions for secondary purchases. The basic logic is to discover the habits of a certain group by analyzing the data of a certain group, and then strengthen this habit through marketing, so that more customers will have such buying habits. There are two important points here. First, you cannot create habits by yourself. For example, I think a certain shampoo will be used up in about 20 days. Instead, you must try to dig out the existing habits of customers from the data. This is in line with the reality of customers. situation.
Stage Six: Loyal Customers—High ARPU Value Customers
The so-called ARPU is average revenue per user (ARPU-Average Revenue Per User). High ARPU value is related to two factors: purchase frequency and customer unit price. Above we have roughly talked about how to increase purchase frequency through data marketing. Let’s take a look at how to increase the unit price of No. 1 store through data marketing.
Related recommendations: "Related recommendations" can increase the overall customer unit price, and can be done by the system or manually. Taobao sellers basically do it manually, and the effect is good. Now I see automatic related recommendations in No. 1 Store, but I don’t know how effective it is. Judging from Taobao’s experience in making related recommendations, it is still very difficult to do a good job in automatic recommendations. Manual is a more effective model, and can also be combined with various promotions. The disadvantage is that it is relatively labor-intensive and can only be done on key products.
Promotional recommendations: In fact, for people who "go to" supermarkets, buying "more cost-effective" products is a very important psychological need. Cross-category activity recommendations or promotional product recommendations, especially some consumer goods, should be good in theory. , the reason is that there are not many people who can browse one category after another, so if the promotional information currently being carried out on the website can be pushed to customers more accurately, it will directly bring sales. For example, if I buy biscuits, potato chips, milk and other food products, if you tell me that paper products are on sale, and a large pack of Qingfeng paper towels originally priced at 35 yuan is now 20 yuan, only for one day, then I may buy it. Sex will be great.
Providing comparison: If we go to the supermarket, we will find that many people compare which packaging is more cost-effective when buying things. 500ml of milk is 10 yuan, and 800ml of milk is only 13. It’s a profit! At the same time, he also increased the price per customer. Therefore, the website must provide customers with this opportunity for comparison. The method is actually very simple. Taobao already provides a comparison of standard weights in the food category, but it does not yet provide it in the toiletries category:
Recommendation for free shipping on purchases over 100 yuan: When observing the unit price of No. 1 Store, you must first take a look at the "free shipping on purchases over 100 yuan" policy. This policy determines the customer's psychological bottom line. If you buy about 100 yuan, you can basically call it a day. So 2 questions will arise next:
1. For a customer who is optimistic about the product and the price is much lower than 100 yuan, how to stimulate him to buy more than 100 yuan?
2. For customers who are optimistic about the product and the price exceeds 100 yuan, how to stimulate them to buy more?
For the first type, we can recommend some products with 10 yuan for postage, 30 yuan for postage, and 50 yuan for postage. At this time, there is no need to be cost-effective. As long as these things are definitely used by him in the future, the effect can be achieved. .
For the second type, we can give some higher-level discounts, enjoy *** when you spend 200 or more, and enjoy *** when you spend 400 or more. It is very simple to do, just have you thought about it.
I have written so much, and the reason why I take Yihaodian as an example is that on the one hand, supermarkets are a b2c industry that everyone is familiar with and easy to discuss. On the other hand, b2c data marketing also requires relatively large technical investment. Generally, Taobao Stores and small-scale B2C are not very popular yet, so this is just an example. I don’t know much about the specific situation of No. 1 Haodian. I’m wrong in some places. Please forgive me for my classmates at No. 1 Haodian~~
Article source: Paidai.com please indicate the source link when reprinting.