How to improve e-commerce data conversion through third-party tools?

Driven by the "Internet +" policy, the Internet has developed into the 3.0 era. Driven by policies and trends, the mobile Internet has developed rapidly. Various mobile apps have appeared one after another, which not only greatly promoted the development of mobile lifestyles, but also created a lot of new business value. Among them, mobile e-commerce , as an important part of daily life, has demonstrated strong vitality.

Umeng data shows that from July 2015 to June 2016, e-commerce shopping guide apps ranked fourth in growth rate in the vertical field.

On the one hand, from a macro perspective, the rapid development of the mobile e-commerce industry has provided ordinary users with more choices, brought more benefits, and made life more efficient and simple; on the other hand, the fierce competition has made survival extremely difficult for developers. Therefore, it is crucial for mobile e-commerce to identify its own advantages, conduct differentiated competition, and occupy the commanding heights of the market. In this process, in addition to high-quality products and services, operational strategies are also very important. Next, let’s talk about the operational strategies of mobile e-commerce.

Build a complete user portrait system

The emergence and rapid development of mobile e-commerce apps have made shopping behavior more fragmented. Through mobile apps, users can browse, order and trade products anytime and anywhere, and it is difficult to capture a series of user behaviors. Therefore, it is very important to have a clear understanding of user composition and predict user behavior. Through big data analysis , developers can not only have a clear understanding of user composition and user behavior, and ultimately restore the real needs of users, but also discover the problems in current user operations and then optimize and improve them in a targeted manner, truly achieving targeted results and achieving twice the result with half the effort.

Next, let’s talk about user portraits.

What is a user portrait?

In short, a user portrait is a labeled user model abstracted from information such as the user's social attributes, living habits, and consumption behavior. The core work of building a user portrait is to "label" the user, and the label is a highly refined feature identifier obtained by analyzing user information.

A user may have many attributes, such as nickname, gender, age, income, interests, region, job, family status, user behavior, etc. However, in the actual operation process, it is not necessary to refine all attributes for label construction. For e-commerce apps, from the perspective of promoting transactions, the dimensions of gender, age group, income, purchase history and purchase interests are sufficient.

The role of user portraits

If a user's purchase history and purchase cycle can be analyzed through data, then the merchant can determine when and what products to recommend to the user in order to more easily close an order. It can be seen that at least precise marketing can be achieved based on user portraits. In addition, there are many other applications as follows.

1. User population analysis

You can clearly know the user population distribution of the product, thereby verifying whether the product positioning is appropriate and making timely adjustments

2. Precision Marketing

Promote apps and conduct marketing campaigns through emails, SMS, and in-app push notifications, greatly improving product usage and order conversions

3. Data mining, multi-dimensional association

Through data mining, it is possible to find the relationship between beer and diapers, which can help improve product sales strategies.

4. Customized services or products

Provide customized product and service recommendations for each user and conduct refined operations

Precise push

After the novelty period or having gone through a complete purchasing cycle, in the fragmented time, if there is no new purchasing demand, users generally will not open the e-commerce app. In order to inform users of the latest discount information or new product launch information, it is necessary to wake up users through certain means. Email, text messages, and push notifications are all commonly used methods. However, in actual operation, the push effect is not very ideal. Next, let’s talk about two key issues in push.

How to be precise

What is precision push? Precision push means "pushing the right content to the right people." The key words, the right content, and the right people. So first of all, we need to create appropriate content based on the actual situation, and then push the content to the right people. There are several specific strategies as follows.

Push notification based on device dimensions. This generally refers to in-app push.

Based on the device dimension, it also includes APP-related attributes. For example, model, region, operating system, channel, APP version, system language, etc. Here are some typical cases for you.

For example, for a fresh food e-commerce application, if grapes are promoted in northern cities after they are on the market, the information should be pushed to the mobile phones of users in the north.

Push notifications based on user attributes. This includes in-app push notifications as well as text messages and emails.

The user attributes here actually refer to user portraits. The dimensions here include gender, age group, spending power, and spending behavior.

For example, in a promotional activity, razors are pushed to male users, and cosmetics are pushed to female users . This kind of push belongs to the basic dimension. A more advanced case is to predict the user's purchase expectations based on the user's purchase history, purchase cycle and product browsing behavior, so as to push the right product content to the user at the right time. This level of push is relatively difficult and requires the construction of a relatively complete data analysis system and data model. But the effect will indeed be much better.

How to increase open rates

Although there are many push methods now, there is an obvious problem that the opening rate is very low. For in-app push notifications, the opening rate is only 3%-10%, and more users choose to swipe away the push notifications or simply turn off the push notifications. The open rates for SMS and email marketing are even pitifully low, less than 1%. Therefore, the final result is that despite the investment of manpower and material resources, it is difficult to bring about transaction conversion. The reason for this situation, in addition to inaccurate push notifications, is that the push links provided to users do not take users to the application page corresponding to the content they are browsing, but instead take users to the application's homepage or h5 page. If users want to find the page they want to browse or the product they want to buy, they still need to perform corresponding searches in the application. This results in a particularly bad user experience. Over time, users' cognition becomes solidified and they will no longer try to click on pushed content.

However, with the advent of deep linking technology, this problem can be solved. Through deep links pushed via email, SMS, and in-app push, you can not only call up the app with one click, but also jump directly to the corresponding page. It can greatly optimize the user experience and save user time. If deep links can be combined with user portraits to achieve accurate push, the corresponding open rate and transaction volume will inevitably be greatly improved.

Social Promotion

For any promotion, whether it is event promotion or application promotion, social networks are one of the best ways, because social networks, as a super entrance for user traffic, can easily produce chemical reactions, thereby causing large-scale growth. Pinduoduo is the best example.

Sharing through WeChat, Moments, and Weibo not only increases the exposure of products or activities, but also quickly builds user trust in the product through the celebrity effect and acquaintance effect, thereby effectively driving the growth of user and order data.

But at present, the promotion of WeChat, Weibo, and Moments will keep user traffic on these social networks. When we see a product or promotion shared by a friend, we can usually only stay on the H5 page to browse the event or purchase the product, which is a poor experience. Many users give up on their purchase when they log in to the H5 page. However, if you jump to the APP through a deep link, the experience will be much better, and the user will not have to search for promotional products in the APP, thereby greatly improving the order conversion data.

At present, third-party tools can help developers better operate and promote APP, and functionality is not a problem. The difficulty lies in improving the accuracy of data analysis. Taking user portraits as an example, it is very difficult to truly achieve 100% matching of portraits, because there are definitely some data dimensions that cannot be collected. Therefore, in this case, a comprehensive combination of solutions may be more effective. Introducing some new technologies based on the application of traditional operating tools may have better results.