With 1 report, you can master data analysis and leverage stock conversion!

Are you easily dazzled by the complex data indicators? The following article is the author's own optimization of the operational report in data analysis. Let's take a look!

There is an important difference between traditional marketing and Internet marketing, which is whether data analysis can be done well in details. Many people find it mysterious how to do data analysis well. In fact, it is not easy to do data analysis well.

It is easy to be dazzled by the complex data indicators. Conventional data analysis includes data collection, data cleaning, data analysis, and data optimization. However, when facing actual projects, how to do a good job of analysis in actual combat requires you to respond to each situation.

After 2018, I started to focus on studying the transformation of corporate inventory. In the process, combined with my previous experience in data analysis, I made a lot of optimizations to the operational reports in data analysis. Today I will share some things about data analysis in inventory operations.

1. Stock Operation, How the Conversion Funnel Evolves

The core value of data analysis is to provide a data conclusion for operations. After a series of operational actions are implemented, what is the specific data feedback? Optimize operational strategies through data results. If you want to do a good job of data analysis, you must first understand a classic data analysis model: the conversion funnel model.

If we simplify the conversion funnel, it can be divided into four links: channel, display, click, and conversion . Here we take the example of user participation in activities as a reference.

1. Channels

This refers to the sources of various types of traffic. Public, private, online, offline and other places where traffic can be obtained can be understood as channels. If it is stock operation, common channels include user touchpoints such as APP/Mini Program/Community/Official Account.

2. Display

The activities you plan and launch through these channels cannot guarantee that they will reach all users. For example, if you launch a banner ad in an APP, only a small number of people may see the creative idea of ​​the ad, because many people do not open the APP frequently. For example, if a public account pushes an article, the number of readers may be only 10,000 for a user base of 1 million. In this way, there will be a loss between the potential traffic of the channel and the actual display. The channel to display is the first layer of the funnel.

3. Click

Even if your activity is seen by others, it is difficult to guarantee that 100% of the users who are willing to participate will participate. If the advertising creativity is attractive, the activity bait is stimulating enough, and the small details are not done well, there will be few participants. After going through layers of screening, there will be fewer participants, and finally only a small number of valuable users can participate. Display to click is the second layer of the funnel.

4. Conversion

This is the final effect of this activity. Common ones include attention, such as following public accounts, following Xiaohongshu accounts, following Douyin account data, downloading APP, registering and leaving calls, online communication, online consultation, online paid purchases, etc. Click to conversion is the third level of the funnel. You can see the traditional conversion funnel model on the left, and each level of the funnel is gradually decreasing.

With the help of stock to increase volume, the conventional funnel model can be changed to a certain extent, mainly focusing on the display to click link. The traffic in the click link is amplified through the social relationship chain. Compared with paid investment, the traffic cost brought by word-of-mouth communication is lower and it is easier to obtain conversion.

Because old users bring in enough new users, the data in each of the four links, namely display, channel, click, and conversion, will change significantly. The shape of the funnel will change slightly. This point has also been verified in past project practice.

In the previously planned activities of bringing in new users, if the activity gameplay is well chosen and the user participation path is smooth enough, old users can generate a lot of new users, which directly increases the number of users participating in the external sharing page. Having a high-quality user base is a prerequisite for later conversion.

2. Inventory operation, how to design user reports

If you want to achieve good stock growth, you must do a good job of user hierarchical management. How can the hierarchical management of users be reflected through reports? First, let's understand a few concepts, user first order conversion rate, user second repurchase rate and user multiple repurchase rate. If these indicators can be displayed in the report, the subsequent data analysis will save a lot of time.

To increase the stock, users must be operated in layers. Designing a scientific user report can quickly understand the user's first order conversion rate, second purchase rate, and multiple purchase rate, and intuitively feel the user's feedback on the platform through data, such as the first order conversion rate and the proportion of repeat purchase GMV.

There is no set of data reports that can solve all problems. The actual operation of the project basically needs to be designed around the project characteristics. The key is to understand the design logic behind the report. For the operation of existing users, how to intuitively display valuable data through reports. That is, the value data of first purchase, repeat purchase, and multiple repeat purchase mentioned above. The following template is for reference only.

Observe this report template in order from left to right and from top to bottom. The headers from left to right are category, data indicator, data indicator quantification, data notes, and time.

The category column is the business line, that is, the product line. For example, if the company sells 3C digital products, the category column can present the business data that needs to be displayed. It can also be modified to "Business Line = 3C Digital" to be more specific. If the business line is complex, the report design here needs to be adjusted.

The data indicator column breaks down the user stratification data, including first purchase, second purchase, multiple purchase, etc. It also breaks down several core indicators of stock user operations.

The quantification of data indicators is the disassembly of the GMV formula, which breaks down the GMV data, order volume data, average customer unit price, and the proportion of this data indicator to the total GMV. It is equivalent to disassembling each element in the GMV formula and displaying it in a two-dimensional state in the data report.

Data notes are supplementary explanations for the quantitative indicators of the data indicators. They are optional. The table provides corresponding explanations for the data indicators, which is relatively intuitive.

Time. Here we take month as an example. It can be changed to day, week, quarter, or year according to business needs.

This operational report has two cores:

1. The first is the logic of the report, which revolves around the stratified operations of existing users.
2. The second is the data indicators, which break down the user's first purchase, second purchase, and multiple purchases. This report is just a reference for ideas, and the actual project should be designed in accordance with business needs.

3. Existing Operations, User Report Case (Part 1)

No matter how beautifully designed the report is, it is useless if it cannot solve the problem. How to apply the above user report to daily work? Here we will analyze it with a real case to deepen the understanding of data analysis for business drivers.

When doing data analysis, you need to have a sense of data accumulation. Only when the data accumulates to a certain level can you do analysis. On the one hand, the stability of the data can be judged only when the data accumulates to a certain level. On the other hand, if the amount of data is not sufficient at the beginning of the project, it is impossible to make a judgment. Let the data run for a while first.

The data in the table are partial data of a certain project. The project started at the end of March, so the data volume in March was relatively small. However, the growth rate of the project was still relatively fast, from tens of thousands in March to 550,000 in April, 760,000 in May, and over one million after June.

Since we want to analyze the behavior of existing users through data, we need to sort out several core indicators. Here we analyze the average value of the data in the user conversion dimension, and we can actually add multi-dimensional analysis.

The data from the user conversion dimension shows that the average conversion rate of first payment for existing platform users is 22%, the average first repurchase rate for platform users is 3%, and the average multiple repurchase rate for platform users is 74%. The data shows that the GMV data of users who purchased more than twice accounts for a high proportion, indicating that users' repurchases contribute greatly to revenue.

As long as platform users pay, their retention and repurchase rates are good. At the same time, it is also found that the conversion rate of users' first orders and the conversion rate of first orders to second purchases need to be improved. The problem that the project needs to solve is how to improve the conversion rate of users' first orders and the conversion rate of users' first orders to second purchases.

I wonder if you get the idea of ​​extracting effective information from these dense data and then analyzing it.

IV. Existing Operations, User Report Case (Part 2)

"Discover problems - solve problems", find out what can be optimized from the data, and then make strategic adjustments. Data drives business growth and leverages the value of data. The data conclusions drawn above are just the fulcrum.

Since the first-order conversion rate and the first-order conversion rate of existing users need to be further optimized, we can make strategic adjustments around these two issues. The following strategy is developed based on the characteristics of the project at the time and the existing resources, and is for reference only.

1. First order conversion: three perspectives: traffic, product, and service

1) On the traffic side, increase the exposure of advertising space within the APP, adjust the PUSH push frequency from 1 time/day to 2 times/day, increase the number of SMS pushes to old users, embed fixed advertising space in public account tweets to expand product exposure, increase investment in channels with good results, and stop those with average results in a timely manner.

The core strategy of traffic is to increase the exposure of the activity. It is like one of the secrets of advertising. Repeat! Repeat! Repeat!

2) On the product side, optimize the discount range of goods, similar to package prices/combination prices. Appropriately package them into more creative activities to give users a sense of freshness. In addition, especially within the APP site, the product link is more concise after optimization, shortening the user payment process and making the user experience smoother.

3) On the service side, we will strengthen the one-to-one community service and create VIP communities for paying users. We will push activity information to the Moments at a fixed rhythm to cultivate user habits and strengthen user awareness of the product.

2. Converting first order to second purchase: Activity operation and community operation dimensions

Activity operation, designing old users to bring new users activities, encouraging old users to spread platform information. If the new users invited successfully pay, cash back can be given to old users, and old users get more rewards. The activity itself does not require additional cost investment, just take part of the profit from product sales.

Community operation, finding high-quality KOCs, using video live broadcasts to cultivate more valuable distribution experts.

Through the above strategy adjustments, continuous optimization and iteration are carried out in combination with data.

High-quality data analysis is an important lever to improve operational conversion. Operations without data support are like blind men touching an elephant. Data drives business growth. Although it may seem like a few short words, the work behind it is tedious and detailed. The secrets of data are hidden in the details, such as "1.01 to the power of 365" and "0.99 to the power of 365". If you want to find the hidden "0.01" through data analysis, it takes a lot of effort.

Author: Hu Xianwu; Official Account: Wenli Marketing Notes