Product data reporting is an essential task for product and operator personnel. Whether it is a weekly report, monthly report, or analysis report on the performance of a new version, it is necessary to organize, analyze and extract key points of the data based on the reporting objectives, and finally form an instructive, easy-to-read and beautiful data report.

This article summarizes how to write a product data report from the perspective of writing steps.

The article outline is as follows:

1. From the reader's perspective, clarify the report's positioning;
2. Disassemble core indicators and establish data models;
3. Data compilation and analysis;
4. An example of a monthly report template is given below.

1. Clarify the report positioning

A report is made to a certain group of people, so first of all, we must identify the target of the report, organize the content, structure, and focus of each module in the report from the perspective of the target of the report.

For example:

If the report is for the company's leadership, for example: routine reports on the company's business lines, or reporting on the performance of new products or new versions to the product line leader; at this time, the report should highlight whether the key indicators have met expectations and why each key indicator has this performance. It is necessary to briefly explain where the problem lies or the reasons for the excellent performance by breaking it down into detailed indicators, and finally summarize the team's next improvement plan.

If the report is for business colleagues in the team, the focus of the report is on discovering problem points and proposing improvement plans or suggestions, and it is intended to play the role of using data to drive (che) the team.

If it is a report published to the public, it generally focuses on results and trends rather than processes. This category is the paradigm of the reports that consulting agencies publicly release.

2. Disassemble core indicators and establish data models

After clarifying the positioning of the report, you can then combine the report positioning with indicators such as product goals, activity operation goals, etc. to break down the core indicators and form a data model for the report.

For example, if it is an e-commerce product and is reported to the leader, then the core indicator may be GMV, which is the number of users multiplied by the average order value. Then we break it down step by step as shown in the following figure:

It should be pointed out that the core indicators and breakdown of the report are dynamic, and the data model also needs to be adjusted at different stages of the product.

For example: In the early stages of a product, the focus may be on user scale. After reaching a certain user scale, it is necessary to start focusing on user value, and the data model needs to be adjusted accordingly.

3. Data collation and analysis

1. Data Acquisition

There are many sources of data, and it is necessary to select stable and reliable data sources according to the needs of different indicators.

Common data sources include:

• The company has its own data statistics system;
• Third-party data statistics platforms, such as Umeng , Baidu, etc.;
• Third-party business platforms, such as advertising alliances , etc.
• Publicly available data sources.

Many of the data come from multiple sources. For example, basic user data is generally collected by the company itself and from third-party platforms, while data from external cooperation is collected from both the company itself and from third-party business platforms.

For data related to money, such as the number of orders, amount details, etc., generally speaking, there must be a strict reconciliation system to check and reconcile accounts.

However, if there are differences in user data, it is generally necessary to analyze the statistical caliber and statistical plan. If the differences are caused by statistical means, they can be ignored.

It should be noted here that if a data source shows large abnormal fluctuations, it is often possible to use another data source for comparative analysis. If the fluctuations on both sides are in the same direction and magnitude, it is necessary to analyze from a business perspective. If the two sides are very different, there is most likely a problem with the source of data statistics.

2. Data organization and cleaning

Data organization and cleaning mainly involves eliminating dirty data and data with statistical anomalies, structuring data, etc., which will not be elaborated here.

3. Analyze data : What matters is thinking

For data analysis , thinking is more important than tools and methods. First, you need to be clear about what problem you want to find, then propose a hypothesis and then check based on the hypothesis, rather than aimlessly looking for problems in massive data.

I think the summary of the analysis steps has been very thorough, and they are listed here. If you are interested, you can read the article "The core of data analysis ability is thinking" for more details.

4. Methods for analyzing data

Disassembly method:

Split a large problem into smaller indicators. If no problem is found, continue to break it down until the problem is found, so as to find the corresponding solution.

BCG Matrix:

According to different business scenarios, two coordinates are selected as coordinate axes to divide the business or users into different types for analysis.

Year-on-year analysis method:

Compare the same type of data from different businesses together.

User Analysis:

User analysis includes indicators such as usage breadth, usage depth, and usage stickiness. These indicators are generally a combination of several user indicators. For example, usage breadth includes the total number of users and MAU, while usage depth includes usage time, dwell time, etc.

There are many ways to analyze data, and you need to choose the right method based on the report’s positioning and objectives.

First of all, this is a monthly product operation report for leaders, so it is necessary to highlight the completion of core indicators, as well as the status of the secondary and tertiary indicators, so as to analyze the completion of this month's indicators from top to bottom.

Secondly, this is an e-commerce shopping guide product, so the first item in the report is to break down the data indicators and establish a data model based on product characteristics, so that readers can know the content framework of the report and the completion status of each data at a glance. Only then will the report be gradually expanded to clarify the specific situation, problem points, and improvement plans.

Indicator decomposition example:

Currently, the MAU is in the millions, so the product stage is still focused on user growth , followed by revenue;

Monitoring indicators can be adjusted at any time according to actual conditions to illustrate the problem.

Example of a data summary table:

After breaking down the data indicators and building the model, the next step is to display the key data and form a data summary table so that readers of the report can know at a glance the completion status of the core indicators, as well as which data exceeds expectations and which data have problems.

As shown in the following table, the target completion percentage of the part that did not meet expectations is highlighted in black to guide everyone to focus on the analysis of the problem:

Examples of core indicators:

There are a few points to note when displaying indicators:

The graphic style needs to be selected according to the focus of the report. For example, a bar chart can be used to show volume changes, a line chart can be used to show trends, and a clustered bar chart can be used to show the comparison of the same indicator of different businesses, etc.

A table can display at most one primary data and one secondary data, for example: MAU + month-on-month growth rate;

It is best not to give more than two reasons for the data situation. Avoid listing a bunch of reasons that will make it difficult for readers to find the main point.

Core indicator analysis:

If the core indicator in the above example is MAU, then it is necessary to analyze the composition of active users in that month, and you can also figure out the monthly retention rate .

Channel analysis example:

Here we use the example of new channel analysis, which is suitable for comparing multiple similar data, such as viewing the new users of free channels and paid channels at the same time.

And because we pay more attention to the increase in free natural traffic , we can also include the growth curve of free new traffic for the convenience of explanation.

Since the focus is on free channels, we can continue to analyze free channels in detail:

Retention user analysis example:

In addition to analyzing the total number of retained users, the following trapezoidal table is often used to analyze the changes in retention rate within a dynamic time period:

Example of average active days per person:

Average number of active days per person is an important indicator for examining user activity. Generally, a line chart is used to show the trend of changes:

The average number of active days per person can also be further refined and analyzed based on business needs, with detailed data on different channels, different models, different operating systems, etc.

Income statistics example:

For income statistics, it is more appropriate to use a summary table to show the overall situation first, and then subdivide different income indicators and select appropriate charts for display:

Summarize

When writing a product data report, one must first analyze the readers and the core product indicators, break down and model the indicators, determine the content framework and focus of the entire report, then organize, analyze and tabulate the data. The final step is to beautify the report.

Source: Everyone is a Product Manager