What should I do if I can only report numbers when writing a report?

This article introduces three methods to help data analysts go beyond year-on-year and month-on-month data comparisons when writing data reports, and instead conduct in-depth analysis and interpretation of the data. This article is recommended for data analysis and product operation practitioners.

When writing data reports, many people simply give year-on-year and month-on-month comparison figures and don’t know where to start.

The style of this report is as follows:

• Sales fell 10% year-on-year.
• The user churn rate increased by 5% month-on-month.
• The product’s user retention rate increased by 3% year-on-year.

But if data analysts only report numbers, ChatGPT will soon replace you.

Here are three ways to help you write better data reports:

Method 1: Find out why

The first method is to find out why. You need to explain why the indicator changes, why it goes up, why it goes down?

For example, if you find that sales are up 10%, why are they up 10%?

The reasons here should be driven by drivers rather than structural ones .

The structural reason is that GMV is driven up by channel A. The pull of channel A here is the structural reason.

If you know that the increase in channel A is due to the increased investment in operations, then this reason is the real driving factor.

You can first identify structural problems by breaking down channels, regions, and products, and then find the corresponding responsible colleagues. Ask them if they have taken any business actions and express these changes in concise language.

For example, a batch of waistband ads were placed in the channel, which increased traffic and ultimately led to an increase in transactions. Generally speaking, it can be expressed like this: Due to the increase in traffic driven by the waistband channel placement, the transaction amount increased by 10% month-on-month.

This clearly explains the reason for the increase in transaction volume.

Method 2 What happens next?

The second approach is to find out what happens to subsequent changes in the indicator?

This step is mainly to allow the business to have a certain prediction of the subsequent trends and make corresponding treatments in advance. For example, the following measures can be taken:

• Analyze previous data to understand whether the traffic increase is sustainable and whether it is consistent with the business development direction.
• Consider market changes, such as changes in competing products and changes in user demand, and analyze and predict the impact of traffic changes.
• Develop a response plan in advance and formulate corresponding response measures for different traffic change trends. For example, if a certain advertisement cannot be delivered next month, you can consider looking for other channels or improving website SEO.
• Through these measures, we can better predict the trend of traffic changes and take timely measures to ensure the stability of traffic. At the same time, this can also improve the flexibility and adaptability of the business, which has a positive role in promoting the long-term development of the enterprise.

Method 3 How big is the impact

A third approach is to show how big the impact is.

If GMV increases by 10% month-on-month, is this increase within expectations? How is the current progress compared to the KPI target? Is it behind or ahead? If so, by how much? How much additional GMV do we need to increase next month?

If the traffic of my waistband channel increases, what is the quality of the waistband channel? Has the conversion rate changed? Has the traffic of the core users increased in the overall traffic summary? How much has it increased?

The above is the practice of mature businesses.

If it is an innovative business, we also need to extract the feasibility of the business model from the indicators.

For example, the user retention rate has been maintained at a low level for a long time, and this month's retention rate has been at a low value for three consecutive months. The impact is that the current business model cannot continue to attract users, and the feasibility of the business model is questionable.

summary

The above are three ways to improve your reporting of data by only reporting numbers. Some common methods. The order of execution is generally progressive, usually first explaining why, then what will happen, and finally the impact.

Author: Jason

Source: WeChat public account "Sanyuanfangcha (ID: sanyuanfangcha)"