Two days ago, I talked to Reader K, who is changing jobs. He is an experienced product manager, but he is reluctant to talk to me because he has been rebuffed in interviews.

It turned out that interviewers liked to ask him questions about data, but this was his weak spot, and he had to answer them awkwardly.

“What I thought was a data analyst’s job has become a product manager’s skill set.”

“I have always learned data analysis, but I didn’t expect dACHang’s requirements to be even higher.”

I believe that many product managers also have this feeling, understand that data analysis is more and more important to the job, many large companies even in the recruitment requirements will indicate that they need excellent data analysis ability.

It can be argued that a product manager who does not understand data will not be a good product manager in the future. But this raises the question: How well do product managers really know data analytics? I think the following four skills are essential for product managers:

1. Understand the ins and outs of data

To understand data, product managers need to understand the context of data first and foremost. The following questions need to be clarified: what are the indicators of the business, the sources of data, the buried data, the types of data, the corresponding caliber of data, and the update mechanism of data?

In an organization, data generation, storage, processing and application are basically as follows:

2. Understand the data index library

When we do business analysis, if the indicators are insufficient or the data are uneven, it will make the data analysis very difficult, and it will take a lot of time to gather the desired data and do data cleaning, which requires the product manager to build a systematic data index database in advance. In large companies, it is generally the responsibility of a dedicated data product manager to build the database, but for other product managers, it is necessary to understand the database planning.

Take jingdong’s index database as an example. They decompose key target data into several sub-indicators, and then disassemble sub-indicators layer by layer to obtain specific indicators to be collected.

The segmentation indicators are disassembled layer by layer (the following figure is the disassembly of the customer acquisition cost index) :

Through a database of indicators, we can “know” the business we are responsible for, such as forecasting business data, identifying which indicators to monitor, and identifying the causes of problems in an orderly manner.

3. Know how to raise data requirements

When product managers and data analysts connect, the most commonly mentioned requirements are front-end buried point requirements and report development requirements.

Buried point is a data collection method that records user data on the client and reports it to the server. It is an important source of data acquisition, and the obtained data is usually clean and reliable.

Before submitting the buried requirements, the product manager should follow the user experience process one by one to avoid omission. If the data is not enough or indicators are not found until the service is about to go online, then it is very troublesome to raise the requirements. In addition, when submitting requirements for buried sites, details of requirements should be made clear, as shown in the following table:

Before submitting report development requirements, product managers should first determine whether the upstream data is supported. When raising requirements, they should make clear report themes, data acquisition methods, data types, and definitions of each field. Some places that are not easy to understand should also be annotated.

4. Understand data analysis

Knowing data is important, but knowing how to use it is even more important. Product managers should be skilled in using various classic analysis models, such as pyramid model for transformation analysis, AARRR model for user analysis, KANO model for user requirements analysis and prioritization, etc.

For detailed analysis process, please refer to the universal data analysis template in the following figure.

The traditional data analysis mode in an enterprise is completed by the business side presenting the requirements to the IT side. However, there is a big drawback to this, not to mention whether IT people can fully understand the results we want (often requiring back-and-forth changes), sometimes during peak demand times, IT takes days to make a small request.

For product managers in particular, I recommend that small data analysis requirements be completely self-fulfilling. With the rise of data analysis tools, data analysis is no longer a skill beyond the reach of non-technical people. Many people agree that data analysis will be as fundamental a skill as Office in the future.

FineBI, for example, which is used by many large companies, dramatically lowers the barrier to learning data analytics thanks to its no-programming, drag-and-drop design.

IT departments only need to prepare data, technical experts and business personnel can easily conduct self-service analysis through FineBI, and can also make real-time adjustments according to their own needs. In this way, the new collaboration mode is more flexible and effective, saving the time of repeated communication with data analysts

Mastering such analytical tools should be a required course for every product manager in the future.

This is just like that, in the past, video cutting was only for a few professionals, and creators needed to communicate with editors repeatedly before they could produce a film, but now video editing software has developed towards the direction of universal benefits, and every ordinary person has the opportunity to express themselves through videos.

Today, all walks of life emphasize fine operation and big data, data analysis has become a necessary skill for product managers. Today, I analyzed at least 4 data analysis skills that product managers should master, hoping to make progress together with everyone!

Tools to share

Finally, share your analytics tools and click “Data Analytics” to get them all!