Shueisha Builds Data Analysis Platform on Azure to Assess "Power of Works"! Aiming for Further Data Utilization Through On-Site Leadership

Data Analysis Platform to Assess "Power of Works" Becomes Important Amid Increasing Digitization and Media Franchising

Since its founding in 1926, the general publisher Shueisha has developed its publishing business in a wide variety of fields, including comic magazines, journals, and fashion magazines such as "Weekly Shonen Jump," "Weekly Playboy," and "non-no," as well as comic books, literary books, bunko-format paperbacks, shinsho-format paperbacks, art books, photo books, dictionaries, and encyclopedias. In recent years, the company has expanded its business domain with a focus on contents (IP), through digital publishing, animation, movies, merchandise, events, and overseas development. In preparation for its 100th anniversary in 2026, the company is working to "create contents and services that people can get absorbed in" under the company motto of "originality," "confidence," and "cooperation." 

As the digitization of goods and services accelerates, the nature of the paper-based business of publishing is changing dramatically. However, many tasks such as internal communication, sales tracking, and communication with bookstores and agencies remained paper-based as in the past, making it difficult to keep up with the times. Mr. Akihiro Suto, Section Chief of the Information Management Office, Information Systems Department, explains as follows.

"For example, in the case of paper magazines and books, when we wanted to see the sales of each work such as "Demon Slayer: Kimetsu No Yaiba" and "Spy × Family," we could grasp the trend by looking at the sales data provided by bookstores and agencies. However, for digital media, the indicators we need to refer to are much more diverse than in the paper media, such as sales by episode on apps, sales by coins or points, and free reading campaigns by each online bookstore. Sales may also increase rapidly as a result of animation broadcasts or the launch of merchandise at convenience stores. It became difficult to determine the timing and number of copies to be reprinted based solely on sales data. In addition, even if we tried to get data on online bookstores, apps, merchandise, animation, and events, the data was so overwhelming and unorganized that we could not immediately get the full picture. When asked by editors about the initial sales speed of their works, we were increasingly unable to answer the question as a sales team." (Mr. Suto)

According to Mr. Suto, for more than a decade, the company was aware of the challenges posed by the increasing volume and diversity of data. The editorial department, the sales department, and the online managers worked together trying to consolidate and share data. However, as the digital domain continued to expand, the data became more diverse, and exchanging and organizing data became more complicated.

"Editors want to know the power of the works they are in charge of. Until now, we have mainly used paper sales data as an indicator to know the power of a work. However, as the digitalization and media franchising advance, the indicators used to determine the power of a work become more diverse, and it becomes harder to know how to look at the data and make a decision. Therefore, it became important to have a platform for analyzing and sharing a wide variety of data." (Mr. Suto)

Under such circumstances, Shueisha spent about a year building a data analysis platform, which will be operational from May 2022. It was Microsoft Azure (referred to below as "Azure") that was selected for this data analysis platform.

The Sales Department, Often Interacting with Editors and External Contacts, Decides to Use Azure as the Foundation

What triggered the company to build a data analysis platform using Azure was working from home, which had been in full swing since 2020, says Mr. Suto. The work-from-home program was started as an initiative to promote workstyle reforms, etc., and the company had been introducing Microsoft 365 throughout the company as a communication tool for that purpose.

"The COVID-19 pandemic occurred just as we completed the company-wide implementation of Microsoft 365, so we were able to use Microsoft Teams and other tools to perform our usual tasks smoothly. At the time, I was in the Comic Sales Department, and we used Microsoft 365 to exchange sales data with the editorial department. That is when I learned that Microsoft 365 came with Microsoft Power Platform ("Power Platform"), a tool that allows you to make business changes in low-code. We had been using Microsoft Excel ("Excel") for data processing and reporting, but Microsoft Power BI ("Power BI") makes aggregation and analysis easier and allows us to present data to editors more efficiently. That's why I started proposing internally to create a data aggregation platform so that we could take advantage of Power BI." (Mr. Suto)

As more work was being done in an online environment through working from home, the need to create a platform for aggregating and sharing data grew rapidly. In addition, the company's business partners and others were increasingly calling for such a system and making concrete proposals. However, due to the lack of an efficient means to achieve this and the lack of a person in charge of promoting it, the efforts to create such a platform were making little headway. That was when Mr. Suto, who knew the business needs of the sales department, took the initiative in pushing for the construction of a data analysis platform, which jump-started the project.

"We chose Azure because it was a cloud platform that made it easy to use Microsoft 365, Power BI, and Azure Active Directory (AAD), and because our partner companies and the Information Systems Department had knowledge and expertise in Azure. There was also a proposal to introduce a tool from a BI vendor that our other departments were using or another public cloud, but we decided it would be better to uniformly use Microsoft products in terms of licensing and operational costs. Personally, I had been using Microsoft Flow, a workflow tool, before it was integrated into Power Platform, so I felt more comfortable working with Azure." (Mr. Suto) 

While working in the sales department, Mr. Suto has been responsible for providing on-site level support for the modification of in-house business systems, the launch of web systems, and workflow improvements, in cooperation with the Information Systems Department.

"My father is also a PC enthusiast, so I have always been familiar with IT and systems. I also enjoy analyzing data and taking the initiative at work. The timing was right to gain the understanding of management, so the sales department took the lead in creating a company-wide data analysis platform using Azure." (Mr. Suto)

Visualizing Data with Synapse and Power BI, and Cataloging with Purview for Data Democratization

However, despite their determination to lead the system building effort, the sales department was far from having the knowledge and experience in systems and project management that they needed, says Mr. Suto.

"If you don't know what your IT vendor or development partner is talking about, you can't properly communicate the requirements, prices, and deadlines you want to achieve. I thought that it would be no good talking to them if I did not have the knowledge to discuss things on equal terms, so I made time to study hard. Fortunately, there are many learning materials and certifications for Microsoft products. Having obtained several certifications, including the Power BI certification 'Microsoft Power BI Data Analyst (PL-300),' I am now able to talk with the people involved in the project on equal terms." (Mr. Suto)

The data analysis platform consists of Azure Blob Storage as the data storage and Azure Synapse Analytics as the data warehouse (DWH), and users access the DWH from Power BI to view regular reports and individual analysis reports. The data sources currently include sales data of digital contents, product master data, bookstore master data, and editorial department master data in addition to sales data of magazines, books, and comic books as well as inventory and incoming/outgoing data that have been managed by the sales department.

Mr. Hirotaka Igarashi of the Comic Sales Department, who has been working with Mr. Suto to promote the use of the data analysis platform and its deployment within the company, says as follows:

"We are accumulating data, mainly sales data, and in the future we would like to accumulate and utilize data on in-store purchasing segments, history of sales promotion activities, etc. However, the data is used in various places and formats, and it will take some time before we can collect and analyze it all in one place." (Mr. Igarashi)

Mr. Suto follows. "For traditional paper sales data, there is a long history, and the industry has an organized system for standardizing and exchanging data. However, for digital data, there are many different formats and data items to be acquired, such as online bookstores, online agencies, events, animation, movies, merchandise, and so on. How to consolidate these items is an issue that needs to be addressed in the future. There are many things to work on, such as data standardization and protocols for exchanging data." (Mr. Suto)

To address these issues and democratize data, the company has started using Microsoft Purview ("Purview") with the aim of cataloging and properly managing a wide variety of data once the platform is up and running.

"With Purview, we can quickly see on the screen which data was created, where it was created, and who is managing it. This is an important function for implementing the data analysis platform across the company and ensuring that everyone shares the data." (Mr. Suto)

Collecting and Sharing Data in One Place Increases Operational Efficiency and Expands the Data Utilization Potential

By building the data analysis platform using Azure, the company has begun to see a variety of positive effects.

First, by consolidating data in one place, the speed of data provision has increased, and all parties involved can now view the same data for discussion.

"Being able to see the same data at the same time is a simple yet extremely important improvement. In the past, when an editorial department member wanted data, they had to check with the sales department for sales data, and the sales department would gather the data from any necessary sources, compile it into an Excel file, and provide it to the editorial department via email, or upload the Excel file to cloud storage for their access. Now we provide the basic sales data as a Power BI report, so everyone can see it at any time and on any device they like. The stress created for viewing data, which used to be a slight burden on everyone, has been eliminated, and discussions based on the data can be held immediately." (Mr. Suto)

The company has confirmed significant benefits in terms of reducing the workload. In order to provide data, the work of collecting and processing the data is necessary. Many of these tasks have been eliminated.

"We receive requests for data not only from the editorial department and the executives, but also from various business partners in conjunction with animation, merchandise development, and organizing events, etc. In the past, we had to collect data and recreate reports for each request, which means that we had to do data-related work almost every day. Now, when we need to process data, we can easily recreate reports with Power BI and provide them to our editorial department or business partners quickly and easily." (Mr. Igarashi)

In addition, being able to combine and look at multiple data sets has made it easier to take on new initiatives.

"We can now look at not only paper sales data, but digital sales data in combination. We can now see the power of a work from a variety of perspectives and find, for example, that digital sales are higher than paper sales. Currently, digital sales data is limited to sales data per book at online bookstores, but we plan to add data in various formats from sources including websites, apps, animation, movies, and events. Also, in the future we will see importance in indicators such as 'reading completion rate and retention rate within the app', not just in sales results, such as 'millions of copies sold.' By combining various types of data, we hope to provide new perspectives on works." (Mr. Suto)

Aiming for an "Environment Where Business Departments Can Use Data Freely"

Amid the increasing digitalization and media franchising, the types and volume of data handled will continue to expand. In this regard, expectations for Azure Synapse Analytics ("Synapse") and Purview are growing.

"The advantage of Synapse is that it can handle a wide variety of data sources without changing the underlying architecture. As we add more data sources, we have to configure ETL tools and add new queries, but with Synapse, we can easily add new settings and data sources without having to update the system each time. Also, with Purview, when people want to use data, they no longer have to worry about who to tell. Only the data catalog can be published and shared within the company without allowing direct access to the raw data. The system also makes it easy to manage the data as it grows, by for example giving alerts to the user that the data is personal information and should be handled with caution. We have just started using Synapse, but from the beginning we are convinced that this is what we have been looking for." (Mr. Suto)

According to Mr. Suto, Shueisha's data analysis platform aims to create "an environment where business departments can use data freely."

"We want to create an environment where data is not managed centrally by the IT department, but can be used freely and securely by those who want to use it, when they want to use it. After completing the construction of the data analysis platform, I was transferred from the sales department to the Information Management Office of the Information Systems Department, and the Information Management Office is working on such company-wide information management. Synapse will allow us to handle the increase in data sources without changing the architecture, and Purview will automatically scan and manage new data as it is generated to ensure governance and compliance. In that sense, we have high expectations for these two services." (Mr. Suto) 

After Mr. Suto's transfer, Mr. Igarashi has taken over system promotion and data utilization in the sales department. As a new initiative, he is also working on demand forecasting.

"We are examining whether we can efficiently determine the number of copies to be reprinted based on the daily incoming/outgoing data stored in Synapse. The number of reprint copies of comic books is estimated for each item and multiple members of the sales department share the task. Since there are several thousand items in the comic book lineup, even if five or six members of the sales department work on the task, one person is responsible for over 1,000 items. It may take two to three days just to collect sales data of 1,000 items, print them out on paper, and review them. We would like to make it possible to analyze the data from Synapse and display the projected number of copies to be reprinted on a form. As it is easy to integrate the data into a Power BI dashboard, we also aim to make it possible in the future to efficiently determine the number of copies by looking at the projected figures. We are also considering using Azure Machine Learning for data analysis in the future." (Mr. Igarashi)

Mr. Suto says that by having people on the site take the lead in data utilization, initiatives that are in line with actual operations can be implemented quickly and efficiently. Mr. Suto, who has moved from the sales site to the IT site, is now advancing his efforts from the perspective of data management. Shueisha's efforts to build and improve the data analysis platform will further accelerate in order to realize "an environment where business departments can use data freely."