How many of these apply to you? 5 reasons why data analysis doesn't work

It has been said for a long time that "business management should be based on data, not on KKD (intuition, experience, and courage)," and many companies have introduced DWH (Data Ware House) and BI (Business Intelligence) to tackle data analysis. However, how many companies are actually achieving results through data-based management? Even now, with the technology of data analysis having made great advances, there are still many cases where companies are faced with various problems and challenges and are unable to fully utilize the data in management.
Today, we will introduce to you a ranking of common problems that our clients, who have some sort of problem with data analysis and are not "data-driven," face. Companies that can laugh it off and say, "That's not true!" will be fine. If you think, "This could be true for us too...", then start looking into ways to improve right away!

No. 5: Leaving BI tool operation entirely to a specific person

Some companies start small, purchasing a single BI tool account and entrusting data analysis to an employee who appears to be adept at using it. While starting small is not a bad idea, leaving the use of the BI tool to a specific employee (or a permanent IT vendor employee) is a no-no. Even if that person knows how to use the BI tool, they often don't fully understand what should be analyzed with it. This can lead to a negative impression, leading to the tool being left unused without being evaluated. Only by understanding the company's challenges and considering how to improve them can one determine the type of analysis that should be performed. Since the person using the tool and the person considering the challenges are not necessarily the same person, trial analysis should be carried out as a team involving both parties.

No. 4: Lack of cross-organizational collaboration

Simply forming a team consisting of both the tool user and the problem solver is not enough. Data analysis requires collaboration across the company, involving multiple organizations, including the user departments that will actually use the tool, the department promoting the project, and the information systems department. We often hear of failures where a BI tool is introduced without sufficient internal coordination, resulting in the information systems department failing to cooperate with post-implementation use, leaving the user departments reliant on Excel for analysis as usual. Because this project involves multiple departments, it is important to establish a cooperative framework at the department head level in advance. Furthermore, in anticipation of the need for cross-departmental decision-making, it is ideal to have an executive officer-level employee participate to act as a leader within the company.

No. 3: Project duration is too long

Because implementing a DWH or BI tool requires a significant investment, projects are often approached with caution and time. Many organizations face the reality that a mountain of prioritized tasks and issues makes immediate action impossible, or that progress can only be made in spare time, resulting in a time-consuming process. However, if it takes six months or even a year to begin data analysis, the business environment may rapidly change during that time, rendering the analysis results meaningless by the time they are obtained. To avoid these risks and ensure successful project results, we recommend a short-cycle approach, first attempting an analysis that can be completed in a maximum of three months, and then adding additional analysis based on the findings. Key concepts such as "small start," "quick win," and "agile" should be kept in mind as common terms within the company.

No. 2: Building a dashboard has become the goal

This is actually quite common, but it's common for the original purpose of a project to be forgotten. Why and how is data-driven necessary in the first place? The original purpose should be to improve productivity by reviewing operations based on the results of data analysis, leading to improved performance in areas such as sales and profits. However, in many cases, the goal ends up being simply to introduce a DWH or BI tool or to create a dashboard (an analytical screen created with a BI tool). This makes it difficult to accurately determine what kind of screen should be created or to accurately assess the results. It's important to first clarify the purpose and set as quantitative performance targets as possible.

No. 1: Necessary data is not available and cannot be provided to users

The number one reason why data-driven approaches don't work is that the data isn't in order. Even if you go to the trouble of introducing a BI tool, you can't perform data analysis if the necessary data isn't available. There are many cases where data is scattered across various systems, making it difficult to collect, data cleansing is necessary because the data items and granularity are inconsistent, and the numbers don't add up. Our company has also experienced having to start over in several implementation projects because the data wasn't in order. Preparing the data is more difficult than visualizing it with a BI tool, yet it is an absolutely unavoidable and important step.

So far, we have introduced the top 5 reasons why data-driven approaches don't work.

• 1st place
  Necessary data is not available or cannot be provided to users
• 2nd place
  The purpose is to create a dashboard (analysis screen).
• 3rd place
  The project duration is long
• 4th place
  No cross-organizational structure has been established
• 5th place
  Leave it to someone who can operate BI tools

If any of the above points apply to your company's project, please feel free to contact us.

The person who wrote the article

Affiliation: Data Integration Consulting Department, DX Consultant

Kei Konno

Since April 2017, he has led a business supporting customers' digital transformation with a focus on data integration and data utilization. Since April 2025, he has been working as a digital transformation consultant, proposing solutions to customers' issues and planning new service menus. He is also active in a wide range of fields, speaking at various seminars, writing columns, and participating in actual projects as an observer.
(Affiliations are as of the time of publication)

Related Content