The Background Behind the Urgent Need for Core System Modernization
In recent years, many companies have been forced to modernize their core systems.
In a rapidly changing business environment, companies must leverage digital technology to reform business processes and maintain or improve their competitive edge.
However, the core systems of many companies lack the flexibility required to adapt to these changes.
Core systems that have been in use for many years have become complex due to repeated modifications and functional additions, leading to increased maintenance and management costs.
In many cases, proper maintenance has become difficult due to the retirement or shortage of engineers who understand the system architecture.
These issues are known as the "2025 Digital Cliff."
This issue, highlighted by the Ministry of Economy, Trade and Industry, warns that from 2025 onwards, the maintenance and operation of aging systems will reach their limits, potentially resulting in economic losses of up to 12 trillion yen.
Why Is Core System Modernization Necessary? A 4-Step Guide to Data-Driven System Migration ▶︎
What Are the Reasons for Core System Modernization Failure?
So, what are the reasons why core system modernization fails? The following are some of the most common examples.
- The Purpose and Expected Outcomes of the Modernization Are Not Clearly Defined
- Outsourcing Entirely to External Vendors
- Misalignment Between Management, Operations, and IT Departments
We will explain each of these points individually.
The Purpose and Expected Outcomes of the Modernization Are Not Clearly Defined
When initiating a core system modernization project, it is not uncommon to start with vague justifications such as "the system is aging" or "maintenance has become difficult."
However, without setting clear objectives and expected outcomes, the project lacks direction, leading to ad-hoc responses.
Furthermore, evaluating the return on investment becomes difficult, making it challenging to secure understanding and support from management.
Additionally, there is a risk that project success criteria will remain ambiguous, preventing effective progress management.
To lead a modernization project to success, it is essential to clarify "why the system is being modernized" and "what is to be achieved," while setting specific, measurable goals.
Outsourcing Entirely to External Vendors
Since core system modernization is a large-scale and complex project, it is common to engage external IT vendors.
However, if the project is led by vendors who do not fully understand the company's operations and challenges, there is a risk that the resulting system will not be suitable for actual business requirements.
Furthermore, there are cases where vendor convenience or technical constraints take precedence, causing the features and outcomes you originally intended to achieve to be deprioritized.
In successful modernization projects, it is essential to establish a structure where your company proactively controls the project while collaborating with vendors.
It is required that your internal team members are deeply involved in critical phases such as requirements definition, progress management, and testing.
Misalignment Between Management, Operations, and IT Departments
Core system modernization involves diverse stakeholders, including management, operational departments, and the IT department.
However, because each department has different positions and perspectives, misalignments often occur.
For example, operational departments often find daily data entry cumbersome and burdensome, and because they only interact with their own department's interface, they may struggle to grasp the importance of company-wide data integration.
The IT department may face situations where systems and tools are fragmented across departments, making overall maintenance and integration difficult.
Meanwhile, management wants to view data integrated across departments, but this is difficult to achieve under current conditions.
Because the positions and perspectives of each department differ in this way, it becomes difficult to align the objectives and direction of the core system modernization.
As a result, partial optimization tailored to individual department needs progresses, and the modernization project proceeds without achieving overall optimization, ultimately leading to failure.
Data Normalization Is Especially Important When Modernizing Core Systems for DX Promotion
To promote DX and achieve a truly effective core system modernization, data normalization is indispensable.
Below, we will examine why data normalization is necessary for the modernization of core systems.
Core Systems Consolidate Vast Amounts of Corporate Data
Core systems handle various operational data that form the backbone of a company. Data from diverse departments, including sales management, inventory control, financial accounting, and human resources and payroll, are accumulated on a daily basis.
However, this data is not always stored in an organized state.
It is necessary to properly perform data matching to ensure that data from different departments, various business systems, and both legacy and new records do not become fragmented.
Furthermore, data degrades over time. Maintaining outdated data requires significant costs and man-hours.
Despite this, many companies do not fully recognize the importance of data maintenance, often leading to cases where expected results are not achieved simply because only the system was updated.
With the advancement of DX, the amount of necessary data for customer management, marketing, procurement, and more continues to increase.
If this data is not properly integrated, matched, and continuously maintained, DX initiatives may end up being nothing more than a pipe dream.
Therefore, when upgrading core systems, forward-looking data maintenance is essential.
[Understand in 5 Minutes] What Is Data Cleansing? An Easy-to-Understand Guide to Objectives and Practical Examples! ▶︎
Expected Benefits of Data Normalization
Data normalization facilitates smooth data integration between departments, thereby improving operational efficiency.
Because normalized data can be easily shared and utilized across different departments, it enables the reduction of redundant tasks and the centralized management of information.
By resolving data duplication and inconsistencies, data quality is improved. This reduces the risk of making decisions based on incorrect data, enabling more reliable decision-making.
As a result, it becomes easier to align perspectives across the entire organization, from management to frontline staff.
In addition, normalized data expands the scope of utilization, potentially leading to new value creation and problem-solving.
For example, when leveraging advanced technologies such as AI and machine learning, you can perform more precise data analysis and forecasting.
Examples of Data to Be Normalized During Core System Modernization
When data is siloed by department, making company-wide optimal decisions is difficult. However, by centralizing data management, you can enhance efficiency and competitiveness across the entire organization.
So, what specific data should be integrated?
Below, we introduce the key data that will be normalized during the modernization of core systems.
Customer Data
Customer data is vital information for managing the relationship between a company and its customers.
By normalizing this data, you can gain a deeper understanding of customer behavior and preferences, leading to the development of effective marketing strategies and improved customer service.
It will enhance the accuracy of customer segmentation and enable personalized approaches tailored to individual customers.
Examples of Customer Data
- Basic Information (Name, Address, Contact Details, etc.)
- Transaction History
- Inquiry History
- Customer Attributes (Industry, Size, Needs, etc.)
- Customer Satisfaction Data
Product Data
Product data is essential for managing information regarding the products and services provided by a company.
Normalization enables the streamlining of inventory management, the optimization of pricing strategies, and the acceleration of new product development.
Furthermore, by combining this with customer data, it identifies opportunities for cross-selling and up-selling, thereby contributing to increased sales.
Examples of Product Data
- Product Code
- Product Name
- Pricing Information
- Inventory Information
- Supplier Information
- Product Attributes (Category, Specifications, etc.)
Transaction Data
Transaction data refers to information regarding a company's daily commercial activities.
Normalizing this data improves the accuracy of cash flow management and forecasting, which helps maintain financial health.
Examples of Transaction Data
- Order Information
- Billing Information
- Payment Information
- Delivery Information
- Return Information
Financial Data
Financial data serves as the foundational information for understanding a company's economic status and making strategic decisions.
To grasp management conditions in real time and conduct more precise financial analysis, the organization of financial data is essential.
Normalization contributes to the advancement of budget management and the improvement of investment decision accuracy, which will lead to the optimization of a company's financial performance.
Examples of Financial Data
- Sales Data
- Cost Data
- Expense Data
- Asset Data
- Liability Data
Proprietary Simulation: Labor Hours Required for Normalizing Core System Data
Since core system modernization typically spans a long period, it is essential to understand the required labor hours in advance.
So, how much time is actually required?
At our company, we have simulated the labor hours required for core system data normalization as follows.
Prerequisites (Source: uSonar Research)
- Total Data Records: 100,000
- Data Including Duplicates or Inconsistencies: 40% (40,000 records)
- Processing Time per Record: 5 minutes (Average processing time for all tasks, including corporate research, correction of inaccurate data, and duplicate checks)
Simulation
- 40,000 records × 5 minutes = 200,000 minutes = 3,333 hours ≈ 416 man-days
In other words, if one person works 8 hours per day, it is estimated to take approximately 416 days (over one year).
However, these labor hours include not only simple mechanical tasks but also the following operations.
- Data Identification and Analysis
- Data Item Definition and Standardization
- Formulation of Data Cleansing Rules
- Development of Data Transformation Logic
- Post-Transformation Data Validation
In practice, this task requires advanced business knowledge and data analysis skills, which may necessitate significant additional time and effort.
At the same time, this simulation reveals that data maintenance accounts for a substantial portion of the workload.
uSonar: Supporting Core System Modernization Through Data Alignment
While data normalization is critical during core system modernization, it remains a significant challenge for many companies.
uSonar maintains one of Japan's largest corporate databases, containing 12.5 million records. By utilizing this as a "dictionary," companies can automatically consolidate and maintain their core system data.
This significantly reduces the man-hours required for data maintenance and ensures that data remains accurate and up-to-date at all times.
Furthermore, because uSonar solutions can be implemented without altering existing operational workflows, companies can improve data quality while minimizing disruption to their business operations.
Summary
Modernizing core systems is a critical initiative for driving corporate digital transformation, yet achieving success involves numerous challenges.
Data normalization is particularly vital.
By properly organizing key data such as customer, product, transaction, and financial information, companies can expect to improve operational efficiency and the quality of decision-making.
However, because data normalization requires an enormous amount of man-hours, it is necessary to leverage external experts and systems.
By doing so, it is possible to efficiently improve data quality through automated data maintenance and phased implementation.
Recognize that the success of core system modernization lies not merely in updating systems, but in data preparation and its effective utilization; approach this strategically.
