LBC: Japan's Largest Corporate Information Database with 12.5 Million Records
The LBC Corporate Information Database is a comprehensive corporate database covering 12.5 million business locations, independently collected and built by uSonar. Based on the philosophy of organizing information for every company in Japan, the database contains over 38 million records when including historical data.
Background of the LBC Corporate Information Database
In recent years, the pace of environmental change has accelerated, necessitating the redesign of business processes to adapt to these rapid shifts. Nevertheless, many companies have historically avoided addressing the challenge of business process redesign, attempting to manage without it.
On the other hand, some companies have faced these challenges head-on, viewing them as opportunities to build mechanisms for next-generation growth. We understand that these companies did not start from scratch; rather, they built the foundation for growth by optimizing internal resources and effectively utilizing external resources within existing frameworks.
To support this, uSonar has built and provided the LBC Corporate Information Database to serve as the foundation for sales and marketing activities. A foundational database must be built to be Mutually Exclusive and Collectively Exhaustive (MECE). Databases with duplicates lead to wasteful redundant approaches, while incomplete databases result in missed opportunities, which remains a challenge for many companies.
By using the LBC Corporate Information Database, you can achieve:
- Consolidation of Duplicate Data
- Integration and Centralized Management of Scattered Databases
- Enrichment of Customer Data with Attribute Information
- Maintenance of Data Freshness and Reduction of Internal Operational Burden through Automatic Corporate Information Updates
- Standardization of Data Entry Rules and Reduction of Operational Burden
- Customer Trend Analysis (Profiling)
- Grasping Transaction History by Company or Group and Identifying Untapped Locations
- Market Share Analysis
- Creation of Approach Lists
is realized.
Database Specialized for Customer Data Management and Marketing
One of the most prominent features of the LBC corporate database is its volume of 12.5 million records. This is because the LBC corporate database is specifically designed for customer data management and marketing purposes.
For instance, when managing customer information, the LBC database serves as a reference guide. If this reference guide (LBC) lacks sufficient data, it becomes impossible to accurately manage or clean your customer records. High-precision customer data management cannot be achieved under such conditions.
Furthermore, when planning marketing strategies or executing outreach, it is essential to first view the entire market to identify untapped markets and non-client companies. To do this, you need a database that provides a comprehensive view of the entire market.
For these reasons, we have built a database containing 12.5 million records. Additionally, the LBC corporate database includes a vast amount of data beyond private enterprises, such as central government agencies, local municipalities, schools, hospitals, public interest corporations, social welfare corporations, and religious organizations. While corporate credit research firms are well-known for their business databases, data on non-private entities is often less critical for credit assessment purposes. In contrast, because the primary purpose of the LBC corporate database is customer data management and marketing rather than credit assessment, we actively collect and include this diverse range of data.
LBC Corporate Database Feature (1): Data Cleansing Expertise
We achieve a high matching rate by utilizing our accumulated knowledge (cleansing expertise) during the matching process between your customer database and the LBC corporate information database. This allows for the maintenance and organization of a larger volume of data.
Furthermore, using our long-cultivated data cleansing technology, we assist in visualizing identical or affiliated companies by assigning common keys* and latitude/longitude information to the registry data (over 10 million records) used in statistical surveys conducted by the Ministry of Economy, Trade and Industry.
(Ministry of Economy, Trade and Industry Begins Using the LBC Corporate Database)
*: Common Key
A unique identifier used to specify a head office or business location (including sole proprietorships that do not possess a Corporate Number).
1. Textbooks and Dictionaries for Data Maintenance
To maintain customer data, rules for data maintenance must be established in advance. This is because data integrity cannot be ensured without consistent management rules. In reality, however, database management rules are rarely consistent; they are often inconsistent not only across multiple databases but even within a single database. Under these conditions, data cannot be properly maintained. Therefore, data maintenance rules are essential. Consider the LBC corporate information database as the textbook for these rules. By maintaining your customer database according to rules aligned with this textbook, you can achieve data consistency.
2. Matching Dictionary
To apply the textbook rules to your customer database, a matching process between the customer database and the textbook is required. However, since the customer database and the textbook operate under different rules, they will not match correctly if processed as-is. Therefore, a dictionary (Knowledge Master) is necessary to bridge the customer database and the textbook. By entrusting your customer data to uSonar, we convert the rules according to our dictionary, enabling a successful match between the disparate rules of your customer database and our textbook.
3. Dictionary 1: Historical Information
A common cause of failure when matching databases is the issue of data freshness, such as when one dataset is outdated. For example, if customer data contains a former company name or an old address, it cannot be matched with the LBC corporate information database in its current state. Therefore, the LBC corporate information database utilizes historical information as a dictionary. Specifically, by using a master (dictionary) that accumulates past events such as company name changes, relocations, and mergers, we automatically convert and match customer data even if it contains old company names or addresses.
4. Dictionary 2: Company Name Knowledge
Similar to historical information, matching failures can occur due to the input of abbreviations or commonly misspelled company names. For example, this includes cases where 'NEC' is entered instead of 'Nippon Electric Company' or 'All Nippon Airways' is entered as 'ANA'. At uSonar, we have accumulated these 'data entry habits' as knowledge through the matching tasks we have performed in the past. Even if your company's customer data contains abbreviations or common misspellings, this knowledge allows us to automatically convert them to official company names and perform accurate matching.
5. Dictionary (3): Formatting
While this may seem trivial compared to the previous section, minor variations in notation can also lead to unsuccessful data matching. At uSonar, we use data cleansing technology to standardize address notations (such as converting common address formats to official ones). Furthermore, we can also process the conversion of old addresses to new addresses resulting from municipal mergers.
As demonstrated above, even if customer data contains typos, omissions, former company names, old addresses, or pre-merger information, we utilize our proprietary dictionaries (various knowledge bases) to perform matching with the LBC corporate information database, thereby achieving high matching accuracy.
By using the 'dictionaries' of the LBC corporate information database in this way, the duplicate detection rate is improved. So, what level of improvement can be expected? While this is for reference only, let us compare the results of deduplicating the business card information entered by uSonar sales representatives—which we have been using as an example—with and without the use of the LBC corporate information database.
When using the LBC corporate information database, 1,038 out of 22,128 records (4.7%) can be identified as duplicates, whereas only 458 records (2.1%) can be identified without it. This difference is due to the dictionaries that enhance cleansing accuracy.
LBC Corporate Information Database Features (2): Coding System
The LBC corporate information database groups business locations that have relationships such as headquarters-branch or parent-subsidiary. This enables data management at the group level, allowing you to consolidate sales information by group to understand gross transaction volume, or identify non-trading locations within a group for cross-selling targets.
When organizing customer information, linking data that shares relationships such as headquarters-branch or parent-subsidiary is extremely important. Without this linkage, it is impossible to grasp transaction amounts at the group level or identify non-trading locations of existing customers to plan outreach strategies.
Generally, a method is adopted where a parent code such as '0001' is assigned to the headquarters, and branch codes such as '0001-001' are assigned to business locations to represent the group. However, with this type of management, the coding system gradually breaks down due to mergers, acquisitions, or the closure of business locations, eventually becoming unmanageable.
To avoid such situations, the LBC corporate information database assigns a unique, immutable code to all business location data and adopts a coding system that links relationships such as headquarters-branch or parent-subsidiary. This linkage is the reason for the name 'Linkage Business Code' (LBC).
Specifically, the details are as follows.
Rule 1. A Unique Identification Number Is Assigned to Every Business Location
First, a unique code (identification number) is assigned to every business location. This code is referred to as "LBC." LBC serves as both the name of the corporate information database and the name of the management code. To avoid confusion, we will distinguish between the two: the master data itself will be referred to as the "LBC Corporate Information Database," and the management code will be referred to as the "LBC Corporate Management Code." Since the LBC Corporate Management Code is assigned individually to each business location, no two locations share the same number, and these numbers are permanent and immutable.
*1 The actual code consists of 11 digits.
*2 Because the 12.5 million records are based on business locations rather than companies, separate codes are assigned to different sites of the same company, such as a headquarters and a factory (the LBC Corporate Information Database contains data consolidated at the site, company, and corporate group levels).
Rule 2. The Headquarters of Any Business Location Is Represented by the Headquarters LBC
For example, regarding the LS Oil Matsuyama Branch (0012), the Headquarters LBC field shows 0007, indicating that it is a business location of LS Oil (0007). Similarly, for LS Oil (0007), the Headquarters LBC is its own code (0007), which indicates that LS Oil (0007) is the headquarters.
Rule 3. The Parent Company of Any Business Location Is Represented by the Parent Company LBC
For LS Oil (0007), the Parent Company LBC field shows 0002, indicating that it is a subsidiary of LS Bank (0002). Likewise, the Parent Company LBC field for LS Bank (0002) is 0001, indicating that it is a subsidiary of LS Trading (0001).
Rule 4. The Top of the Capital Group Is Represented by Its Own Code in All Fields
For LS Trading (0001), the Headquarters LBC, Parent Company LBC, and Group LBC fields all show 0001, indicating that it is at the top of the group hierarchy.
Rule 5. In the Event of a Group Relationship, Only the Grouping Information Is Updated
Even if changes occur in corporate affiliations due to mergers, acquisitions, or reorganizations, only the grouping information is updated, while the Corporate Management Code LBC remains unchanged (the Corporate Management Code LBC is immutable). By keeping the assigned identification number (Corporate Management Code LBC) constant and updating the attribute information, we maintain the consistency of the code system.
By utilizing this code system, you can group the client data entrusted to us by our customers, enabling data consolidation at both the individual company and corporate group levels.
