Agrex announced its AI Customer Data Quality Diagnosis Service on September 28, 2026, to analyze corporate customer data and recommend improvements. It launches October 1, 2026, at prices starting from ¥800,000 before tax; fees vary with the volume of data assessed. Agrex says the service shortens data quality assessments that previously took about two weeks to as little as two days.

The service checks customer data stored in customer master databases, CRM, and SFA systems for duplicates, missing information, and inconsistent formats. It scores quality across five dimensions: completeness, uniqueness, validity, degree of normalization, and address quality, helping companies assess whether their data is ready for use. It also performs basic record linkage by matching names, addresses, and phone numbers to estimate the scale of duplicate data.

Using the quality analysis and record-linkage results, generative AI produces a diagnostic report outlining improvement priorities and possible responses. Agrex says the service aims to show companies what they should improve. Development combined Agrex’s expertise in data cleansing and record linkage with Lighthouse’s know-how in applying generative AI. The companies cited a problem in which some organizations fail to achieve expected results from generative AI and RAG (retrieval-augmented generation) because their data quality is insufficient.

Agrex announced its AI Customer Data Quality Diagnosis Service on September 28, 2026, to analyze corporate customer data and recommend improvements. It launches October 1, 2026, at prices starting from ¥800,000 before tax; fees vary with the volume of data assessed. Agrex says the service shortens data quality assessments that previously took about two weeks to as little as two days.

The service checks customer data stored in customer master databases, CRM, and SFA systems for duplicates, missing information, and inconsistent formats. It scores quality across five dimensions: completeness, uniqueness, validity, degree of normalization, and address quality, helping companies assess whether their data is ready for use. It also performs basic record linkage by matching names, addresses, and phone numbers to estimate the scale of duplicate data.

Using the quality analysis and record-linkage results, generative AI produces a diagnostic report outlining improvement priorities and possible responses. Agrex says the service aims to show companies what they should improve. Development combined Agrex’s expertise in data cleansing and record linkage with Lighthouse’s know-how in applying generative AI. The companies cited a problem in which some organizations fail to achieve expected results from generative AI and RAG (retrieval-augmented generation) because their data quality is insufficient.