Everything you wanted to ask about EPA.

EPA (Excel Process Automation) turns your existing Excel-based work — forms, reports, approvals — into a governed web application with a database, multi-user access, audit trails, and AI-assisted form generation. You keep the Excel UX your team already knows; you gain enterprise scale.

Excel becomes a bottleneck when your team grows: version conflicts, lost data, no audit history, fragile macros, dead files in shared folders. EPA fixes this without forcing your team to learn a new tool — Excel stays as the front door; the back end becomes a proper, secure database.

RPA automates mouse clicks and keystrokes on top of legacy systems. EPA automates the underlying data process — generating the database, the input form, and the report from your existing Excel output. RPA glues old systems together; EPA replaces the spreadsheet sprawl underneath.

Those are visualization tools on top of a data warehouse you already have. EPA builds the data layer itself — taking the Excel file you actually use as the spec, generating the DB tables, input UI, and reports. EPA is upstream of BI; you can absolutely pair it with your existing BI tool.

Finance teams, operations, plant managers, project controls, HR, anyone with 30+ Excel files circulating between people. Specifically: organizations where Excel is the real system of record but it's becoming a liability.

That is exactly the combination we are built for. EPA turns your spreadsheets into a real database with a governed web UI for data entry. TRINITY/PRISM sits on top and answers questions in plain language. The same workflow that took 3 weeks in Excel — input, consolidation, reporting — becomes a one-click pipeline. Each product stands alone; running them together is where the magic happens.

They really use it. EPA's input UI looks like Excel because users already know Excel. No new tool to learn — they keep typing into grids, but now there is a real database underneath with audit trails and permissions. Admin certification is a half-day for IT; end-user training is one hour. Most teams skip the training session after seeing the first form.

You hand EPA the Excel output you currently produce. Define the columns you need (Korean or English column names + 1–2 sample rows is enough). EPA's AI generates the CREATE TABLE DDL, the input form UI, and the report logic — automatically. Your team uses the new web app exactly the way they used to type into Excel.

About one week from first conversation to business users actually entering data. EPA pre-automates ~17 setup steps (DDL, schema, form, list view, permissions, audit, deployment). What used to be a 3-month custom dev project is a 1-week guided rollout. This is the headline number — we say it because it is true.

Yes. The point isn't to replace Excel — it's to scale it safely. Users keep entering data in Excel-style grids; reports still export to Excel; formulas still work. The difference: everything is now in a real database, audited, and accessible by your whole team.

EPA assumes real-world Excel. You can define the data structure manually from any output sample — even one with merged cells or cross-tab layouts. The 1st phase converts it to a clean list-based input; the 2nd phase can add cross-tab input back if your team prefers it. You don't have to clean Excel first.

The AI ingests your Excel output spec + sample data → infers column types (text / number / date / FK) → suggests primary keys → generates English column names → emits DDL and a default form/list UI. A human reviews and approves before deployment. Sample data improves accuracy dramatically.

Oracle, MS SQL Server, PostgreSQL, MySQL, MariaDB, SAP HANA, Tibero — and any other JDBC-compatible source. EPA can either create its own dedicated DB or run alongside your existing schemas (read-only or read-write).

Yes. EPA can read from / write back to SAP, Oracle ERP, Dynamics, custom in-house ERPs, and CRMs (Salesforce, HubSpot). Common pattern: ERP holds master data, EPA handles the operational data layer that ERP doesn't cover well (project tracking, ops logs, ad-hoc workflows).

Every EPA-generated app exposes a standard REST API for its tables and forms (GET/POST/PUT/DELETE), with OpenAPI/Swagger documentation auto-generated. Webhooks are supported for real-time downstream pushes.

Yes — bi-directional. Users can import an existing Excel sheet to bulk-insert into the database, and export any list/report back to Excel (.xlsx) one-click. EPA also supports CSV and HWP (Korean office format).

Lists and dashboards refresh in real time as users save records. For external integrations, EPA supports webhooks, scheduled syncs, and DB triggers. Sub-second response on typical OLTP workloads.

Role-based access control (RBAC) at the table, row, and column level. Inheritable role hierarchies, dynamic groups (e.g., "managers of cost center X"), and field-level masking for sensitive data. Permissions are versioned and auditable.

Yes — SAML 2.0, OIDC, OAuth 2.0, LDAP, Active Directory. Common SSO providers (Azure AD/Entra ID, Okta, Google Workspace, KakaoWork) are pre-tested. Provisioning via SCIM also supported.

Every insert, update, delete, login, permission change is logged with user, timestamp, before/after values, and IP. Logs are queryable in-product or exportable to your SIEM. Retention is configurable per data class.

At-rest: AES-256 encryption (DB and file storage). In-transit: TLS 1.2+. Column-level encryption for PII/PCI fields. Configurable masking by role. Optional integration with HSM / KMS (Korean GFK, AWS KMS, etc.).

BI MATRIX is KOSDAQ-listed (413640). EPA deployments support K-ISMS, ISMS-P, GDPR (data residency options), SOC 2 readiness. Many of our public-sector and financial customers run EPA inside their own VPC/data center with no external connectivity.

Three tiers: Personal / Teams / Enterprise. Pricing scales with users + processes. 14-day free trial on all tiers. For enterprise rollouts (50+ users or self-hosted), we issue a tailored quote — contact sales for an NDA-friendly conversation.