Gerhard G. StocKinger · Published on January 27, 2025

Many people consider MS Access to be a dinosaur. Nevertheless, I encounter the platform regularly in my day-to-day business life—often in companies where it has reliably supported a specific process for years. At an Austrian SME with about 50 employees, that was exactly the starting point: a database that had grown over time, a wealth of practical knowledge, and the desire to process PDF invoices, quotes, contacts, and master data more quickly.

Understand the value first, then modernize

A complete replacement sounds like a good idea, but it isn't necessarily cost-effective. Existing solutions contain rules, exceptions, and lessons learned that aren't documented anywhere. Anyone who replaces the system too hastily risks losing that knowledge.

The pragmatic approach, therefore, was a controlled expansion: documents are read via an AI interface, relevant fields are structured, and then reviewed by a human before being imported into the database. Access remains the user interface; the AI complements a clearly defined capability.

Four Questions to Ask Before Every AI Expansion

• Is the existing process understood from a technical perspective?
• Are data fields and quality rules clearly defined?
• What types of content are allowed to leave an external API?
• Who reviews unclear or contradictory results?

The result isn't a spectacular demonstration, but a practical improvement. Modernization isn't a race to adopt the latest tool. It's the ability to sensibly enhance existing systems while simultaneously laying the groundwork for a realistic migration path.

Conclusion: Legacy does not necessarily mean outdated. A system becomes problematic when it is undocumented, insecure, or can no longer be maintained. As long as it provides value, a carefully defined AI integration can be the economically sound next step.

This post is based on my LinkedIn post from January 27, 2025, and has been expanded for the blog.