Ask the question.
Get past the SQL queue.
A secure text-to-SQL solution lets business users query internal data in plain language. First user tests showed answers arriving in minutes rather than weeks.
A large public institution responsible for social security oversight
The challenge
In a large public institution responsible for social security oversight, employees regularly needed access to company data for inspections, audits or policy work. Typical questions included:
"Show all company registrations in the Ghent region from the last quarter."
These requests were traditionally routed to the central data analysis team, who translated them into SQL queries by hand. This created bottlenecks: employees often waited days or even weeks for answers, which slowed down decisions and increased their dependency on technical staff.
The solution
Together with the organisation, Common Sense AI developed a text-to-SQL solution powered by generative AI.
Users ask questions in natural language. The system automatically converts them into SQL queries and runs them on internal databases, giving non-technical users direct access to real-time data without writing code.
Key features:
- Full local deployment to keep the data secure
- A working demo based on open flight data, to show the system's real-time capabilities
- Tailored integration with internal systems, for reliability and scalability
The result
First tests with business users, including users without SQL knowledge, showed immediate and measurable benefits:
- Query response time reduced from weeks to minutes
- A significant reduction in workload for the data analysis team
- More satisfied internal users, thanks to more autonomy and faster insights
The project was a successful step towards democratising data access within the organisation. Future iterations are planned to improve the user experience and expand coverage to more data domains.