Answers from council data, without the report queue.
An Australian govtech company gave council staff a way to ask questions about their data in plain English and get tables or charts when useful.
See what changedInside the workflow
Australian govtech client
- 01Prepare the data
- 02Enforce permissions
- 03Answer and visualize
The change
Self-service data access
Client
Australian govtech company
Use case
Council data access
Deployment
Client AWS environment
01 / The challenge
The work before the agent
The client’s platform held years of council operations data. Council staff could only get answers by asking the vendor to build a report or dashboard.
Every request consumed the vendor team’s time and made council staff wait. Users had no way to explore the data themselves, even for routine questions.
02 / The approach
What we built
We built a conversational agent on top of the existing data infrastructure. It interprets a question, queries the data available to that user and can return an answer, table or chart.
Prepare the data
An ETL pipeline curates queryable columns and adds computed fields that match how council staff think about the data.
Enforce permissions
Access rules are applied at the query layer. A parks manager, for example, can see parks data without gaining access to HR payroll.
Answer and visualize
The interface accepts plain language questions and returns a table or chart when that makes the answer clearer.
Learn from gaps
Unanswered or poorly answered queries are captured for review, giving the client a way to improve the system and understand what users need.
03 / The result
What changed
Direct access
Council staff can ask questions about their data in plain language.
Useful formats
Answers can be presented as text, tables or charts.
Permission aware
The query layer applies the client’s access rules.
Council staff can explore their own data without waiting for a custom report. That frees the vendor team to spend more time on the product as more councils use it.
The deployment in the customer’s own cloud lets its team apply existing infrastructure controls and review data residency in that environment.
Under the hood
The implementation
A closer look at the systems and controls behind this story.
- Natural language to SQL
- Dynamic charts
- ETL and schema curation
- Role-based query access
- Unanswered-query review
- Client-managed AWS deployment
Your workflow next
Can your customers find their own answers?
Show us the report requests and the access rules behind them. We will map how customers get answers from their own data.