Market report summariser
Twenty market reports, one briefing.
- Client
- A large property investment and development business
- Sector
- Property and development
- Status
- Delivered
Business problem
What problem are we actually solving?
Investment and development decisions rely on market intelligence spread across broker reports, internal research and third-party data. Reading all of it for every submarket was impractical, so teams sampled.
Target users
Who is this for, and what does their day look like?
Development managers, investment and research teams.
Current process
How does it work today, and where does it hurt?
Manual reading and ad hoc note-taking; insight quality depended on the reading time available. No consistent structure across submarkets or reports.
Desired outcomes
What does better look like?
- Consistent briefing per submarket: supply, demand, absorption, rents, pipeline, notable transactions.
- Ability to compare across reports and flag where sources disagree.
- Source attribution on every data point.
Constraints
What limits, systems, data or rules apply?
- Multiple report formats and publishers.
- Every figure must be attributed to its source and date, because market data ages fast.
- No extrapolation beyond what the reports state.
Success metrics
How will we know it worked?
- Reports covered per briefing, where teams had previously sampled.
- Briefing preparation time.
What was built
A multi-document summarisation agent with a fixed briefing schema, source-and-date attribution, and a conflict flag where reports disagree.