Iris gives you analyst-grade answers when you tell it four things: what you want, where to look, which time period, and how to show the answer.
Iris is the research assistant inside Multibagg AI. It can read official company documents, pull numbers from financial tables, and search the web. A vague question gets a vague answer. A clear question gets a clean table or a short summary you can use straight away.
This guide shows you how to ask, step by step.
The golden rule
Every good question to Iris answers these four things.
Miss one, and Iris has to guess. Include all four, and the answer comes back the way you need it.
A template you can copy
Fill in the blanks and send. That is all it takes.
Pull [metric or topic] for [company] from [document or source] for [time period]. Show as [format].
Filled in, it looks like this:
Pull Revenue, EBITDA and PAT for Q3 FY26 from the Investor Presentation of DMART. Show as a table with QoQ and YoY.
Here is how that example maps to the golden rule:
- What: Revenue, EBITDA and PAT
- Where: DMART's Investor Presentation
- When: Q3 FY26
- Output: a table with QoQ and YoY
Where: tell Iris where to look
Iris can read several sources. Naming one makes the answer far more accurate, because Iris knows exactly which document to trust.