# From Results to Decisions

> Read research results, explore their evidence, and decide whether to generate a report.

Source: https://imario.ai/docs/run-and-analyze/overview

A completed research run produces a result in the conversation. You can inspect it before requesting a narrative report.

## Read in this order

1. **Question and stimulus:** confirm what participants answered or saw.
2. **Audience and count:** confirm whose responses are included.
3. **Distribution or themes:** read the main pattern and the disagreement.
4. **Individual answers:** check the evidence behind the summary.
5. **Warnings and limits:** note missing material, partial results or limited evidence.

The result's shape follows the method. A closed question shows choices; an open question shows responses and coded themes, led by the values people named when the question asked for one; a focus group includes a discussion. Do not expect the old Quick Test GO/REVISE/KILL scorecard on every current result.

## Explore before summarizing

Open [Explorer](https://imario.ai/docs/run-and-analyze/explorer) to select responses, inspect subgroups and use breakdowns. A pooled result can hide opposing patterns across audiences.

Ask for an explanation of existing evidence when a finding is unclear. Ask a new research question when the evidence does not answer it. [Follow-up research](https://imario.ai/docs/research/run-and-follow-up) has a new scope and may incur another charge.

## Create a report when needed

Use [report generation](https://imario.ai/docs/run-and-analyze/reports) to synthesize completed runs in the conversation. Choose the included runs deliberately. A report is a separate output with its own price, not a prerequisite for viewing research results.

Before sharing, keep the audience definition, question wording and synthetic nature of the responses with the finding. See [Share and export](https://imario.ai/docs/run-and-analyze/share-and-export) and [Evidence](https://imario.ai/docs/methods/evidence).