# What Exactly is an Insight? A Literature Review

arXiv: https://arxiv.org/abs/2307.06551 · HTML: https://mlsanity.com/paper/2307.06551/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Jul 2023.

## Built on by (most cited first, 3 of 3)

- [Where Are We So Far? Understanding Data Storytelling Tools from the Perspective of Human-AI Collaboration](https://mlsanity.com/paper/2309.15723/) (2023)
- [Data-to-Dashboard: Multi-Agent LLM Framework for Insightful Visualization in Enterprise Analytics](https://mlsanity.com/paper/2505.23695/) (2025)
- [Beyond Description: A Multimodal Agent Framework for Insightful Chart Summarization](https://mlsanity.com/paper/2602.18731/) (2026)

## Abstract (arXiv)

Insights are often considered the ideal outcome of visual analysis sessions. However, there is no single definition of what an insight is. Some scholars define insights as correlations, while others define them as hypotheses or aha moments. This lack of a clear definition can make it difficult to build visualization tools that effectively support insight discovery. In this paper, we contribute a comprehensive literature review that maps the landscape of existing insight definitions. We summarize key themes regarding how insight is defined, with the goal of helping readers identify which definitions of insight align closely with their research and tool development goals. Based on our review, we also suggest interesting research directions, such as synthesizing a unified formalism for insight and connecting theories of insight to other critical concepts in visualization research.