> For the complete documentation index, see [llms.txt](https://docs.coupler.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.coupler.io/ai/data-analysis.md).

# Data analysis

Once you've connected an AI tool as a destination, you can explore and analyze your data through natural language conversation — no SQL or formulas required. You ask questions in plain English, and Coupler.io runs the query, verifies the calculations, and returns validated results.

{% hint style="info" %}
Every answer is backed by Coupler.io. Your AI tool never queries your source systems directly — it asks Coupler.io, which executes the query and returns only checked results. This keeps numbers consistent and trustworthy.
{% endhint %}

## What you can do with AI

* **Ask questions in plain language** — "What were my top 5 campaigns by spend last month?" or "Which deals are stalled in the pipeline?" — and get answers, summaries, and insights back in the conversation.
* **Discover what data is available** — ask what datasets you have and what each one contains.
* **Understand the structure** — get a plain-language breakdown of the columns, their meaning, and how to use them.
* **Rely on built-in domain expertise** — your AI tool is automatically primed to focus on the right metrics and dimensions for your data.
* **Build lasting context** — document a dataset once so every future conversation already understands it.
* **Generate reports** — turn your analysis into a structured, decision-ready report that's sanity-checked for arithmetic, units, and consistency before you see it.
* **Refresh your data** — pull the latest data on demand without leaving the conversation.

## Available in

This works in every Coupler.io AI destination: Claude, ChatGPT, Cursor, Custom MCP, Gemini CLI, OpenClaw, and Perplexity.

## What your AI tool can see

{% hint style="info" %}
How much data is visible depends on how you're working:

* **In your AI tool (a destination)** — it sees only what you send to it: the specific dataset (or datasets) delivered to that destination.
* **With the Coupler.io AI agent** — it sees every dataset in a data flow, or all datasets across your Coupler.io account, so it can reason across many datasets at once.
  {% endhint %}

## Explore and ask questions

Start a new conversation in your AI tool and just ask — you don't need to name a dataset or know where the data lives.

{% stepper %}
{% step %}
**Ask your question.** Describe what you want to know in plain language. The AI figures out which dataset best answers it. If nothing matches — or several datasets could fit — it asks you which to use.
{% endstep %}

{% step %}
**It brings in domain expertise.** As it works, the AI automatically loads knowledge for your data's domain — for example, ecommerce or marketing analytics — so it focuses on the metrics and dimensions that matter for your field instead of treating your data as generic rows and columns. There's nothing to configure.
{% endstep %}

{% step %}
**Get your answer — and dig deeper.** The AI returns answers, summaries, and insights. Follow up naturally to filter, group, compare periods, rank, or drill into the results.
{% endstep %}
{% endstepper %}

## Context that gets better over time

You can ask your AI tool to **document a dataset**. It explores the data and asks you a few questions to capture the business context it can't infer on its own — how a metric is defined, which rows to deduplicate, or what a placeholder value really means.

What it learns is attached to the dataset itself, not just the current chat — so every future conversation starts already understanding it, even in a different AI tool. You never have to re-explain your data.

{% hint style="info" %}
Where something needs a human to confirm, the AI leaves a clear **to-do for you** — marked to verify with your team — so you know exactly what to fill in.
{% endhint %}

You can also ask it to **improve the column labels and descriptions** for a dataset, so fields are clearer and easier to reason about in every future analysis.

## Generate a report

Ask your AI tool to write up your analysis as a report — a weekly performance summary, a monthly business review, an executive summary, or a TL;DR. The result is a scannable, structured document, and the numbers are validated (arithmetic, units, and claims are checked) before it's delivered.

## Get fresh data on demand

If your data might be out of date, ask your AI tool to refresh the data flow. It triggers a run and fetches the latest data from the source, so your next question is answered against current numbers.

{% hint style="success" %}
Want it to stay current on its own? Ask your AI tool to run the flow on a schedule — daily, hourly, every 15 minutes, or monthly.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.coupler.io/ai/data-analysis.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.