> For the complete documentation index, see [llms.txt](https://docs.drivepoint.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.drivepoint.io/using-the-web-app/modeling-agent.md).

# Modeling Agent

Ask Neo chat to preview a what-if on a plan, then save it when you confirm.

Ask a what-if in plain language. Drivepoint previews the change on your plan's key drivers and shows the effect on revenue, margin, and cash. The workbook stays as it is until you confirm a save.

{% embed url="<https://www.loom.com/share/e7718eb7777749cc9280490eef20f89b>" %}

## Ask from Neo chat

1. Open **Chat**.
2. Name the plan when you do not mean the live model. The live model is the current plan.
3. Ask the question. For example:
   * "What if AOV drops by $3?"
   * "What if we cut Meta spend by 20% next quarter?"
   * "What if shipping costs for DTC double?"
   * "What if subscription orders grow 15%?"
4. Review the preview. It lists the driver changes and the recalculated key results. The workbook is not modified.
5. Say so if you want to keep it. You can write the change onto the same plan, or save a new plan in the same folder. Drivepoint asks you to confirm before it writes.

Open a plan under **Models** and click **Summarize plan** to start a chat about that plan's key drivers, key results, risks, and next actions.

### In the legacy app

The Plans page still has an **Agent** that edits one plan at a time. Neo chat uses the same preview and save, and it can also summarize the plan and save the result as a new plan.

## Scenarios that work well

* Revenue and demand: AOV, CAC, and the mix of first-time and returning orders
* Cost and margin: COGS percent, fulfillment cost, and ad spend
* Operating expenses: headcount, software, and marketing cuts
* Cash: payment terms, days payable outstanding, and shipping costs

{% hint style="info" %}
Be specific. "Cut Meta ad spend by 20% next quarter" works better than "reduce marketing." Include a time frame, such as "next quarter" or "in 2026."
{% endhint %}

## Teach it your drivers

When you add or update key drivers on a plan, the agent picks them up, usually within a few minutes.

1. Open the **Product** tab in the plan.
2. Mark the new inputs as key drivers. Examples are unit price, product launch date, and discount rate.
3. Ask the what-if again. The agent uses those drivers without a separate setup step.

Use this for new products or channels, pricing, and cost levers that are specific to your company.

[Learn more about key drivers and results](https://docs.drivepoint.io/help/working-in-the-smartmodel-tm/add-key-drivers-+-results).

## FAQ

<details>

<summary>Does it write the change automatically?</summary>

No. The first answer is a preview. The workbook changes only after you confirm a save onto the existing plan or a new plan.

</details>

<details>

<summary>What AI models are you using?</summary>

Drivepoint is integrated with Gemini, OpenAI, and Anthropic.

</details>


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.drivepoint.io/using-the-web-app/modeling-agent.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
