> For the complete documentation index, see [llms.txt](https://docs.zigpoll.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.zigpoll.com/integrations/mcp/mcp-chatgpt.md).

# Connect to ChatGPT

Connect your Zigpoll surveys to **ChatGPT** as a custom MCP connector using the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/). Once connected, ChatGPT can read your survey responses, run analytics, and create surveys — just ask in plain English.

For other AI clients, see the [MCP overview](/integrations/mcp.md).

{% hint style="info" %}
Custom MCP connectors require **Developer Mode**, available on ChatGPT **Pro, Plus, Business, Enterprise, and Education** plans on the web (not the mobile apps). Menu labels evolve as ChatGPT updates — "Apps" and "Connectors" refer to the same area. On **Business and Enterprise** plans, a workspace admin may first need to allow custom MCP connectors in the workspace permission settings.
{% endhint %}

## Step 1: Enable Developer Mode

1. In ChatGPT on the web, open **Settings → Apps & Connectors** (may appear as **Connectors**).
2. Open **Advanced settings** and turn on **Developer mode**.

## Step 2: Add the Zigpoll connector

1. Back in **Apps & Connectors**, click **Create** (or **Create app**).
2. Enter the details:
   * **Name:** `Zigpoll`
   * **MCP Server URL:** `https://mcp.zigpoll.com/mcp`
3. Set the authentication type to **OAuth**.
4. Check the box confirming you trust the provider, then click **Create**.
5. From the connector's details page, click **Connect** (or **Connect Zigpoll**). A Zigpoll authorization tab opens:
   * **If you're already logged into Zigpoll** in that browser, authorization completes automatically — you'll briefly see a Zigpoll loading screen, then be returned to ChatGPT with a "connection successful" message. No API key or extra steps needed.
   * **If you're not logged in**, you'll see a Zigpoll sign-in page. Sign in with your email and password, Google, or Shopify, and authorization completes from there.

{% hint style="warning" %}
A successful connection is **not** the last step. The connector must also be enabled inside each conversation (Step 3 below) — otherwise ChatGPT will say it can't access your Zigpoll data even though the connection succeeded.
{% endhint %}

{% hint style="info" %}
`mcp.zigpoll.com` uses a standard browser-based OAuth login tied to your Zigpoll account. You never paste an API key into ChatGPT, and a static API-key header is not supported — the browser flow exchanges your Zigpoll login for a secure token.
{% endhint %}

## Step 3: Enable Zigpoll in your chat

Custom connectors are only active in conversations where you turn them on:

1. Start a **new chat** and click the **+** (plus) button in the message box.
2. Choose **More → Developer mode** (may appear directly as **Developer mode**).
3. Toggle on the **Zigpoll** connector for the conversation.
4. Ask away. Being explicit helps ChatGPT pick the right tool, e.g.:

> "Use the Zigpoll app to list my accounts"

> "Use Zigpoll to summarize the responses from my NPS survey"

A good first prompt is *"List my Zigpoll accounts"* — it confirms the connection works and that you're connected to the right Zigpoll account.

## Example prompts

* "List all my surveys and their response counts"
* "Where are my survey respondents located?"
* "Compare order values for each answer on my attribution question"
* "Create a new customer satisfaction survey"

For the full list of tools and more examples, see the [MCP overview](/integrations/mcp.md). For ideas on what to build, see [Ideas for Using Zigpoll MCP](/tutorials/zigpoll-mcp-use-cases.md).

## Troubleshooting

* **"ChatGPT says it can't access Zigpoll" or "not connected" after a successful setup** — this is the most common issue, and it almost always means the connector isn't enabled in the current conversation. Start a new chat and enable Zigpoll via **+ → Developer mode** (Step 3). The connector stays on for the rest of that conversation, but each new chat needs it enabled again.
* **The authorization tab flashed and closed without asking for anything** — that's expected. If you were already logged into Zigpoll, authorization completes automatically; the "connection successful" message means it worked.
* **Developer mode not visible** — confirm your plan supports it and that you're on the web app, not the mobile app. On Business/Enterprise plans, ask a workspace admin to enable custom MCP connectors for the workspace.
* **Connected to the wrong Zigpoll account** — authorization uses whichever Zigpoll account is logged into your browser. Log into the correct account at [app.zigpoll.com](https://app.zigpoll.com), then disconnect and reconnect the connector in ChatGPT.
* **Authentication errors** — the connector authenticates with an API key labeled "MCP" that Zigpoll creates automatically. If it was revoked in **Settings → Manage Integrations → API Key**, disconnect and reconnect the connector to issue a new one.


---

# 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.zigpoll.com/integrations/mcp/mcp-chatgpt.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.
