# ChatGPT and Perplexity brand tracker

Give your brand, your competitors and the questions your buyers ask. It asks OpenAI and Perplexity, with web search on, and measures how often your brand is named, in which place, against whom, and which sources the answers cite.

**From USD 5.00 per 1,000 answers.** On the Apify Store since 18 November 2026. Checked 18 November 2026. Apify Store page: https://apify.com/tan-solutions/ai-visibility

## What it does

More buyers now ask a chat assistant which tool to pick before they ever open a search page. If the answer names three of your competitors and not you, you never hear about that buyer. This tool asks the questions your buyers ask, several times each, and counts what comes back.

You bring your own OpenAI or Perplexity key, or both. The tool uses it only inside your own run to ask that engine, and you pay the vendor for that usage on your own account. Tan Solutions never receives the key. The tool charges only for its own work: reading each answer, matching the brands, and the summary and trend rows.

- **Asks each question several times.** Answers vary from one ask to the next, so each question goes to each engine 1 to 10 times (3 by default) for a stable rate.
- **Uses the engines with web search.** OpenAI through its Responses API with web search on, and Perplexity through its Agent API with web search on. You can pick the model; the instructions and search settings are fixed.
- **Finds your brand and your competitors.** Whole words only, case and accents ignored, with the other names you give. "Notion" matches "notion" and "Notión" but not "notional".
- **Sums up per question.** One free summary row per question and engine: mention rate, average position, share of voice against your competitors, citation rate and the 5 most cited domains.

## Who it is for

- Marketing and SEO teams who want to know whether chat assistants recommend them, and who they recommend instead.
- Product teams who want to see which sources the answers in their category trust, so they know where to be mentioned.

## What it costs

**From USD 5.00 per 1,000 answers.** You pay per result. The price depends on your Apify plan:

| Apify plan tier | Price per 1,000 answers |
|---|---|
| FREE | USD 8.00 |
| BRONZE | USD 7.00 |
| SILVER | USD 6.00 |
| GOLD | USD 5.00 |

Free, never charged: summary rows, one per question and engine; every error row: no key, a rejected key, no credit left, timeouts and answers that cannot be read; engines you gave no key for; questions not reached after a stop or your spending limit. Apify also charges a small start event per run.

## Input

The fields you will most likely change:

| Field | What it does | Default |
|---|---|---|
| `brand, brandAliases` | Your brand as people write it, and up to 10 other names that count as your brand. | none |
| `domain` | Your website. Citations of it or its subdomains count as your site cited. | none |
| `competitors` | Up to 10 names. Add other names after a bar, such as Google Workspace | Google Docs. | none |
| `prompts` | Up to 50 questions, each at most 300 characters. | built-in questions for your category |
| `category, promptTemplates` | Used when the questions are empty: built-in or your own templates with {category}, {brand} and {country}. | none |
| `openaiApiKey, perplexityApiKey` | Your own keys, stored encrypted as secret input, never logged and never in the results. Give at least one. | none |
| `repeats` | Asks per question and engine, 1 to 10. | 3 |
| `monitorId` | The name of a tracked series. Summary rows then show the change against the previous run. | none |

A small input to start from:

```json
{
  "brand": "Acme CRM",
  "domain": "acme.example",
  "competitors": [
    "HubSpot",
    "Pipedrive",
    "Zoho CRM | Zoho"
  ],
  "category": "CRM tools",
  "monitorId": "acme-weekly",
  "openaiApiKey": "your-own-openai-key"
}
```

## Output

Three kinds of rows: answer rows (one per answer, charged), summary rows (one per question and engine, free) and error rows (free). The main fields:

- `engine, model, prompt, repeat`: Which engine and model answered which question, and which ask it was.
- `brandMentioned, position`: Whether your brand was named, and its place: 1 plus the number of your competitors named before it.
- `competitorsMentioned`: The competitors the answer named.
- `citedDomains, citedUrls, brandCited`: The sources the answer cites, and whether your own site is one of them.
- `answerExcerpt`: Up to 500 characters around the first mention of your brand, as evidence.
- `n, mentionRate, avgPosition`: In summary rows: the number of answers, the share that name your brand, and its average place.
- `shareOfVoice`: Answers naming your brand, divided by the answers naming any of the brands you track.
- `citationRate, topCitedDomains`: The share of answers that cite your site, and the 5 most cited domains with their counts.
- `change`: With a monitor ID: the change in each rate and in position since the previous run.

Only answer rows are charged, one per answer read. Summary rows and error rows are free. An error row names the reason, such as NO_API_KEY, ENGINE_AUTH_FAILED (key rejected) or QUOTA_EXCEEDED (no credit on your vendor account), and a rejected key stops that engine for the rest of the run. Download as JSON, CSV or Excel, or read them through the API.

## How to run it

### In Apify Console, no code

1. Type your brand, your website and 3 to 10 competitors.
2. Add 5 to 20 questions a buyer would ask, or leave the questions empty and type your category, such as CRM tools.
3. Paste your own OpenAI or Perplexity key, or both. Each engine runs only with its own key.
4. Click Start, then read the rows as JSON, CSV or Excel, or through the API. For a trend, set a monitor ID and schedule it.

Start from the tool's page (https://apify.com/tan-solutions/ai-visibility). You need an Apify account.

### From the API

One call runs the tool and returns the results. Replace `<YOUR_API_TOKEN>` with your token from Apify Console (Settings, API and Integrations). Send `{}` for a short demo run.

```bash
curl -X POST "https://api.apify.com/v2/acts/tan-solutions~ai-visibility/run-sync-get-dataset-items?token=<YOUR_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"brand":"Acme CRM","domain":"acme.example","competitors":["HubSpot","Pipedrive","Zoho CRM | Zoho"],"category":"CRM tools","monitorId":"acme-weekly","openaiApiKey":"your-own-openai-key"}'
```

### From an AI agent, through MCP

Apify's MCP server offers this tool alone to your MCP client at `https://mcp.apify.com?tools=tan-solutions/ai-visibility`. The client signs in to Apify on first connect, or sends your token in an `Authorization: Bearer` header.

```json
{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=tan-solutions/ai-visibility"
    }
  }
}
```

## Run it on a schedule

Answers change from week to week. Keep the same monitor ID and brand on every run, and each run's summary rows show the change against the run before. In Apify Console: Schedules, Create new, cron `0 7 * * 1` (every Monday at 07:00 UTC), then add this tool with your input. Or as one API call (`yNbRVmtZuP3vf7gSG` is this tool's Actor ID):

```bash
curl -X POST "https://api.apify.com/v2/schedules?token=<YOUR_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
  "name": "ai-visibility-schedule",
  "cronExpression": "0 7 * * 1",
  "timezone": "UTC",
  "isEnabled": true,
  "isExclusive": true,
  "actions": [
    {
      "type": "RUN_ACTOR",
      "actorId": "yNbRVmtZuP3vf7gSG",
      "runInput": {
        "body": "{\"brand\":\"Acme CRM\",\"domain\":\"acme.example\",\"competitors\":[\"HubSpot\",\"Pipedrive\",\"Zoho CRM | Zoho\"],\"category\":\"CRM tools\",\"monitorId\":\"acme-weekly\",\"openaiApiKey\":\"your-own-openai-key\"}",
        "contentType": "application/json"
      }
    }
  ]
}'
```

## Limits

- The answers come from the OpenAI and Perplexity APIs with web search, not from the ChatGPT or Perplexity apps. They are close, but can differ from what a logged-in user sees.
- You need your own key for each engine, and you pay that vendor for your usage on your own account.
- At most 50 questions, 10 repeats and 2 engines, so up to 1,000 answers per run. 100 answers take about 6 to 17 minutes, 1,000 answers about 1 to 3 hours.
- Matching is by name. A brand named after a common word is also counted when the word is used in its plain meaning; add a more specific name in such cases.

## Questions

### Why ask each question more than once?

The same question gets different answers. Three or more asks per question give a mention rate you can compare from week to week.

### Do I need my own key?

Yes, at least one: an OpenAI key, a Perplexity key or both. Each engine runs only with its own key. Without any key, the run ends at once with one free row that says what to add.

### Who sees my key?

Nobody at Tan Solutions. It stays in your Apify account, stored encrypted as secret input, used only inside your own run to call the vendor, and never logged or written to the results.

Not affiliated with, endorsed or approved by OpenAI or Perplexity. ChatGPT, OpenAI and Perplexity are marks of their owners. Built and run by Tan Solutions.

Source: https://www.tansolutions.org/tools/ai-visibility/ · Tan Solutions
