AI Search Research API for Structured Answer Data

Collect AI answers across engines, countries and locations as structured JSON, with the sources each answer cites, for academic research, content strategy and GEO studies.

Start free Read the docs ↗

8AI engines in one response format
9 × 9location grid around any point
500free credits to start

Who it's for

Academic researchers

Study how AI assistants answer the same question across markets, with the country and collection time stored with every answer.

Content strategists

Find the sources AI answers cite for your topics and the searches the engines run while they answer.

Data and analytics teams

Load answers, citations and search queries into your warehouse and join them with your search and sales data.

How it works

Start free Read the docs ↗

01
Design the sample
Pick the prompts, engines and countries. ChatGPT, Gemini and Google also take a latitude and longitude, and prompt research suggests questions ranked by search demand.
02
Collect in real browsers
Each task opens the engine's consumer app from the country you set and returns the answer a person there sees.
03
Analyze structured fields
Every answer arrives with its text, cited sources and engine fields such as search queries and shopping cards, ready for a notebook or a warehouse.
04
Repeat to measure change
Run the same sample again on a schedule. Each run is a fresh observation, so comparing runs shows how answers move over time.

Start from one request

Send it with your API key. The reference lists every field.

Read the docs ↗

POST/v1/async/task/grid

curl -X POST https://api.querying.ai/v1/async/task/grid \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "taskType": "CHATGPT",
    "payload": {
      "prompt": "dermatologist near me",
      "country": "US"
    },
    "grid": {
      "center": {
        "latitude": 40.758,
        "longitude": -73.9855
      },
      "size": 5,
      "radiusMeters": 2000
    }
  }'

Products it runs on

AnswersAI Answer Scraping API for GEO DataCollect consumer AI answers from ChatGPT, Gemini, Perplexity, Google, Meta AI, Qwen, Doubao and Naver through one async API, with source citations and market settings.SearchSearch Results, Shopping and Reddit Scraping APICollect Google, Bing and Naver search results, product searches and product pages from 21 storefronts such as Amazon, Coupang and Olive Young, and Reddit posts as JSON through one async API.Citation AnalysisCitation Analysis API: Diagnose Your Brand's AI VisibilityAudit one AI answer for your brand: whether it is named and at what rank, which pages the engine searched and cited, which brands those pages back, and prioritized GEO actions, all through one API.EnginesSupported AI EnginesCollect answers from ChatGPT, Gemini, Perplexity, Bing Copilot, Google AI Overviews and AI Mode, Meta AI, Qwen, Doubao, and Naver AI Brief and AI Tab through one async API. Pick an engine to see its response fields and a real response.

Frequently asked questions

How do I collect answers for a specific location?

Set payload.location to a latitude and longitude on ChatGPT, Gemini, Google AI Overviews, Google AI Mode or Google search results. POST /v1/async/task/grid asks one prompt from every point of a grid up to 9 × 9.

How should I sample answers that change between sessions?

Each task is a fresh session, so ask each prompt several times and read rates across the samples. Monitors take 1 to 10 samples per prompt in each run.

How do I move large result sets into a warehouse?

Result exports deliver every finished task as gzip-compressed JSON Lines in S3 or Google Cloud Storage, or as BigQuery table partitions, once per hour or once per UTC day.