DeepFetch documentation

DeepFetch is a web scraping platform with two ways to get data:

• Dashboard — point-and-click scraping. Pick a site template, enter a search query, and get structured results you can export or publish as a live URL.
• API — programmatic access for developers. Fetch any web page, extract structured data, take screenshots, or run AI-powered extraction from your own code.

Both methods share the same credit balance. You can see your remaining credits in the sidebar and on the API page.

1. Go to New Scrape, choose a site (e.g. Amazon UK or Bing Search), and enter what you want to find — like wireless headphones under £50.
2. Run the scrape. When it finishes, open the job to view results as a table, JSON, or CSV.
3. Copy a Live API Endpoint URL from the job page to use the data in spreadsheets, apps, or automations — without writing scraper code.

DeepFetch uses credits instead of per-request cash billing. Each action consumes a number of credits based on complexity. Failed dashboard scrapes and blocked API requests are refunded — you only pay for successful, non-empty results.

The New Scrape form is the fastest way to collect structured data:

• Search query — what you want to find (products, articles, videos, etc.). For a single product page, paste the URL in the custom URL field instead.
• Site— a pre-built template tuned for that website's layout. Choose Any site (AI) when no template exists.
• Max results — how many items to return (1–50). More results use more credits.
• Output format — table view in the dashboard; export as JSON or CSV from the job page.

Before you run a scrape, the form shows an estimated credit costand your current balance. If you don't have enough credits, the scrape won't start.

Completed scrapes appear on the Endpoints page. Click any row to see full results, re-run the scrape, or copy live URLs.

Templates use site-specific rules for speed and reliability. When a template doesn't return enough data, DeepFetch can fall back to AI extraction automatically.

Every completed scrape can be published as a live URL — a simple HTTP link that returns your data as JSON. Find these on any job detail page under Live API Endpoint.

• Cached data — returns the last successful scrape instantly. Ideal for dashboards and apps that poll periodically.
• Fresh data — triggers a new scrape on each request, then returns updated results. Rate limited to protect your credits.

Use the Refresh button on a job page to re-run a scrape manually without going through the API URL.

The REST API gives full control over page fetching, JavaScript rendering, proxies, screenshots, CSS/XPath extraction, and AI-powered queries. Manage keys on the API page.

Authenticate every request with one of:

• Authorization: Bearer YOUR_API_KEY
• ?api_key=YOUR_API_KEY

Common use cases:

• Fetch a page with JavaScript rendered — SPAs, React sites, lazy-loaded content
• Extract fields with CSS rules — product names, prices, links from any HTML page
• Ask a natural-language question — ai_query=price of the product
• Search Google or Bing — /api/v1/store/google?search=...
• Take full-page or element screenshots
• Check credit usage — GET /api/v1/usage

The API page includes copy-paste curl examples. Full parameter reference is available at /api/openapi for OpenAPI-compatible tools.

Get notified when a dashboard scrape finishes. Configure your webhook URL in Account settings. Works with Zapier Catch Hook, Make, and custom HTTPS endpoints.

• DeepFetch POSTs JSON when a job completes or fails. Payload includes endpointDataUrl, endpointFreshUrl, dashboardUrl, and a dataPreview array for no-code tools.
• Events: job.completed, job.failed, and job.test (from Send test webhook in settings).
• Optionally set a signing secret. Verify with X-DeepFetch-Signature (SHA-256 HMAC) or X-Hook-Secret for Zapier.

Zapier: Trigger → Webhooks by Zapier → Catch Hook → paste URL into settings → Send test webhook → Test trigger → add your actions (Sheets, Slack, Airtable, etc.).

The Analytics page shows how your account is performing over the last two weeks:

• Daily scrape volume and credit usage
• Success vs failure rate
• Average scrape duration
• Recent errors — useful for spotting blocked sites or bad queries

Use analytics to decide when to switch from a template to AI mode, enable premium proxies via the API, or adjust scrape frequency on live endpoints.

Manage your profile, webhook settings, and account deletion from Account settings.

• Update your display name
• View credit balance and plan tier
• Configure webhook URL and signing secret
• Permanently delete your account and all associated data

Sign in with Google. Your session stays active until you sign out from the sidebar.