How to Scrape Bing Search Results as Structured JSON

I watch rankings for a few sites, and Bing kept being the blind spot. Google has a dozen rank trackers; Bing has almost none, and after Microsoft retired its public Search API endpoint, pulling Bing results in code meant writing a scraper and babysitting a browser. I wanted a clean call that hands me the results page as data, so I built one.

What it is

The Bing Search API is a Bing scraper that returns organic search results as structured JSON. Give it a query and it returns each result with its position, page, title, snippet, the real destination URL, the displayed URL, and a date when Bing shows one.

What it returns

• Organic results: position, page, title, snippet, url, displayedUrl, date

• Optional ad rows, so you can see who is bidding above you

• Market, location, safe-search, and device controls

• One row per result, CSV export built in

Run it in three steps

Set a query, pick a market, and choose how many pages to pull. Run it. Read the JSON, one row per result. It is billed per page of results, about $0.015 a page, because Bing does not let callers set a page size and organic depth swings from 2 to 10 per query. The page is the honest unit, so a thin page never costs more than it is worth.

Use it as a REST API

Every Actor on Apify has a REST endpoint. Start a run with your query, poll for the finish, then read the dataset as JSON or CSV. No key from Microsoft, no proxy pool to rent, no HTML parsing on your side. The Bing Search API page lists every input.

Wire it into Claude with MCP

The Actor runs over MCP, so Claude, Cursor, or any MCP client can call it as a tool. Ask where your site ranks on Bing for a keyword and the agent runs the search and reads the position field back, with no scraping code. This is the cheapest way to give an AI assistant live second-engine visibility.

What people build with it

Rank tracking dashboards, keyword position checks, SERP research datasets, and competitor ad monitors. A few runnable examples: Bing Rank Tracking, Scrape Bing Search Results, and Bing SERP API. The full set lives on the project page.

FAQ About Scraping Bing

How fresh and complete is the data this Bing scraper returns?

Results come back live at run time, straight from a real Bing results page for the market you pick. Organic depth varies by query, usually 2 to 10 results per page, which is why billing is per page. It returns what Bing actually shows; it does not invent positions or backfill missing ranks.

How do I call the Bing Search API from my own code?

Start a run through the Apify REST API with your query, wait for it to finish, and read the dataset as JSON or CSV. Pricing is about $0.015 per page returned, with no subscription and no Microsoft API key. Ten pages of results costs around fifteen cents.

Can I use it from Claude or another AI agent over MCP?

Yes. The Actor is exposed over MCP, so any MCP client can discover and call it as a tool. Point Claude at it and ask a ranking question in plain English; it runs the search and reads back the structured rows, no code required.

Can an AI agent pay for this scraper in USDC with x402?

Yes. The Bing Search API supports agentic payments via the x402 protocol, so AI agents and MCP clients can pay for runs in USDC on Base with no Apify account or API token. Point your agent at the Apify MCP server and it can discover, pay for, and run the scraper autonomously; the Apify x402 announcement has the details.

Issues get a same-day reply. If Bing changes a field, the Actor gets patched.