Bing Search Results Scraper

Returns: one row per Bing result with type, query, position, title, url, domain, snippet, sourceUrl. Input: {"queries": ["best running shoes"], "country": "US", "maxPagesPerQuery": 1, "maxItems": 10} Best for: SEO rank tracking, competitor research, web search for AI agents.

{"type": "organic", "query": "laptop kaufen", "position": 1, "title": "Laptops & Notebooks kaufen bei MediaMarkt", "url": "https://www.mediamarkt.de/de/category/laptops-notebooks-362.html", "emphasizedKeywords": ["Laptops", "Notebooks kaufen"], "sourceUrl": "https://www.bing.com/search?q=laptop%20kaufen&cc=DE&setlang=de"}

What does Bing Search Results Scraper do?

This Bing scraper returns structured Bing SERP data: one row per organic result, text ad or related search. Supply queries, a country and a language to use it as a Bing API for search workflows.

It is built for three common jobs:

• SEO rank tracking. Every organic record carries a rank (position), the page it appeared on, the decoded target URL and the domain, so you can see where a site ranks for a keyword in a given market and compare runs over time.
• Competitor ad monitoring. Text ads are returned with their headline, landing URL, displayed URL, advertiser and placement (top, bottom or other), so you can watch who bids on a keyword.
• Web search for AI agents and pipelines. One record is one result, which makes the output easy to feed to a language model, a spreadsheet or a database.

The Actor reads the public Bing results page the same way a visitor's browser would. It does not log in, and it never opens the result pages or clicks the ads. Only Bing search pages are requested, and ad landing URLs are decoded from the ad link rather than visited, so no advertiser is charged for a click.

Results depend on the market you choose. The Actor sends the country to Bing and also routes the request through a residential connection in that country, because Bing returns degraded or empty pages when the connection country and the market do not match.

What data can you get?

Fields that do not apply are generally null; breadcrumbs is emitted only for organic records.

Field	Type	Description
id	string	Stable 16-character id built from the type, market, query and URL (or suggestion text)
type	string	organic, ad or relatedSearch
query	string	Search query sent to Bing
country	string	Upper-case 2-letter country code used
language	string	Language code used
page	integer	Results page (1-based) where the item was first seen
position	integer	Organic: absolute rank across pages. Ad and related search: order of first appearance within the query
positionOnPage	integer	Organic only: rank within its page
title	string	Result headline, or the suggested query for a related search
url	string	Decoded target URL, ad landing URL, or the Bing search URL for a related search
displayedUrl	string	The breadcrumb-style URL line Bing shows under the title
breadcrumbs	array	Organic only: the path segments of the displayed URL, split on the separator Bing shows
faviconUrl	string	Organic only: icon URL shown in the result header, when present; never fetched
emphasizedKeywords	array	Organic only: distinct bold text spans in the title and snippet, preserving Bing wording
domain	string	Hostname of url without www.
snippet	string	Description text under the result
date	string	Organic only: the date Bing shows before the snippet, when present
siteName	string	Organic only: the site label Bing shows above the title
adUrl	string	Ads only: the raw Bing click link (never requested)
advertiser	string	Ads only: the advertiser domain shown on the ad
placement	string	Ads only: top, bottom or other
resultsTotal	integer	Bing's "About N results" estimate for the page, when shown
sourceUrl	string	The exact Bing results page URL the record came from
scrapedAt	string	ISO timestamp of collection

emphasizedKeywords contains distinct bold spans from the organic title and snippet, preserving capitalization. It reports Bing's wording, without inferring relevance or operator support. It is null when neither element has bold spans.

faviconUrl is the icon URL in the organic result header, when available. It is not a thumbnail. The Actor never fetches it; an application displaying it decides whether to load the image. These nullable fields add context without creating extra records or requests.

A few notes on how the fields behave:

Organic ranks are deduplicated by URL across pages, and results that Bing shows inside AI summary boxes are not counted as organic rank positions. Ads are deduplicated within a query, because the same ad often shows on several pages.

• The date field is only filled when Bing prints a date in front of the snippet, which is the case for a minority of results.
• Ads and related searches mostly appear from the second results page onward, so with maxPagesPerQuery set to 1 you will usually get organic results only.

How to use

1. Open the Actor and go to the Input tab.
2. Type one search query per line under Search queries, for example best running shoes. Alternatively supply Bing search URLs as strings or {"url":"…"} objects under Bing search URLs.
3. Enter a 2-letter Country code and a Language code, such as US and en.
4. Set Max pages per query and Max items to control how much is collected.
5. Choose which result types you want: organic results, ads and related searches.
6. Leave the proxy setting on its default. Residential proxies are required for complete pages and are applied automatically.
7. Click Start. Results appear in the dataset as they are collected, and you can export them as JSON, CSV, Excel or XML.

Queries are processed up to three at a time, and the pages within one query are fetched one after another with a short pause. A run with a single query and three pages typically takes under a minute.

Search operators

Search terms are passed to Bing exactly as you type them. Operators such as quotes, -term or site: are included in the request, but Bing does not always honor them for anonymous visitors, so do not rely on operator filtering for rank tracking.

Start URLs

If you already have Bing search URLs, put them in Bing search URLs. The Actor reads q (the query), cc (the country) and setlang (the language) from each URL. Those values override the Country and Language inputs for that URL. Other parameters, such as filters, are kept. URLs that are not Bing search URLs, or that have no q parameter, are skipped with a warning.

Nonsense and misspelled queries

Bing silently replaces queries it cannot match with a different query and shows results for that instead. With Skip substituted queries on (the default), the Actor checks whether any word of your query appears in the first-page results. If none does, it returns nothing for that query and reports why in the run status message and in the RUN_SUMMARY record. Turn the option off if you want whatever Bing returns.

Input

Field	Type	Default	Description
queries	array of strings	prefill best running shoes	Search terms, one per item. Also accepted under the names searchQueries, keywords and query
startUrls	array	none	Full Bing search URLs. Plain strings and { "url": ... } objects both work
country	string	US	2-letter market code, also used as the proxy country
language	string	en	Language code sent to Bing (for example en, de, fr, zh-hans)
maxPagesPerQuery	integer	3	Results pages per query, 1 to 20
maxItems	integer	100	Maximum records across all queries and types, 1 to 10000
timeRange	string	any	Freshness filter for search terms: any, day, week or month
includeOrganic	boolean	true	Return organic results
includeAds	boolean	true	Return text ads
includeRelatedSearches	boolean	true	Return related searches
skipUnrelatedResults	boolean	true	Drop queries that Bing replaced with a different query
proxyConfiguration	object	Apify residential	Proxy settings. Residential is enforced

Example input:

{
  "queries": ["project management software", "crm software"],
  "country": "US",
  "language": "en",
  "maxPagesPerQuery": 2,
  "maxItems": 80,
  "includeAds": true,
  "includeRelatedSearches": true,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

The maxItems limit is exact: when it is reached, the Actor stops requesting pages and stops adding records. Within each page, records are written in this order: organic results first, then ads, then related searches. A small maxItems therefore returns organic results from the first page.

Output

This organic record came from a local run of laptop kaufen in the German market. The null icon reflects that response variant; other Bing layouts provide the icon URL.

{
  "id": "7ae1878cf2283980",
  "type": "organic",
  "query": "laptop kaufen",
  "country": "DE",
  "language": "de",
  "page": 1,
  "position": 1,
  "positionOnPage": 1,
  "title": "Laptops & Notebooks kaufen bei MediaMarkt",
  "url": "https://www.mediamarkt.de/de/category/laptops-notebooks-362.html",
  "displayedUrl": "https://www.mediamarkt.de › de › category",
  "breadcrumbs": [
    "www.mediamarkt.de",
    "de",
    "category"
  ],
  "faviconUrl": null,
  "emphasizedKeywords": [
    "Laptops",
    "Notebooks kaufen"
  ],
  "domain": "mediamarkt.de",
  "snippet": "Laptops & Notebooks für alle Bedürfnisse Homeoffice Bildbearbeitung und vieles mehr - Jetzt bei MediaMarkt online kaufen!",
  "date": null,
  "siteName": "mediamarkt.de",
  "adUrl": null,
  "advertiser": null,
  "placement": null,
  "resultsTotal": 46000,
  "sourceUrl": "https://www.bing.com/search?q=laptop%20kaufen&cc=DE&setlang=de",
  "scrapedAt": "2026-10-12T23:51:46.947Z"
}

An ad record has the same fields, with type set to ad, the decoded landing page in url, the raw Bing click link in adUrl, and advertiser and placement filled in when Bing shows them.

Besides the dataset, each run stores a RUN_SUMMARY record in the default key-value store. For every query it lists the pages fetched, the number of organic results, ads and related searches, the number of retries and the reason a query returned nothing, if any.

How much does it cost?

Every dataset record is one charged result. maxItems caps the records returned. See the Pricing tab for current billing details. Runs stop successfully with a clear message when the maximum cost per run cannot cover another result.

To keep a run small, lower maxItems and maxPagesPerQuery, and turn off the result types you do not need.

Use with AI agents (MCP)

You can call this Actor from any MCP-compatible client through the Apify MCP server. Add the Actor as a tool with a configuration like this:

{
  "mcpServers": {
    "apify": {
      "url": "https://mcp.apify.com?tools=axiomworks/bing-search-scraper"
    }
  }
}

Example prompts for your agent:

• "Search Bing for 'best running shoes' in the US and list the top 10 organic results with their domains."
• "Use Bing to check which advertisers show text ads for 'crm software' in the UK."
• "Get the related searches Bing suggests for 'home insurance' and group them by topic."

Each record is one result, so an agent can call the Actor with {"queries": ["your query"], "maxPagesPerQuery": 1} and read the dataset items directly.

Integrations and API

You can run the Actor from code with the Apify API clients, or connect it to other tools through Apify integrations and webhooks. Replace YOUR_API_TOKEN with your own token (in cURL, set it in the APIFY_TOKEN environment variable and send it in the Authorization header).

Python:

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")
run = client.actor("axiomworks/bing-search-scraper").call(run_input={
    "queries": ["best running shoes"],
    "country": "US",
    "language": "en",
    "maxPagesPerQuery": 2,
    "maxItems": 30,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["type"], item["position"], item["title"], item["url"])

JavaScript:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_API_TOKEN' });
const run = await client.actor('axiomworks/bing-search-scraper').call({
    queries: ['best running shoes'],
    country: 'US',
    language: 'en',
    maxPagesPerQuery: 2,
    maxItems: 30,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
for (const item of items) process.stdout.write(`${item.position} ${item.title}\n`);

cURL:

curl -X POST "https://api.apify.com/v2/acts/axiomworks~bing-search-scraper/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"queries": ["best running shoes"], "maxPagesPerQuery": 1, "maxItems": 10}'

FAQ

Why do I get fewer results than I asked for?

Bing shows between about 6 and 14 organic results per page, and the number of pages it serves is limited. If a query has fewer results than maxItems, or the page limit is reached first, the run ends with fewer records. Raise maxPagesPerQuery to get more.

Why are there no ads or related searches on page 1?

A fresh visit to Bing usually shows a first page without ads and without related searches. They tend to appear from page 2. Set maxPagesPerQuery to at least 2 or 3 if you need them. Some queries simply have no ads in some countries.

Why did a query return nothing?

Check the run status and RUN_SUMMARY. Try broader terms, check country/language and residential proxy access, or set skipUnrelatedResults=false to accept substitutions. Enable includeOrganic; ads and suggestions may need more pages.

Why do my results differ from what I see in my browser?

Bing personalizes results by location, including the city of the connection, and by session. Results can also change between runs. Expect some variance, especially for local queries.

Does the Actor support search operators?

Operators are sent as typed, but Bing may ignore them for anonymous visitors. Do not depend on them.

Which countries and languages are supported?

Use a 2-letter country code and a language code such as en or en-US.

What happens when Bing blocks a request?

The Actor retries blocked or failed requests up to three times with a pause between attempts, each time on a fresh proxy session. A target returning no records has a 75-second request budget. Unresolved targets are reported in RUN_SUMMARY; the run continues successfully.

Is it legal to scrape Bing?

The Actor reads publicly visible search result pages and does not log in or access any private data. Whether scraping is appropriate for your use depends on your jurisdiction, your purpose and the terms of the website. Search results can contain personal data, so handle any personal information in line with the privacy rules that apply to you, such as GDPR. Microsoft's terms of service restrict automated querying of Bing, so check them and take legal advice if you are unsure. This Actor is an independent tool and is not affiliated with or endorsed by Microsoft or Bing.

Feedback

If a field is missing, a market behaves unexpectedly or Bing changes its page layout and results look wrong, open an issue from the Actor page and include the query, the country and the run link.