PrismCrawl Google Search API

What does PrismCrawl Google Search API do?

PrismCrawl Google Search API runs live Google searches through the PrismCrawl API and saves the normalized JSON response to an Apify dataset: ranked results, SERP features (featured snippets, People also ask, knowledge panels, news, shopping, local results, AI summaries, and more when Google shows them), and pagination metadata.

This Actor is a thin integration between Apify and PrismCrawl. It does not scrape Google itself: every search is executed by PrismCrawl's Google Search endpoint, and the Actor passes your parameters through and stores what PrismCrawl returns. PrismCrawl searches are live and uncached: each successful request runs against Google, and PrismCrawl does not serve a cached keyword database.

Running it on Apify adds scheduling, API access, webhooks and integrations, run monitoring, and dataset exports on top of PrismCrawl.

• 🌐 PrismCrawl website: www.prismcrawl.com
• 📖 Full API reference: www.prismcrawl.com/docs

Why use PrismCrawl Google Search API?

• Rank tracking and SERP monitoring: schedule the same keyword with fixed location, language, and device, and compare positions over time.
• SEO and competitor research: see which pages rank and which SERP features appear for a keyword set.
• Local and international search: search as if from a country, a named city, or exact coordinates, on any supported Google domain, in any supported interface language.
• AI agents and research tools: give an agent current search results with titles, snippets, and source URLs. The Actor also supports Apify's Standby mode for low-latency HTTP calls.
• No infrastructure: PrismCrawl handles retrieval and parsing, so you don't run browsers, proxies, or parsers.

You need a PrismCrawl API key

This Actor uses your own PrismCrawl account. To get a key:

1. Create a free account at prismcrawl.com. New accounts get 100 free credits and don't need a credit card.
2. Copy your API key from the PrismCrawl dashboard.
3. Paste it into the PrismCrawl API key field of this Actor.

The key field is a secret input: Apify stores it encrypted, and the Actor never logs it or writes it to the dataset. It is only sent to api.prismcrawl.com in the x-api-key header.

How much does it cost to search Google with PrismCrawl?

There are two separate costs:

• PrismCrawl credits (billed by PrismCrawl to your PrismCrawl account). Each successful search costs one credit, including a completed search that returns no results. Each additional results page is a separate search. Failed searches are free: invalid requests, authentication errors, rate limits, and server errors don't use credits. PrismCrawl sells prepaid credit packages with no subscription. See PrismCrawl pricing for current packages and rates.
• Apify platform usage (billed by Apify) for the Actor's compute time. The Actor only makes HTTP calls, so a run is short and uses very little memory.

With Maximum result pages set to 1 (the default), one run uses at most one PrismCrawl credit.

How to use PrismCrawl Google Search API

1. Open the Actor's Input tab.
2. Paste your PrismCrawl API key.
3. Enter a Search query, for example best coffee shops in Baltimore.
4. Optionally set a country, language, location, device, search tab, or time filter.
5. Click Start. When the run finishes, open the Output tab to see the results, SERP features, or full PrismCrawl responses.

You can also start the Actor through the Apify API, schedule it, or connect it to webhooks and integrations.

Input

All fields other than the API key and query are optional. Parameter names in parentheses are the PrismCrawl API names, so you can look each one up in the PrismCrawl API reference.

Field	PrismCrawl parameter	What it does
PrismCrawl API key	x-api-key header	Your secret PrismCrawl key. Required.
Search query	query	The Google search. Required, up to 8,192 UTF-8 bytes.
Maximum result pages	–	Pages to fetch (1–10, default 1). The Actor stops early when there is no next page. Each page uses one credit.
Start offset	start	Zero-based result offset for the first page.
Country	gl	Two-letter result country (default us). Boosts results from that country.
Interface language	hl	Google interface language (default en-US).
Google domain	google_domain	For example google.co.uk (default google.com).
Location	location	A named place such as Baltimore,Maryland,United States (Google geo-target Canonical Name).
Latitude / Longitude	coordinates	Search from exact coordinates.
Radius	radius	Meters around the coordinates (1–199 for desktop, up to 1,000 for mobile or tablet).
Device	device	desktop, mobile, or tablet.
Search vertical	tbm	News, Videos, Images, Shopping, Local, or Books.
Search tab	udm	Local, Images, Forums, Videos, News, Web, Shopping, Books, or AI Mode.
Time and advanced filters	tbs	For example qdr:d (past day), qdr:w (past week), or cdr:1,cd_min:8/1/2026,cd_max:8/12/2026.
SafeSearch	safe	active or off.
Restrict to languages / countries	lr / cr	For example lang_fr|lang_de or countryUS|countryCA.
Encoded location	uule	A raw Google-encoded location, for advanced users.
Spelling correction, duplicate filtering, personalization	nfpr, filter, pws	Google's native 0/1 controls.
Zero trace	zero_trace	PrismCrawl won't store the source HTML or parsed JSON; only audit and billing metadata is kept.

Some fields can't be combined. Use only one of location, coordinates, or UULE. Radius requires coordinates. Use either a search vertical or a search tab, not both. The Actor checks these rules before sending anything, so an invalid combination fails immediately and costs nothing.

Example input

{
    "prismcrawlApiKey": "<YOUR_PRISMCRAWL_API_KEY>",
    "query": "best coffee shops in Baltimore",
    "gl": "us",
    "hl": "en-US",
    "location": "Baltimore,Maryland,United States",
    "device": "mobile",
    "maxPages": 2
}

Output

The Actor stores one dataset item per results page. Each item is the PrismCrawl response exactly as returned, plus two fields added by the Actor:

• page: the page number within this run (1, 2, …)
• credits_remaining: your remaining PrismCrawl credits, from the X-Credits-Remaining response header

The Output tab has three views of the same data:

• Search results: one row per ranked result (rank, type, title, URL, snippet, domain, source)
• SERP features: one row per feature (type, title, text, items, links)
• Pages: one row per PrismCrawl request with the complete response

You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

Example output item

Shortened; live values vary.

{
    "page": 1,
    "credits_remaining": 99,
    "success": true,
    "request_id": "66ca8deb-1dc2-4017-865f-c4fb0e3c7c0d",
    "data": {
        "format": "json",
        "content": {
            "search_parameters": {
                "q": "best espresso machines",
                "type": "search",
                "engine": "google",
                "device": null,
                "google_domain": "google.com",
                "gl": "us",
                "hl": "en-US",
                "location": null,
                "tbs": null,
                "tbm": null,
                "udm": null
            },
            "has_next_page": true,
            "results": [
                {
                    "id": "s_972c538a8734a8b7",
                    "rank": 1,
                    "type": "organic",
                    "title": "The 14 Best Espresso Machines, Tested & Reviewed",
                    "url": "https://www.seriouseats.com/best-espresso-machines-5185482",
                    "display_url": "seriouseats.com › best-espresso-machines-5185482",
                    "snippet": "Our favorite espresso machine is the Breville Bambino Plus.",
                    "domain": "seriouseats.com",
                    "favicon": "https://seriouseats.com/favicon.ico",
                    "source_name": "Serious Eats",
                    "position": { "absolute": 1 },
                    "engine": "google",
                    "domain_info": { "tld": "com", "sld": "seriouseats", "category": null },
                    "classification": null
                }
            ],
            "serp_features": [
                {
                    "id": "f_5236e4e73b2337d4",
                    "engine": "google",
                    "type": "people_also_ask",
                    "title": "People also ask",
                    "text": null,
                    "items": [
                        { "title": "Which espresso machine brand is most reliable?", "text": "Which espresso machine brand is most reliable?", "link": null }
                    ],
                    "links": [],
                    "source_result_ids": ["s_972c538a8734a8b7"],
                    "position": { "absolute": 1 },
                    "confidence": 0.8,
                    "extracted_at": "2026-07-01T16:45:24Z"
                }
            ]
        }
    }
}

Main data fields

Field	Description
request_id	PrismCrawl request ID. Use it to find the request in your PrismCrawl request history or when contacting support.
data.content.search_parameters	The parameters PrismCrawl used for the search.
data.content.results[]	Ranked results: rank, type (e.g. organic, ad, news, shopping), title, url, display_url, snippet, domain, source_name, position, domain_info.
data.content.serp_features[]	SERP features: type (e.g. featured_snippet, people_also_ask, knowledge_panel, ai_summary), title, text, items, links, source_result_ids, confidence.
data.content.has_next_page	Whether Google shows another results page.

The full, authoritative schema is in the PrismCrawl API reference.

Tips and advanced options

• Control cost with Maximum result pages. Each page is one credit. Start with 1 page and increase only if you need deeper results.
• Keep targeting fixed for rank tracking. Compare runs that use the same country, language, location, and device.
• Use a time filter for fresh results, for example qdr:d for the past day, or pick the News vertical (tbm) or tab (udm).
• Use Zero trace for sensitive queries. PrismCrawl then keeps only audit and billing metadata. Results still appear in your Apify dataset.
• Standby (HTTP) mode. For low-latency calls from your own code, use the Actor's Standby URL: POST / with the same JSON fields as the input (without prismcrawlApiKey and maxPages), and send your PrismCrawl key in the x-prismcrawl-api-key header. The response is the PrismCrawl response body and status code, unchanged. Each call fetches one page.
• Source HTML (html, rewrite_links), pre-encoded queries (query_encoded), and color_scheme are PrismCrawl parameters that this Actor doesn't expose, because they don't change the normalized results. Use the PrismCrawl API directly if you need them.

Errors and retries

The Actor reports PrismCrawl errors in the run's status message and log, including the PrismCrawl request ID:

HTTP status	Meaning	Retried?
400	Invalid request parameters	No
401	Missing, invalid, expired, or revoked API key	No
402	No PrismCrawl credits remaining	No
429	Rate limit exceeded	Yes, up to 3 times, waiting until the reported reset time (max 60 s)
5xx	PrismCrawl could not complete the request	Yes, up to 3 times with exponential backoff

Connection failures that never reached PrismCrawl (DNS errors, refused connections) are also retried. Retries only happen for failed requests, which PrismCrawl doesn't charge for, so a retry doesn't cause a duplicate charge. If a request times out, the Actor does not retry, because the search may have completed on PrismCrawl's side. Check your PrismCrawl request history before running again. If an error happens after some pages were saved, the run fails but the saved pages stay in the dataset.

FAQ, disclaimers, and support

Is this an independent Google scraper? No. All searches are performed by the PrismCrawl API under your PrismCrawl account and subject to PrismCrawl's terms. This Actor forwards your parameters and stores the response.

Are results cached? No. PrismCrawl runs a live search for every successful request.

Which SERP features will I get? It depends on the query, market, and device. Features appear only when Google shows them and PrismCrawl recognizes them.

Where can I learn more? The PrismCrawl API documentation covers every parameter and response field. PrismCrawl also offers other endpoints (Bing, DuckDuckGo, Google Maps, Google Shopping, Amazon, app stores, and reviews) at prismcrawl.com.

Support. For problems with this Actor, open an issue in the Issues tab. For account, credit, or API questions, contact PrismCrawl at support@prismcrawl.com and include the request ID.

Make sure your use of search data complies with applicable laws and the terms of the services involved.