Google Trends Scraper

Get daily trends, interest curves, related queries & country comparisons from Google Trends — without a browser, proxy, or API key.

HTTP-only (JSONP + official RSS) · no Playwright · no cookies persisted · rate-limit aware

In one line: Scrape 5 Google Trends datasets (daily trends, interest over time, related queries, related topics, multi-geo compare) → one flat, schema-backed JSON/CSV record per trend → for content teams, SEOs, and analysts who need current trend data in minutes.

Input	a mode + keywords and/or geo codes
Output	one clean record per trend; fields per mode (see Output)
Cost	~2 HTTP requests per keyword (no browser, ~256 MB); 20 records ≈ near-zero CU
Speed	daily trends (2 geos) ≈ seconds; 50 keywords interest ≈ ~2–3 min (rate-limit aware)
Runs on	Apify Console · CLI · API

Who it's for

• Content & editorial teams — know what's trending per market before you pick a headline.
• SEO / keyword researchers — related queries & topics give a ready seed keyword list.
• Regional growth analysts — multi-geo compare shows where demand concentrates.
• News & media monitors — daily RSS trends as an automatic morning data feed.

What you get

• Daily trending searches by country — top-10 per geo from Google's official trending RSS (source: rss), no cookie needed, near-instant.
• Interest-over-time curves — per-point value (0–100) with formattedTime, time (unix) and hasData for clean charting.
• Related queries (Top + Rising) — volume and momentum signals per keyword.
• Related topics — entity-level trends, not just strings.
• Multi-geo comparison in ONE run — pass a country list, get region-by-region interest values in a single pass (competitors often cap at one geo per run).
• Flat, sanitized records — control chars stripped, length-capped, ready for CSV / BI / AI pipelines with zero cleanup.

Why this Actor?

Feature	What you get
5 modes in one Actor	daily + IOT + related + topics + geo-compare, no second tool
HTTP-only (JSONP/RSS)	no Chromium, no proxy, no fingerprinting → lower CU + no proxy line item
Request budget guard	hard cap stops runaway cost & protects against IP bans
Per-keyword error records	one failing keyword doesn't kill the run — an error row is logged, the rest continue
Auto NID cookie bootstrap	session warms once in-memory (never persisted/printed) to pass Google's token checks
Backoff + re-warm	429s auto-retried 2→60s with jitter; missing-cookie 429 auto re-warms
Flat CSV-ready schema	mode-qualified single-record view via dataset schema

How it compares

Capability	This Actor	apify/google-trends-scraper	data_xplorer/google-trends-fast-scraper
Daily trends per geo	✅ (RSS)	❌	✅ (browser, 125+ countries)
Interest over time (batch keywords)	✅ (up to 50)	✅	⚠️ one keyword per run
Related queries + topics	✅ (HTTP, always on)	✅ (browser)	❌
Multi-geo compare in one run	✅	✅ (viewedFrom proxy)	❌
No browser / no proxy	✅	❌ (pageLoadTimeout, maxConcurrency)	❌ (browser + proxy rotation)
Per-request pricing anchor	~$0.001–0.002 (estimate)	$0.003/result	$0.002/result + $0.025 start

Quick start

Apify Console

1. Open the Actor and click Start.
2. Accept the default (mode: dailyTrends, geos: ["US","ID"]) or switch mode.
3. Results land in Storage → Dataset (JSON or CSV export).

{
  "mode": "interestOverTime",
  "keywords": ["coffee", "tea"],
  "geos": ["US"],
  "timeRange": "today 12-m"
}

How it works

Input JSON → validate (trust boundary) → warm NID cookie (in-memory)
  → [explore] GET /api/explore → widgets+token
  → [widget] GET /api/widgetdata/{multiline|comparedgeo|relatedsearches} → JSONP
  → [daily]   GET trending/rss?geo=XX  (official RSS)
  → strip `)]}'\n` envelope → parse → sanitize → batch pushData → Dataset

Input

{
  "mode": "dailyTrends",
  "geos": ["US", "ID"],
  "timeRange": "today 12-m",
  "maxKeywords": 50,
  "maxRequestCap": 400,
  "includeTopics": false
}

Field	Type	Required	Default	Description
mode	string	✅	dailyTrends	dailyTrends, interestOverTime, relatedQueries, relatedTopics, multiGeoTrends (see schema enum)
keywords	string[]	mode-dependent	—	1–50 keywords for interest/related modes (max 50/run to stay in IP budget)
geos	string[]	mode-dependent	["US","ID"] daily / ["US"] others	ISO 3166-1 alpha-2, region sub-codes, or WORLD; max 50
timeRange	string	—	today 12-m	now 4-h · now 1-d · today 3-m · today 12-m · today 5-y
category	integer	—	0	Google Trends category id; 0 = All (19 Entertainment · 21 Health · 27 Finance · 29 Food)
maxKeywords	integer	—	50	Hard cap on keywords processed this run (protects IP budget)
maxRequestCap	integer	—	400	Hard cap on total HTTP requests per run (budget guard)
includeTopics	boolean	—	false	relatedQueries mode: also fetch related topics

Output

One flat record per trend/mode in the dataset. Example (real dailyTrends run, US):

{
  "mode": "dailyTrends",
  "query": "US",
  "geo": "US",
  "date": "2026-10-21",
  "formattedDate": "2026-10-21",
  "trendTitle": "john husted",
  "trendUrl": "https://trends.google.com/trending/rss?geo=US",
  "approxTraffic": "200+",
  "source": "rss",
  "hasData": true
}

Field	Modes	Type	Description
mode	all	string	dataset name
query	all	string	originating keyword (or geo for dailyTrends)
geo	all	string	ISO code / WORLD
time, formattedTime	interestOverTime	number / string	unix sec + human date
value, formattedValue	interest / multi-geo / related	number / string	scalar value (0–100)
hasData	all	boolean	false on empty/error rows
rankKind, rank	relatedQueries/topics	string / number	TOP / RISING + position
entityTitle	relatedTopics	string	entity name
date, formattedDate	dailyTrends	string	trending date
trendTitle, trendUrl, approxTraffic	dailyTrends	string	+ official RSS link + approx traffic
geoCode, geoName, maxValueIndex	multiGeoTrends	string / number	region compare
link	related*	string	query/topic link
error	any	string	present only on per-keyword failure rows

Output schema: .actor/output_schema.json (dataset link). Dataset view: .actor/dataset_schema.json.

Cost, limits, proxy

Item	Value
Min memory	256 MB (max 512 MB set; actor.json)
Verified run (3 geo daily-trends, 30 records)	3.55 s, 0.00099 CU, 63.4 MB peak → ~$0.0002 platform cost
Expected run (50-keyword interest)	≈ 100 requests, ~2–3 min, rate-limit aware
Proxy	none required (datacenter IP OK for token flow; heavy runs may 429)
Retries	backoff 2→60s + 30% jitter; hard 429 fail-fast after 2 (per-keyword error row)
Cost per 1,000 results	≈ $0.0066 platform cost (verified: 30 records / 0.00099 CU)

Recipes

Daily trends for your markets (default)

{ "mode": "dailyTrends", "geos": ["US", "GB", "ID", "IN"], "maxRequestCap": 20 }

→ Morning news/content feed, top-10 per country.

Keyword momentum (Top + Rising)

{ "mode": "relatedQueries", "keywords": ["ai", "crypto", "saas"], "geos": ["US"], "includeTopics": true }

→ Seed lists for content/SEO in one run.

Multi-geo demand map

{ "mode": "multiGeoTrends", "keywords": ["laptop"], "geos": ["US", "DE", "ID", "JP", "BR"] }

→ Where to localize / prioritize ad spend.

Troubleshooting

Symptom	Cause	Fix
error: RATE_LIMITED (429 missing cookie) on a keyword	session cookie stale	auto re-warm triggers once; reduce maxKeywords/pause between runs
error: RATE_LIMITED (429 IP budget)	IP exhausted on Google's side	backoff 2–60s; wait IP reset, lower maxRequestCap
error: EMPTY_RESULT row	valid empty (OF-8) — no data that period	legit; e.g. relatedTopics returns empty on some sessions
Dataset has 0 items	geo blocked / wrong geo code	verify geo is ISO alpha-2; check warm() status in log
dailyTrends returns < 10 items	RSS top-10 is the public ceiling	by design — no category filter on RSS

FAQ

Q: Do I need a Google account or API key? No — public JSONP/RSS endpoints, NID cookie is auto-bootstrapped in memory.

Q: Why no Playwright/Puppeteer? HTTP-only = 10–20× cheaper CU + faster; all 5 modes work over JSONP/RSS.

Q: Is daily trends data "official"? It uses Google's public trending RSS feed — same data the trends page exposes; it's top-10 with no category/volume numbers.

Q: Why did one keyword return an error record? Rate limit or transient failure after 2 retries — the rest of the run continues; check the log + error value.

Q: Can I get historical data? Yes — interest/related/multi-geo support today 5-y or custom ranges.

Q: What does includeTopics do? In relatedQueries mode it also fetches related topics for the same keywords.

Limitations

• Public data only — you are responsible for Google's ToS and applicable privacy law.
• RSS daily trends: top-10, no category filter, no absolute search volumes (approx buckets only).
• Values are relative interest (0–100), not absolute search counts.
• Google may geo-block datacenter IPs; heavy runs can hit 429 → auto-retry + budget guard keeps it safe.

Run it

Open the Actor on Apify Console and click Start. Every run writes flat records to your Dataset — export JSON or CSV directly.