Traffic Generator

Generate realistic website traffic with human-like behavior simulation. Drive pageviews to any website, YouTube video, Etsy shop, or Behance profile. Each visit includes natural scrolling, mouse movements, variable dwell times, and browser fingerprint randomization to appear as genuine user traffic.

What can this traffic generator do?

• Simulate real traffic -- Visit any URL with human-like browsing behavior including scrolling, mouse movements, and idle time
• YouTube video views -- Play YouTube videos with realistic viewing behavior and configurable watch duration
• Crawl and browse -- Follow links on target pages to simulate deeper browsing sessions across multiple pages
• Fingerprint randomization -- Each session uses a different browser fingerprint (user agent, viewport, timezone, WebGL parameters)
• Stress testing -- Rapid request mode for load testing your own websites
• Resource blocking -- Block ads and trackers for faster page loads and reduced bandwidth
• Kill switch -- Automatically stop after a set time limit to control costs
• Detailed logging -- Track every page visit with load times, scroll depth, and interaction metrics

Use cases

• SEO and analytics -- Increase website traffic metrics for testing and validation
• Load testing -- Test how your website handles traffic using stress mode
• YouTube creators -- Generate views on YouTube videos
• Etsy shop owners -- Drive traffic to your Etsy shop listings
• Behance designers -- Increase portfolio views
• Campaign testing -- Verify analytics tracking is working before launching campaigns
• Website monitoring -- Regularly visit pages to check availability and performance

Input

Field	Type	Required	Default	Description
startUrls	URL[]	Yes	--	Target URLs to visit
mode	string	No	PAGEVIEW	PAGEVIEW (human simulation) or STRESS (rapid requests)
enableCrawling	boolean	No	false	Follow links on visited pages for deeper sessions
crawlingLinkSelector	string	No	a[href]	CSS selector for links to follow when crawling
maxPagesPerUrl	integer	No	10	Max pages per start URL when crawling (1-1,000)
waitOnPage	integer	No	30	Seconds to spend on each page (0-600, varies +/-20%)
endAfterSeconds	integer	No	300	Stop after N seconds (0 = unlimited)
enableYoutube	boolean	No	false	Enable YouTube video playback simulation
blockResources	string[]	No	--	Additional URL patterns to block
enableAdvancedFingerprinting	boolean	No	true	Randomize browser fingerprint per session
proxyConfiguration	object	No	--	Optional proxy settings for geographic targeting

Example input

{
    "startUrls": [
        { "url": "https://example.com" },
        { "url": "https://www.youtube.com/watch?v=dQw4w9WgXcQ" }
    ],
    "mode": "PAGEVIEW",
    "waitOnPage": 30,
    "endAfterSeconds": 300,
    "enableCrawling": true,
    "maxPagesPerUrl": 5,
    "enableYoutube": true
}

{
    "startUrls": [
        { "url": "https://www.etsy.com/shop/YourShopName" }
    ],
    "waitOnPage": 45,
    "endAfterSeconds": 600,
    "enableCrawling": true,
    "maxPagesPerUrl": 10
}

Output

Each page visit produces a record in the dataset with these fields:

Field	Type	Description
url	string	URL of the visited page
status	string	Visit result: success, failed, blocked, or timeout
pageTitle	string	HTML title of the page
loadTimeMs	number	Page load time in milliseconds
timeOnPageSec	number	Seconds spent on the page
scrollDepthPercent	number	How far the page was scrolled (0-100)
linksFound	number	Links found on the page
linksFollowed	number	Links followed from this page
referrer	string	Previous page URL (empty for start URLs)
userAgent	string	User agent used for this visit
viewport	string	Viewport dimensions (e.g., 1920x1080)
timestamp	string	ISO 8601 timestamp of the visit

Sample output

{
    "url": "https://example.com",
    "status": "success",
    "pageTitle": "Example Domain",
    "loadTimeMs": 342,
    "timeOnPageSec": 28.5,
    "scrollDepthPercent": 85,
    "linksFound": 1,
    "linksFollowed": 0,
    "referrer": "",
    "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
    "viewport": "1920x1080",
    "timestamp": "2026-03-23T10:30:00.000000+00:00"
}

How it works

1. Launches a real Chromium browser with stealth technology to avoid bot detection
2. Creates a unique browser fingerprint for each start URL (user agent, viewport, timezone, WebGL)
3. Visits each target URL and simulates human behavior: scrolling, mouse movements, link hovering
4. Spends the configured amount of time on each page with natural variation
5. Optionally follows links to simulate deeper browsing sessions
6. Records detailed metrics for every page visited
7. Stops automatically when the kill switch timer expires

Tips for best results

• Start with a small number of URLs and short endAfterSeconds to test
• Use PAGEVIEW mode for realistic traffic; STRESS mode for load testing only
• Enable crawling to increase total page visits per start URL
• Adjust waitOnPage based on your needs: 10-30s for quick traffic, 60-120s for deep engagement
• Use proxy for websites that block datacenter IPs or for traffic from specific locations
• YouTube mode works best with waitOnPage set to the approximate video duration
• Block additional resources to speed up page loads when images and scripts are not needed

Limitations

• The actor uses a real browser, so it consumes more compute than HTTP-only tools
• Some websites with advanced bot detection may still block visits
• YouTube view counting has anti-fraud systems; views may not always register in YouTube analytics
• Stress mode may trigger rate limiting on target websites
• The actor cannot solve CAPTCHAs

Frequently Asked Questions

Do I need a proxy? No. The actor works without proxy for most websites. Use proxy when target websites block datacenter IPs or when you need traffic from specific geographic locations.

How many pages can it visit? In PAGEVIEW mode with 30-second dwell time, approximately 2 pages per minute per start URL. With crawling enabled and 10 maxPagesPerUrl, one start URL generates up to 10 page visits. In STRESS mode, it can process 50+ pages per minute.

Does it work with YouTube? Yes. Enable the YouTube mode and set waitOnPage to the desired viewing duration. The actor plays the video and simulates watching behavior.

Can I use it for Etsy? Yes. Provide your Etsy shop or listing URL as a start URL. Enable crawling to visit multiple product pages.

What is the kill switch? The endAfterSeconds parameter automatically stops the actor after the specified time. This prevents runaway costs. Set to 0 for unlimited runtime (the actor will stop after processing all URLs).

Does it store output data? Yes. Every page visit is recorded in the dataset with detailed metrics including load time, scroll depth, and interaction data.

How is the human behavior simulated? Each visit includes random mouse movements, variable-speed scrolling to random depths, link hovering, and idle periods that simulate reading. Timing varies randomly so no two visits look identical.

Can it follow links on the target website? Yes. Enable crawling mode to follow links on visited pages. You can specify a CSS selector to target specific links (e.g., a.product-link). Only same-domain links are followed.

What is the difference between PAGEVIEW and STRESS mode? PAGEVIEW mode simulates realistic human browsing with scrolling, mouse movements, and dwell time. STRESS mode sends rapid requests with minimal delay, useful for load testing your own websites.

Can I run multiple instances? Yes. You can start multiple runs of this actor simultaneously from the Apify platform to multiply traffic volume.