Many pages build their content with scripts after loading. A plain request gets an empty shell; a browser runs the scripts and gets the real page. Tools such as Playwright, Puppeteer and Selenium start a browser and tell it what to click and read.
A browser downloads everything a page asks for, so it uses far more proxy traffic than a plain request. Blocking images, fonts and video in the browser cuts the bill sharply. Each of these tools takes the proxy and its login in a slightly different place.
Headless does not mean invisible to the site, only that no window is drawn. Pages load, scripts run and cookies are kept exactly as in a normal browser, which is what makes it useful for pages a plain request cannot read.
Questions about headless browser
Does a headless browser need a proxy?
Not to run. It needs one when the site limits requests per address or shows different content by country, the same as any other scraper.
Can sites detect a headless browser?
Often, yes: automation leaves small traces in the browser. Stealth settings hide some of them. A clean IP address does not hide any.
Related terms
- Web scraping
- Web scraping is collecting information from web pages with a program instead of by hand, for example prices, listings or search results.
- Proxy traffic
- Proxy traffic is the data that passes through a proxy, counted in gigabytes. Both what you send and what you download count toward it.
- Browser fingerprint
- A browser fingerprint is the mix of details a browser reveals, such as screen size, fonts and graphics card, that together single it out.