In general, every time our proxy sends a request to a website (asking for an url, .js, .css, image files, etc), our checkers count it as one request. If you are using, e.g., scrapebox, it usually downloads only an html page, not a js, css etc. 

Requests are calculated like this: if you've connected to a proxy and authorized yourself - that means a request. We have no way to determine if the request you send was actually successful, because most sites use https now and everything is encrypted. Additionally, request counts are independent of IP swaps or the number of pages visited, as each page may contain a different number of resources. Proxies may encounter limitations based on the frequency and volume of requests. For instance, YouTube may restrict searches if too many requests are made in a short period. To avoid such issues, it is recommended to reduce the search frequency and monitor the proxy's performance.

Residential proxy gateways connect users to a pool of residential IPs. When you assign the gateway to different browsers, applications, or ports, each connection is assigned an IP from this pool. This mechanism ensures that IPs are rotated across connections, providing anonymity and reducing the likelihood of being blocked or flagged by websites. This setup is particularly useful for managing multiple connections simultaneously, as each connection can operate with a unique IP address.

It is important to note that request counts are independent of IP swaps, so managing the total number of requests is crucial for efficient proxy usage.