Behavioral scoring is a risk-assessment method that judges a session by what it does, not only by where it comes from. Defenders assign a score from signals such as click timing, mouse movement, scrolling, navigation paths, request frequency, cookie reuse, and how closely a session matches normal human browsing patterns.
This matters because attackers can hide behind cloud hosts, residential IPs, proxies, or browser-like automation. A bot may look legitimate at the network layer but still behave differently from a real user. Behavioral scoring helps anti-bot systems spot scraping, credential stuffing, and account abuse when simple IP blocking or rate limits are not enough. In practice, sites combine behavioral scores with browser fingerprinting, session continuity checks, and challenge steps to decide whether to allow, slow, or block a request. The goal is not to prove every session is malicious, but to measure how suspicious its interaction pattern really is.