Understanding Bot Management
Effective bot management blends distributed systems, web development, machine learning, security, research, and every discipline in between, while fighting ever adaptive and motivated adversaries.
Key concepts
- Bot - An autonomous program on a network that can interact with computer systems or users, imitating or replacing a human user's behavior, performing repetitive tasks much faster than human users could.
- Good bots - Bots which are useful to businesses that they interact with, such as search engine bots like Googlebot or Bingbot, or bots that operate on social media platforms like Facebook Bot.
- Bad bots - Bots which are designed to perform malicious actions, ultimately hurting businesses, such as credential stuffing bots, third-party scraping bots, spambots, and sneaker bots.
- Bot management - Blocking undesired or malicious Internet bot traffic while still allowing useful bots to access web properties by detecting bot activity, discerning between desirable and undesirable bot behavior,
and identifying the sources of the unwanted activity. - Web Application Firewall (WAF) - A security system that monitors and controls network traffic based on a set of security rules.
The following image shows the difference between good and bad bots:
The module
All bot detection mechanisms are applied on every request in real time during the request processing stage in the bot management module that runs on every machine at Cloudflare's edge locations. When a request comes in, we extract and transform the required request attributes and feed them to our detection mechanisms.
Detection Mechanisms
The Cloudflare Bot Management platform currently uses five complementary detection mechanisms, producing their own scores, which we combine to form the single score going to the firewall. Most of the detection mechanisms are
applied on every request, while some are enabled on a per-customer basis to better fit their needs.
Overall globally, more than a third of the Internet traffic visible to Cloudflare is coming from bad bots, while Bot Management customers have the ratio of bad bots even higher at approximately 43%.
Summary
Cloudflare has the unique ability to collect data from trillions of requests flowing through its network every week. With this data, Cloudflare can identify likely bot activity with machine learning, heuristics, behavioral analysis, and other detection mechanisms. Cloudflare Bot Management integrates seamlessly with other Cloudflare products, such as WAF and Workers.
Updated about 1 year ago