Cloaking is showing search engine crawlers a different version of a page than what real visitors see, most classically a keyword-stuffed, stripped-down page built purely to rank, while human visitors land on the actual, often unrelated content. It's one of Google's most explicitly and severely penalized black hat tactics, since it directly deceives the systems Google uses to evaluate what a page actually contains.
The tactic has largely faded because modern crawlers render pages far more like real browsers than the simple text-parsers they used to be, which makes a straightforward content mismatch between what a bot sees and what a visitor sees much easier to detect than it once was. It's worth distinguishing from legitimate, permitted techniques that can look superficially similar: dynamic rendering, serving a pre-rendered version of a JavaScript-heavy page to crawlers, is explicitly sanctioned by Google as long as both versions show the same underlying content. The line is intent and content parity, not the mere fact that two different versions of a page exist for different audiences; cloaking specifically means those versions genuinely differ in substance, not just in rendering method.
