The terms heuristics and cognitive biases are often used loosely, with definitions that lack clarity. This paper refines these terms by presenting two definitions of heuristics: a narrow and a broad one. The narrow definition reflects Kahneman and Tversky’s view of heuristics as perception-like, natural assessments. From this perspective, people use various types of information as cues to make judgments about things not directly accessible but important to them. Some cues have high indicative value, i.e., the probability of achieving the desired result. Based on them, rules like “If A, then B” are formed, which become useful heuristics. Thus, heuristics are simple, intuitive if/then rules of judgment and decision making that, using such cues, generally enable the achievement of intended goals. Pólya, Simon, Gigerenzer, and others also describe a different type of heuristics – conscious methods that simplify decision making and problem solving. Taking them into account, it is possible to propose a broad definition of heuristics according to which they are simple methods of judgment, decision making, and problem solving that, using little information and cognitive resources, mostly enable the achievement of the intended goal. As for cognitive biases, they are systematic errors in cognition that occur when, having an epistemic goal, we non-consciously deviate from it by relying on irrelevant or partially relevant information and ignoring that which is relevant. They arise from the same cognitive processes as Kahneman and Tversky’s heuristics: just as we, without conscious awareness, use heuristic cues for judgment and decision making, we rely on irrelevant information when pursuing epistemic goals.