Outlier

Outlier is a noun drawn from early 20th-century statistics to describe a data point that sits an abnormal distance from the rest of the values in a dataset. The term captures the idea of something that “lies out” beyond the main cluster of observations. Synonyms include anomaly, aberration, deviation, and exception, while norm, average, and typical value describe the opposite condition.

In hospitality analytics, outliers appear in many forms. If ten comparable vacation rentals on a lake rent for between $200 and $300 per night but one luxury estate commands $4,000 per night, that single data point is an outlier. Left in the dataset, it pulls the average price well above what any typical property in that market actually earns, producing a benchmark that misleads rather than informs. This is precisely why median price is often more reliable than mean price in markets with significant property variation: the median is resistant to outliers in a way the arithmetic mean is not.

Revenue managers and data analysts use outlier detection to serve two distinct purposes. The first is data cleaning, removing booking errors, duplicate entries, or anomalous transactions that would distort forecasts if left in place. The second is investigation, because not every outlier is a mistake. A month where revenue spiked far above historical norms might reflect a one-time event worth understanding rather than a number to delete. Outliers classified as point outliers involve a single anomalous value, contextual outliers are anomalous only within a specific situation, and collective outliers describe a group of values that are unusual together even if no single point stands out alone.

Related terms include mean, median, standard deviation, interquartile range, and data scrubbing.