nonaggregated

Nonaggregated is an adjective used in data management and analytics to describe data, measurements, or results that have not been summarized, rolled up, or combined across broader categories. Nonaggregated data retains the original granularity of observations, such as individual records, events, or timestamps, rather than totals, averages, or other summarized metrics. In contrast, aggregated data presents a higher-level view by applying functions like sum, average, or count to groups of records.

Common contexts for nonaggregated data include transaction logs, sensor readings, user events, and other raw data streams. Analysts may query nonaggregated data to perform ad hoc analyses, drill down into specifics, or train machine learning models. Aggregation is typically applied for dashboards and standard reports to improve readability and performance, but doing so can obscure detail.

Maintaining nonaggregated data has several advantages, including the preservation of detail, the ability to perform flexible analyses, and improved traceability for auditing or debugging. However, it also involves tradeoffs such as increased storage requirements, longer query execution times, higher complexity in data governance, and potential privacy considerations due to the finer-grained nature of the data.

In practice, nonaggregated data is often stored in raw or staging layers of data architectures or data lakes, where it can be retained alongside transformed, aggregated datasets. Related terms include granular data, raw data, and disaggregated data. The term’s exact usage can vary by domain, but it generally denotes data that has not been summarized.