Batching: Why Can Waiting to Do Things Together Be Faster?

Batching means collecting several similar small jobs and executing them as a group. It does not make each job simpler; it reduces the setup, switching and finishing costs that would otherwise be repeated for every item.

A dishwasher offers a clear example. Starting it after every cup would repeat the filling, heating and draining cycle. Waiting for a load lets one setup serve many items, reducing the cost per item. Computing jobs, payments and email delivery can use the same structure.

Batching is related to buffering but is not identical. A buffer merely holds items awaiting work; batching determines how those items are grouped for execution. Nor is it debouncing: debouncing treats a stream of inputs as one change that has not yet finished, whereas batching preserves each item and processes them together.

Larger batches are not automatically better. A longer wait may improve equipment use and throughput, but it also increases the delay experienced by each item. The central question is therefore not simply whether work can be combined, but how much waiting is justified by the reduction in repeated overhead.