Researchers waste 80% of LLM annotation costs by classifying one text at a time
This provides a practical, cost-saving method for social scientists using LLMs for text classification, showing that batching and stacking are safe within certain limits.
Researchers found that batching 25 items and stacking all variables into a single prompt reduces LLM annotation costs by over 80% without degrading coding quality for batch sizes up to 100 and up to 10 stacked variables, with error smaller than typical inter-coder disagreement.
Large language models (LLMs) are increasingly being used for text classification across the social sciences, yet researchers overwhelmingly classify one text per variable per prompt. Coding 100,000 texts on four variables requires 400,000 API calls. Batching 25 items and stacking all variables into a single prompt reduces this to 4,000 calls, cutting token costs by over 80%. Whether this degrades coding quality is unknown. We tested eight production LLMs from four providers on 3,962 expert-coded tweets across four tasks, varying batch size from 1 to 1,000 items and stacking up to 25 coding dimensions per prompt. Six of eight models maintained accuracy within 2 pp of the single-item baseline through batch sizes of 100. Variable stacking with up to 10 dimensions produced results comparable to single-variable coding, with degradation driven by task complexity rather than prompt length. Within this safe operating range, the measurement error from batching and stacking is smaller than typical inter-coder disagreement in the ground-truth data.