Documentation / QueryTek Review glossary / Batch Submission
A batch submission sends several related documents for review as a single request instead of one at a time. The batch keeps them together for tracking and decision-making, while each document is still counted, reviewed, and reported on individually, so grouping changes the workflow around the documents rather than the review of any one of them.
Documents that need review rarely arrive alone. A course has modules, a policy set has chapters, an evaluation run produces hundreds of AI answers. Submitting each separately multiplies clicks and scatters related work across a list. A batch keeps related material together.
What a batch changes and what it does not is the whole idea.
What changes. The submitter makes one request instead of many. The documents share a request context, so the customer can track progress and review outcomes as a set. For integrations, one call carries the whole group, which simplifies retry handling and makes it easier to reconcile what was sent with what came back.
What does not change. Each document is still measured on its own, so the price reflects the actual words in every file. Each is still assigned, reviewed against the rubric, and approved on its own merits, and each receives its own findings. A weak document is not hidden by strong neighbors.
The common misunderstanding is that a batch is reviewed as one blended document, with a single score for the lot. That would save effort and destroy usefulness: authors need to know which chapter failed, not that the set averaged out.
A second misunderstanding runs the other way — that batching should buy a volume discount by default. Pricing follows the words reviewed; commercial terms for large programs are a separate conversation.
Related documents share context: terminology, audience, the standard they are meant to meet. Submitting them together keeps that context visible to the people planning the review and to the customer reading the results, and it turns a pile of individual outcomes into a view of the whole set.
QueryTek Review accepts batches through the portal and through the integrator API, which also supports idempotency keys so a retried batch request is not submitted twice. Each document in a batch is quoted from its own word count, reviewed individually, and delivered with its own findings. Batch size limits are not published.
