Batch Processing
What is Batch Processing?
Batch is the older sibling of real-time processing. When you process records as they arrive, you spread the work over time and get instant results but pay for constant infrastructure. When you batch them, you accumulate the records and process them all at once — usually overnight or on a schedule — which is cheaper and simpler but adds latency. Modern GTM stacks use both: real-time for anything a prospect or rep is waiting on (a form submit, a meeting booking), batch for the heavy lifting that doesn't need to be instant (overnight enrichment, weekly score recalculation, monthly cohort reports). The right answer is rarely 100% one or the other.
Why it matters
- Far cheaper per record than real-time processing for non-urgent work.
- Simpler to operate — one scheduled job replaces a constant streaming pipeline.
- Easy to retry — a failed batch can be re-run; a failed real-time event is gone.
Use cases
- Overnight enrichment. All new accounts created in the day enriched in a single 2 a.m. job.
- Weekly score refresh. The entire lead database re-scored against the updated model.
- Monthly cohort report. Last month's cohort analyzed once for the board.
How turgo helps
turgo runs urgent operations in real time (signals, routing, first-touch) and uses batches for heavy operations like full-database re-scoring, with both observable from the same workflow UI.
See turgo in action →
