Chapter 11 · Every screen, part by part
Batches
What research runs next, what it has already covered, and the button that starts it.
1 — How batches work. A short explanation in place, for when you arrive here without having read chapter 3.
2 — The three numbers.
| Number | What it means |
|---|---|
| This week | Batches run this week, against your weekly ceiling. 0 / 1 means none run and one allowed |
| Next batch covers | How many professors the next run would take on — your professors per batch setting |
| Batches run so far | The lifetime count |
The weekly ceiling exists so that a mistake in your settings costs you one batch rather than your month's budget. Both numbers come from Settings → Research Batches.
3 — Next batch preview. Exactly who the next run would cover, by name and university, before you start it. Read this list before pressing the button — it is the cheapest moment to notice that the queue is aimed at the wrong people.
Selection rotates between universities. Three professors from three different institutions, not three from one. That is deliberate: without rotation, one large imported roster would consume every batch for weeks and your list would grow lopsided without you noticing.
4 — Run this batch. Starts the run. What happens next is not in this app — the batch becomes work your connected assistant picks up, and this page shows its progress. If nothing happens at all, no assistant is connected; see chapter 10.
5 — Batch history. Every run, with what it covered and how it ended. Empty until the first one.
What a batch does, and where it stops#
A batch works through its professors one at a time, and each one goes: research → source check → your review → on the record. It stops on its own at two points and is stopped by the platform at two more — §10.6 has the full table.
The one that surprises people: findings that pass their source check still do not appear on a professor's page until you publish them. If a batch reports success and the records still look thin, that is where they are — Review.