Run a job group
Run a group from Home, review the classified buckets, and export with Process.
You've built a job group — now point a mixed batch at it. Pick a source, let Scanix Desktop sort and read every document in one pass, glance over the classified buckets, and click Process to file everything where it belongs. This page walks the run end to end.
Before you start
You need a job group that has something to run with — at least one member template, an AI service, or both. A group with neither can't act on a document; if you open one in this state, Scanix tells you Nothing configured yet and points you back to the designer to add members or AI services. See Build a job group.
Run a group
Open the group and pick a source
From the Home page (or the Job Groups list), click the group you want to run. A source chooser opens with three ways to feed documents in:
- Scan from Scanner — acquire straight from your connected scanner. This option is greyed out and reads No scanner detected when no device is found.
- Import Files — select PDF, TIFF, JPEG, or PNG files to process.
- Import Folder — process all files in a directory.
Screenshot
The job group source chooser opened from Home, showing the three source tiles — Scan from Scanner, Import Files, and Import Folder — under the group's name. — shot job-groups-run-a-job-group-01
Let the run go end to end
Pick a source and Scanix takes it from there — there's no separate "start" button to press. As soon as the batch lands it runs the whole group automatically: it splits the batch into documents, routes each one to the right member template, reads it with OCR, and — when the group has AI services — runs AI extraction on top.
The activity chip narrates the run as it goes. For a group with AI services it frames the work as AI extraction rather than a plain OCR batch, and steps through each document so you can watch progress.
Bring your own AI
AI extraction runs on the engine you've configured. Cloud calls go straight from your machine to your own provider account (Anthropic or OpenAI) — Scanix is never the account holder and never sees your key or your documents. For a group set to local-only privacy, any cloud call is hard-blocked regardless of how routing is configured.
Review the classified buckets
When the run finishes, the documents are classified and staged for review — not yet written to disk. In the thumbnail strip they're grouped into collapsible buckets, one section per document type the group recognised. Click a bucket header to expand it and scope the canvas to just those documents.
Documents the group couldn't place land in their own visible section so they never disappear silently: an Unclassified bucket for AI-driven runs, or an Unrecognized bucket for routed groups. A bucket that you expected to fill but stayed at zero is a useful signal too — it usually means something was classified somewhere it shouldn't have been.
This is your chance to spot-check before anything is filed. Open a document, confirm its type and extracted fields look right, and move on.
Screenshot
A finished job group run with the thumbnail strip grouped into buckets by type, an Unclassified section at the end, and the Process button ready in the toolbar. — shot job-groups-run-a-job-group-02
Click Process to export
When the buckets look right, click Process. For a job group run, Process is the save step — it writes every staged document to disk according to the group's export settings. If any document is still missing a required field, the run is held back and you're told which pages need attention before it will export.
When it's done, Scanix reports how many files were written and to how many locations.
Where your output lands
How documents are filed depends on the group's export setup:
- With a group-level export config (the designer's Export tab, with a Destination set) — every document the group produces is filed under that destination. With Per-type subfolders turned on, each type goes into its own subfolder — for example
…/Invoices,…/Receipts, and…/Unrecognized— so the output is sorted on disk exactly as it was in the buckets. - Without one — each document is exported using its matched member template's own export settings (the same folder outputs you set on the template's Naming & Filing), so invoices land where that invoice template files them and receipts land where their template files them.
AI-only groups need a destination
A group that has no member templates (AI services only) has no template export settings to fall back on, so it requires a Destination on the Export tab. The designer won't let you save an AI-only group until you set one.
Troubleshooting
Next steps
Build a job group
Add member templates, set their order, pick AI services, and choose where output lands.
Export overview
Formats, destinations, filename patterns, and how documents are filed.
Hot folders
Run a job group automatically as files arrive — no clicks per batch.