Per-page field capture
How Scanix captures one value per page across multi-page documents and carries it into every export.
Treat every page as its own form. When a single document runs many pages — a stack of delivery notes, a multi-page batch where each sheet repeats the same layout — Scanix Desktop captures one value per page for each field, tags it with the page it came from, and carries that page number all the way through to your export. This page shows where per-page values appear — on screen in the Viewer and in every export format.
One value per page, not one per document
A field isn't a single slot on the whole document. Each page captures its own value for each field. So if your template has a field like an invoice number or a date, and your document has eight pages that all share the layout, you get eight captured values — one read from page 1, one from page 2, and so on — not a single value that the last page quietly overwrites.
This matches how operators actually think about a batch: each page is a separate form instance, and the value on page 3 belongs to page 3. It's the core of how Scanix verifies a multi-page run — you can step through the pages and confirm each one captured what it should, instead of trusting one document-level summary.
Why this matters for batches
In a batch where every page repeats the same template, per-page capture is what lets a value that reads cleanly on page 1 stay on page 1 — a blank or noisy continuation page can't reach back and clobber it. Every page keeps the value it actually carried.
Some values genuinely belong to the document as a whole rather than to one page — a value you typed in by hand, or one Scanix derived. Those stay attached to the document instead of to a single page, and they show up alongside whichever page you're viewing.
Seeing per-page values on screen
When you open a multi-page import in the Viewer, the right sidebar shows the Index Fields list — the captured values for the document. Click a page in the thumbnail strip and the panel narrows to that page: the heading changes to Index Fields · this page so you know you're looking at what was captured from the page in front of you, not a blend of the whole document.
Captured values that came from a specific page wear a small page badge — for example P1 for page 1, P2 for page 2 — and hovering it tells you exactly where the value came from ("Extracted from page 2"). The badge is your at-a-glance proof that a value is pinned to the right page.
Screenshot
The Viewer right sidebar showing the Index Fields · this page heading with a captured value carrying a P2 page badge, the matching page selected in the thumbnail strip. — shot data-capture-per-page-fields-01
Because the panel is page-scoped, an optional field that simply read nothing on the page in front of you drops out of the main list rather than showing as a blank row on every page — keeping each page's panel to the fields it actually captured. When a document spans several pages, the heading also offers a quick way to widen back out to every field in the whole document, then narrow to the active page again, without leaving the Viewer.
Multi-file imports work the same way
Per-page scoping also covers imports that span several documents in one session — clicking a thumbnail shows the fields captured on that page, so a long mixed import stays easy to verify page by page.
If a field needs a closer look or AI assistance, this is the same panel where you run extraction again or recover a value — see Analyse a document with AI.
Per-page values in your exports
The page number doesn't stop at the screen — it travels into every export format. Wherever you send your captured data, you get one record per page rather than a single flattened row per document:
Spreadsheet-style and delimited exports write one row per page, with a dedicated Page Number column so you can tell at a glance which page each row of values came from. Eight pages become eight rows, each labelled with its page.
Structured exports keep each page as its own record — a per-page Page element in XML, or a pages array in JSON — so a downstream system can read the values page by page without you having to split anything yourself.
A value that belongs to the whole document — one you typed in, or one Scanix derived — isn't tied to a single page. In a per-page export it's repeated alongside every page record, so each page's record is complete on its own.
The upshot: the page you verified on screen is the page you'll see in the file. Nothing collapses behind your back.
Screenshot
A CSV opened in a spreadsheet showing one row per page, the Page Number column highlighted, each row carrying that page's captured field values. — shot data-capture-per-page-fields-02
For the full list of formats and what each one contains, see Export formats.
Next steps
Analyse a document with AI
Run OCR plus AI extraction in the Viewer to fill or recover a field on the page you're verifying.
Export formats
See how per-page records land in each export format Scanix can produce.