# Read fields from the last or a specific page

Point an index-field rule at the page that actually carries the value — First Page, Last Page, or a specific page number — and know the one silent failure mode when a document is shorter than the rule expects.

Product: capturepoint · Audience: catalog-administrator · Time: 15 minutes · Last verified: 2026-09-05

Canonical: https://help.ademero.com/capturepoint/templates/read-fields-from-the-last-or-a-specific-page

**At the end of this guide, the rule that reads your invoice total looks on the last page — where the total actually lives — instead of the first, and you'll know the silent way a page-targeted rule fails so blank fields don't puzzle you later.**

By default, an index-field rule reads from a document's **first** page. Fine for a header value; wrong for a grand total on the final page of a variable-length invoice, or a signature date on page 2 of a fixed form. The page target is one setting deep.

## Set a rule's page

1. In CapturePoint, open **Index** and choose **Index-Field Automation** — the **Index-Field Rules** screen, listing the rules under your selected **Classification** and **Document Layout**.

2. On the rule to change, choose **PROPERTIES** — the **Rule Properties** dialog — and under "What would you like to configure?", pick **Identification**.

3. Find the **Page Number** choices:
   
   | Choice | Reads from |
   |---|---|
   | **First Page** | The document's first page — the default |
   | **Last Page** | The final page, wherever it falls — the variable-length answer |
   | **Specific Page** | The page number you type — **counted within each document**, not the scanned batch |

4. Save, then prove it: **RECOGNIZE CURRENT** re-runs rules on the open document — the field should now fill from the targeted page.

> **NOTE:** The Page Number choices appear for **Text Extraction** and **Barcode Extraction** rules — the kinds that read *from a page*. Rule types that don't read pages (Static Value, Not Blank, Preceding Document) have nothing to target, so the section stays hidden for them; that's by design, not a missing setting.

## The silent failure to know

A **Specific Page** rule pointed past a document's end — page 3 of a two-page document — doesn't error. It simply yields nothing, and the field arrives blank. So when a page-targeted field is blank on *some* documents and fine on others, count pages before touching the rule: short documents are almost always the story.

Design around it:

- **Last Page** beats **Specific Page** whenever "the end" is what you mean — it can't miss.
- **Specific Page** fits fixed forms where page N always exists — and pairs well with a document layout that identifies those forms specifically.

**Success check:** run a mixed test batch — documents of different lengths, including one shorter than any specific page number you used. The targeted fields fill on every document long enough, and the short one's blank now reads as information, not mystery.

## What's next

- [Train CapturePoint templates](https://help.ademero.com/capturepoint/templates/train-capturepoint-templates)
- [CapturePoint extracted a blank or wrong field](https://help.ademero.com/capturepoint/templates/capturepoint-extracted-blank-or-wrong-field)