Getting Started

How to use this docs page

This page is meant to answer one operational stage of the DullyPDF workflow well enough that you can run a controlled test without guessing. Read the sections below, validate the behavior against one representative document, and only then move to the next linked page.

That order matters because most setup failures come from mixing detection, mapping, fill validation, and sharing into one unstructured pass. A narrower review loop keeps troubleshooting faster and makes the template easier to trust once you save it for reuse.

Quick-start path

1. Upload a PDF (50MB max). Non-PDF or larger files are blocked before upload.
2. Wait for detection results, then check low-confidence items first.
3. If naming is inconsistent, run Rename or Rename + Map (with schema ready).
4. Clean geometry in the editor, then verify field types.
5. Save the template, then either publish Fill By Link or load CSV/XLSX/JSON rows for Search & Fill.
6. Run one controlled Search & Fill or respondent-selection test before production use.

Best-practice order

For consistent results, keep this order:

• Detect first.
• Rename before mapping if labels are inconsistent.
• Map after schema upload so field names align to column headers.
• Finalize geometry and field types before large batch filling/exporting.

Practical credit plan: when you need both operations, use Rename + Map to reduce round trips. Credit cost remains bucketed by page count either way.

First-run checklist

• Confirm each required form area has a field candidate.
• Verify page assignment for fields spanning multiple pages.
• Check checkbox groups/options (`groupKey`, `optionKey`) before filling.
• Run one test record through Search & Fill before saving templates.
• If using Fill By Link, verify the public form questions read clearly on a phone before sharing.
• Validate one date field and one checkbox group in the final output PDF.

First 30 minutes

1. Pick one recurring PDF instead of a full packet.
2. Run detection and clean the low-confidence items immediately.
3. Rename or map only after the field geometry is believable.
4. Fill one realistic record, inspect the PDF, clear it, and fill again.
5. Only after that should you publish a Fill By Link, save packet groups, or expose API Fill.

Most common first-run mistakes

• Uploading several document variations before one canonical template is stable.
• Running mapping before low-confidence geometry and checkbox cleanup are reviewed.
• Judging the workflow from field detection alone instead of from one full fill cycle.
• Publishing links or sharing templates before date, checkbox, and repeated-name fields are tested with a real record.

What good output looks like

• High-confidence fields require little or no geometry correction.
• Mapped field names resemble your schema headers (snake_case style in most cases).
• Yes/no checkbox pairs always end with exactly one selected option after fill.
• Search returns expected records quickly with either contains or equals mode.