> For the complete documentation index, see [llms.txt](https://docs.documentero.com/documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.documentero.com/documentation/ai/convert-document-to-template-ai.md).

# Convert Document to Template - AI

AI Convert is designed to take a completed `.docx` document and turn it into a reusable template with dynamic fields. Its goal is simple: keep your original document structure and wording, but replace the parts that should vary (names, dates, amounts, IDs, rows, options, etc.) with placeholders.

**Instructions**

* Go To Documents/Templates
* Click "AI - Convert document to template"
* Provide your document in .docx format in modal and click "Convert with AI"
* After processing you will be moved to new template overview page with template preview and all detected fields.

<figure><img src="/files/1H29UF5PqVTp6Ne4Ot2N" alt=""><figcaption></figcaption></figure>

#### Before conversion

<figure><img src="/files/91kmg1BdTqRVw0k58FOM" alt=""><figcaption></figcaption></figure>

#### After conversion

<figure><img src="/files/S4PGoLZKJZqztDwxogpu" alt=""><figcaption></figcaption></figure>

#### What AI Convert looks for

1. **Fill-in blanks and empty slots**\
   It detects places where users are expected to enter data, including dotted lines, underlined blanks, and empty cells in forms/tables.
2. **Table context**\
   It analyzes table structure to understand whether a table is:

* a fixed label/value form (for example: Company | \[empty value]), or
* a repeatable data table (for example: multiple item rows, measurements, participants).

3. **Repeated patterns**\
   If multiple rows or list items represent the same type of record, it tries to convert them into repeatable template sections instead of creating duplicated static placeholders.
4. **Checkboxes and selectable options**\
   It identifies real checkbox markers and converts them into dynamic options while keeping labels visible.
5. **Header/footer variables**\
   It also scans header and footer content, where important fields like dates, references, or document numbers are often placed.
6. **Context-based field naming**\
   It tries to name placeholders based on nearby labels and sentence context, so fields are meaningful and easier to use later.