AI and Premium Content Management

What Are Autofill Columns in SharePoint?

Autofill columns are AI-powered SharePoint library columns that analyze file contents and populate metadata according to a configured prompt. They can reduce manual tagging and make documents easier to search, filter, route, and govern. The underlying document-processing service is generally available, while a newer Copilot-guided creation experience remains in preview.

Business LeaderIT DirectorSharePoint AdminAlso: SharePoint Autofill columnsAlso: Autofill columnAlso: AI-generated library metadata
Current as of July 13, 2026: As of July 11, 2026, the pay-as-you-go Autofill service is GA; the Copilot-guided creation experience remains preview and requires eligible Microsoft 365 Copilot access.

What Autofill Columns Means in SharePoint

An Autofill column combines a SharePoint library column with natural-language instructions describing the value to extract or generate from a document. For example, a prompt can identify a contract’s renewal date, classify a proposal by service line, or summarize a document into a short metadata field. Results are stored as SharePoint metadata where users and downstream processes can work with them.

Two current experiences need to be separated. Microsoft’s document processing Autofill service reached general availability and uses pay-as-you-go billing after tenant setup. Microsoft also documents a Copilot in SharePoint experience that suggests and creates Autofill columns through conversational analysis; that guided experience is still preview and is associated with Microsoft 365 Copilot licensing. Their controls, limits, and commercial paths are not interchangeable.

Why It Matters to the Business

Consistent metadata can improve search, views, retention decisions, routing, reporting, and process automation. Autofill is valuable where employees repeatedly open files to copy an obvious field into a library column. At suitable volume, automating that step can reduce effort and make important attributes available sooner.

AI-generated metadata should be treated as a proposed business value with a quality requirement. Extraction difficulty varies with document layout, wording, language, scans, and ambiguity. Process owners need an acceptable error rate, a review path for critical fields, and a rule for what happens when a value is blank or wrong. High-risk decisions should not rely on unverified output.

What SharePoint Administrators Need to Know

For the generally available document-processing service, administrators configure the required pay-as-you-go billing relationship and make the feature available to appropriate sites. They should estimate processing volume, apply cost monitoring, control who can configure columns, and review supported file and column types, limits, regional availability, and current prerequisites in Microsoft documentation.

A pilot should use representative documents and a labeled answer set. Test prompts against common formats, poor scans, missing fields, conflicting dates, and exceptions; then compare results with human-reviewed values. Administrators should also document reprocessing behavior, correction ownership, downstream flow dependencies, retention implications, and how a prompt change will be tested before production use.

What to Consider

Write a specific prompt

Name the desired value, format, context, and fallback. Ambiguous instructions produce inconsistent metadata and make quality hard to measure.

Create a benchmark set

Use representative files with known correct answers, including difficult and missing-value cases. Measure precision and correction effort before scaling.

Separate GA and preview paths

Document whether the solution uses the generally available pay-as-you-go service or the Copilot-guided preview, because access, licensing, setup, and limits differ.

Practical Takeaway

Autofill columns can turn document text into useful SharePoint metadata at scale, provided teams distinguish GA from preview experiences, validate prompts and accuracy, monitor consumption, and protect consequential processes.

How This Shows Up in the Field

Contract renewal date

A legal library uses an Autofill date column to identify the stated renewal date, then routes low-confidence or blank results to a contract owner before reminder automation begins.

Invoice metadata

An accounts-payable library extracts vendor name, invoice number, and total into columns, allowing staff to filter work and trigger a controlled review flow without opening every file first.

Autofill approaches in SharePoint

GA Autofill serviceA generally available document-processing capability that populates configured library columns and uses the applicable pay-as-you-go setup and consumption model.
Copilot-guided Autofill creationA preview Copilot in SharePoint experience that analyzes an initial set of files and helps users propose or create useful Autofill columns.

Frequently Asked Questions

Are Autofill columns generally available or preview?

Both labels can be correct for different experiences. The underlying document-processing Autofill service is generally available. The newer Copilot in SharePoint experience that suggests and creates Autofill columns is separately in preview as of July 11, 2026.

How are Autofill columns licensed?

The generally available document-processing service uses Microsoft’s pay-as-you-go setup. The Copilot-guided preview is associated with a Microsoft 365 Copilot license. Confirm tenant eligibility, consumption pricing, and current terms for the chosen path.

Can Autofill metadata be wrong?

Yes. Results are generated from file content and a prompt, so ambiguous language, scans, unusual layouts, missing facts, or prompt design can produce incorrect or blank values. Test against known answers and review consequential fields.

Official Microsoft References

Editorial review date: July 13, 2026. Product status and licensing can change; confirm current Microsoft documentation before implementation.

Make document metadata useful and trustworthy

6SC can select the Autofill scenario, govern billing, test prompts and accuracy, integrate exceptions, and scale responsibly.

Talk with 6SC
Scroll to Top