Getting recommendations from Fern

Fern analyzes your support tickets and product changes to identify documentation gaps. Recommendations surface early, before patterns become obvious, so you can address issues before they pile up.

Accessing recommendations

Find recommendations in your dashboard in the Inbox on the Fern page.

Click any recommendation to see the evidence behind it specific tickets, commits, or releases that triggered the suggestion, plus Fern's proposed action.

How Fern identifies gaps

Fern monitors two signal types:

Support patterns

Fern analyzes tickets from your connected support tools looking for:

• Recurring questions

• Issues requiring multiple back-and-forth messages

• High-priority or escalated topics

• Failed search attempts

Product changes

Fern tracks commits, pull requests, and releases from GitHub or Linear to catch:

• New features without documentation

• Breaking changes affecting existing users

• UI updates making screenshots outdated

• Deprecations needing migration guides

Prioritization

Fern assigns priority based on:

• Volume - How many users are affected

• Frequency - How often the issue appears

• Recency - Whether it's new or ongoing

• Impact - Whether it blocks users or is nice-to-have

When recommendations generate

On-demand scans: Trigger a scan anytime to analyze a specific lookback period (7 days, 31 days, etc.). Scans take 5-30 minutes and analyze up to ~3,000 tickets, prioritizing recent and relevant data.

Automatic daily monitoring: Once enabled, Fern runs daily scans in the background to catch emerging patterns without manual checks.

Acting on recommendations

From your Inbox, you can:

• Assign to Fern - Creates a task for Fern to draft. You'll review before publishing.

• Reject - Dismiss recommendations that don't apply. These will be used to train Fern.

Use the evidence panel to review the actual tickets or commits before deciding what to prioritize.