How to turn on recommendations from Fern

Recommendations from Fern help you identify documentation gaps based on support tickets and product changes. To receive recommendations, you need to set up monitoring first.

Where recommendations appear

Once monitoring is enabled, recommendations show up on the Fern Inbox. This is where you'll review, prioritize, and assign documentation tasks to Fern.

The Signals dashboard still exists as a separate data view showing raw support requests and product releases that feed into recommendations.

Setting up monitoring

You can start the setup process from several places:

• Dashboard Opportunities card

• Fern Inbox page (/fern)

• Signals dashboard in your sidebar

Look for a "Setup monitoring" button or card. Click it to begin.

Step 1: Connect integrations

The setup dialog will check your integrations. You can connect:

Your codebase

• GitHub

Support tools:

• Crisp

• Intercom

• Zendesk

• See all integrations

You need at least one integration to enable monitoring. Both types are optional, but codebase integrations generate more accurate recommendations by analyzing product changes directly.

Step 2 (optional): Choose a lookback period

Select how far back Fern should analyze for your initial scan:

• 7 days - Recent activity only

• 14 days - Good starting point

• 31 days - Default, balances depth and speed

• 90 days - Quarterly view

• 365 days - Full year (takes longer)

The scan processes up to about 3,000 tickets, focusing on the most recent and relevant items if you have more.

Step 3: Start monitoring

Click "Start monitoring" to begin the scan. Fern will:

1. Import tickets and product changes from your connected tools

2. Read and analyze the data

3. Identify documentation gaps

4. Create prioritized recommendations

The process takes 5-30 minutes depending on data volume. You can leave the page - the scan runs in the background.

After the first scan

When the scan completes, recommendations appear in your Fern Inbox. You'll see:

• Documentation gaps surfaced from support patterns

• Priority rankings based on ticket volume and recency

• Context from the support tickets that triggered each recommendation

Fern automatically runs daily scans going forward, so new recommendations appear as patterns emerge - no manual triggering needed.

Running manual scans

Even with daily automatic scans enabled, you can trigger manual scans anytime:

• After connecting a new integration

• Following a major product launch

• To analyze a specific historical period

Go to the Signals dashboard, click "Check for new tickets," choose your lookback period, and start the scan.

Troubleshooting

Setup button is disabled

You need at least one integration connected before you can enable monitoring. Connect a codebase tool (GitHub or Linear) or support tool (Crisp, Intercom, or Zendesk) first.

No recommendations after scanning

If the scan completes but your Inbox is empty:

• Verify your integrations have data during the lookback period you selected

• Make sure you have some baseline documentation - Fern compares support questions against what exists

• Try a longer lookback period to capture more data

Scan fails or shows an error

If you see an error:

• Check that your integrations are still authenticated (they may need reconnection)

• Verify the integrated tools are accessible (no API outages)

• Try a shorter lookback period if you have very high ticket volume

• Use the retry button to attempt the scan again

Can't find the setup button

If you don't see a setup prompt, monitoring may already be enabled. Check your Fern Inbox for existing recommendations or visit the Signals dashboard to see your monitoring status.

Next steps

Once recommendations are flowing:

• Review items in your Fern Inbox and assign high-priority gaps to Fern

• Explore the Signals dashboard to understand the raw data feeding recommendations

• Connect additional integrations for more comprehensive coverage

The more data sources you connect, the better Fern gets at identifying documentation gaps.