AEP Destination Activation Troubleshooting Guide

Understanding AEP Destination Types

Before troubleshooting, understand the two main destination categories:

Type	Examples	Latency	How It Works
Streaming	Facebook, Google Ads, The Trade Desk	Minutes to hours	API-based, near real-time updates
Batch/File	SFTP, S3, Azure Blob, Data Landing Zone	Scheduled (daily/hourly)	File exports on schedule

Different destination types have different failure modes. Know which type you're troubleshooting.

Common Failure #1: Authentication Errors

Symptoms

• Dataflow shows "Failed" status
• Error message mentions "401 Unauthorized" or "403 Forbidden"
• Activation worked before but suddenly stopped

Root Causes

• Expired credentials: OAuth tokens, API keys, or passwords expired
• Revoked access: Destination platform revoked the integration
• Permission changes: User who set up the connection lost access
• Rate limiting: Too many requests triggered security blocks

How to Fix

1. Go to Destinations → Browse
2. Find the failing destination and click into it
3. Look for "Re-authenticate" or "Update credentials" option
4. Re-authorize with valid credentials
5. Test with a small audience before full activation

Common Failure #2: Mapping Errors

Symptoms

• Partial activation (some profiles succeed, others fail)
• Error mentions "required field missing" or "invalid format"
• Destination receives data but can't process it

Root Causes

• Missing required fields: Destination requires fields not mapped
• Data type mismatch: String sent where number expected
• Format issues: Phone numbers, dates in wrong format
• Null values: Required field has no value for some profiles

How to Fix

1. Review destination's required fields in documentation
2. Go to Destinations → Dataflows and check the mapping
3. Ensure all required fields are mapped
4. Use transformation functions for format conversion:

// Example: Phone number formatting
// Source: "+1-555-123-4567"
// Required: "15551234567"

// Use calculated field to strip formatting
replace(replace(phone, "-", ""), "+", "")

Common Failure #3: Consent/Governance Blocking

Symptoms

• Activation runs but exports 0 profiles
• Monitoring shows "Consent policy violation"
• Segment has profiles but nothing activates

Root Causes

AEP's Data Usage Labeling and Enforcement (DULE) blocks activation when:

• Data labels on fields conflict with destination marketing actions
• Profiles haven't consented to the specific use case
• Governance policies prevent certain data from leaving AEP

How to Diagnose

1. Go to Policies → Browse
2. Check which policies apply to your destination
3. Review data labels on the fields being activated
4. Check profile consent status in Profile Viewer

How to Fix (If Appropriate)

• Update consent records if customers have actually consented
• Modify segment to only include consented profiles
• Review and adjust data labels if they're incorrectly applied
• Create destination-specific consent segments

Common Failure #4: Segment Not Activating

Symptoms

• Segment shows profiles in AEP but destination reports empty
• No errors in monitoring, but no data flowing
• Worked before but stopped after segment changes

Root Causes

• Segment not mapped to destination: Forgot to add segment to dataflow
• Evaluation timing: Batch segment hasn't evaluated yet
• Merge policy mismatch: Segment uses different merge policy than expected
• Identity not present: Profiles lack the identity type destination needs

How to Fix

1. Verify segment is added to the destination dataflow
2. Check segment evaluation schedule (batch vs streaming)
3. Confirm profiles have the required identity (e.g., email hash for Facebook)
4. Test with a known profile using Profile Viewer

Common Failure #5: Export Schedule Issues

For Batch Destinations

File-based destinations export on schedule. Common issues:

• Schedule not configured: Dataflow created but no export schedule set
• Timezone confusion: Schedule set in wrong timezone
• Incremental vs Full: Expecting full export but configured for incremental

How to Check

1. Go to Destinations → Browse → [Your Destination]
2. Check the Scheduled tab for export schedule
3. Verify timezone matches your expectations
4. Check export type (incremental changes vs full snapshot)

Export Type	What It Sends	Use Case
Full	All qualified profiles every time	Complete list sync, data warehouse loads
Incremental	Only changes since last export	Frequent updates, bandwidth optimization

Monitoring and Diagnostics

Using the Monitoring Dashboard

1. Navigate to Monitoring → Destinations
2. Filter by destination or dataflow
3. Check Dataflow runs for success/failure status
4. Click into failed runs for error details

Key Metrics to Watch

• Records received: Profiles AEP tried to activate
• Records activated: Successfully sent to destination
• Records failed: Rejected by destination or AEP
• Records excluded: Filtered by consent/governance

API-Based Monitoring

// Get dataflow run details
GET /flows/{FLOW_ID}/runs

// Response includes:
{
  "items": [{
    "id": "run-123",
    "status": "success",
    "metrics": {
      "recordsReceived": 50000,
      "recordsActivated": 49850,
      "recordsFailed": 150
    },
    "errors": [...]
  }]
}

Destination-Specific Issues

Facebook/Meta Ads

• Audience too small: Facebook requires minimum audience size (~100)
• Hashing mismatch: AEP hashes email/phone; don't pre-hash
• Business Manager issues: Verify correct ad account is connected

Google Ads/DV360

• Customer Match eligibility: Account must qualify for Customer Match
• Match rate delays: Google processing can take 24-48 hours
• Policy violations: Content policies can reject audiences

SFTP/Cloud Storage

• Connection timeout: Firewall blocking AEP IPs
• Permission denied: Write permissions not set on target folder
• Disk space: Target storage is full

Prevention Best Practices

1. Test with small audiences first before scaling to millions
2. Set up alerts for dataflow failures in Monitoring
3. Document credential expiry dates and set renewal reminders
4. Use service accounts not individual users for auth
5. Validate mappings whenever schema changes
6. Monitor match rates at destination to catch issues early