Deliverability Health Check

Audit your email deliverability by analyzing bounce rates, open rate anomalies, and sending patterns across reps and sequences to catch domain reputation issues early.

Instructions

When a user wants to check deliverability health, query the Amplemarket analytics engine for bounce and engagement data, classify the overall health status, and flag any reps or sequences that pose a risk.

Steps

1. Clarify scope and concerns. Ask the user:

• "What timeframe would you like to check? (default: last 30 days)"
• "Any specific concerns? For example: high bounces, low open rates, or suspected spam folder placement."

If the user does not specify, default to the last 30 days and run a comprehensive check.

1. Submit analytics questions by calling mcp__claude_ai_Amplemarket__ask_analytics for each of the following (adjust the timeframe accordingly):

• "What is the bounce rate by rep in the last [timeframe]?"
• "What is the bounce rate by sequence in the last [timeframe]?"
• "What is the overall bounce rate in the last [timeframe]?"
• "What are the open rates by rep in the last [timeframe]?"
• "How many emails were sent vs delivered in the last [timeframe]?"

Store each returned request_id for polling.

1. Wait approximately 20 seconds, then call mcp__claude_ai_Amplemarket__get_analytics_result for each request_id to retrieve the answers.
2. Handle pending results. If any result is still processing, wait another 20 seconds and retry. Repeat up to 3 total attempts per request. Proceed with available data if some results remain unavailable.
3. Compile the deliverability health report with three components:

Overall Health Score:

• Healthy (below 2% bounce rate): Deliverability is in good shape. No immediate action required.
• Warning (2-5% bounce rate): Some issues need attention. Investigate the sources of bounces.
• Critical (above 5% bounce rate): Immediate action required. Domain reputation is at risk.

Per-Rep Breakdown:

Per-Sequence Breakdown:

1. Flag critical issues with specific, actionable callouts:

• Any rep with a bounce rate above 5%: "Marcus has a 6.1% bounce rate. This can damage your sending domain. Investigate his lead sources immediately."
• Any sequence with a bounce rate above 3%: "Cold Outreach v3 has an 8% bounce rate vs. the 1.5% team average. The lead list feeding this sequence likely contains stale or invalid emails."
• Any rep or sequence with an open rate below 20%: "An open rate of 15% may indicate emails are landing in spam folders rather than inboxes. Review the sending domain, subject lines, and email content for spam triggers."

1. Identify patterns and root causes. Connect the dots between metrics:

• High bounce + specific sequence = lead list quality issue
• High bounce + specific rep = that rep may be importing unvalidated lists
• Low open rate across all reps = potential domain reputation problem
• Low open rate for one rep only = that rep's email account may have deliverability issues

1. Generate recommendations based on the health classification:

Critical actions:

• "Pause [sequence/rep] immediately until the bounce rate issue is resolved."
• "Run all lead lists through email validation before importing."
• "Check your sending domain's reputation on Google Postmaster Tools and other monitoring services."

Warning actions:

• "Enable email validation on future lead list imports to catch invalid addresses before sending."
• "Review the lead sources for sequences with above-average bounce rates."
• "Monitor weekly. If bounce rates continue to climb, escalate to critical."

Healthy maintenance:

• "Deliverability looks good. Keep monitoring monthly to catch issues early."
• "Continue validating lead lists before import to maintain this healthy status."

Offer next steps: "Want me to check which lead lists are feeding the high-bounce sequences?" or "Should I run a full team audit to compare deliverability across all reps?"

Important Notes

• Submit all five analytics questions in parallel to minimize wait time.
• Bounce rate is the primary health indicator. Open rate is a secondary signal because privacy features (like Apple Mail Privacy Protection) can inflate open rates, making them less reliable as a standalone metric.
• A bounce rate above 5% is an urgent issue. Always flag this prominently, even if the user asked about something else.
• Very low send volume (under 100 emails) can produce misleading bounce rates. A single bounce out of 20 emails is a 5% bounce rate but may not indicate a real problem. Always consider sample size when classifying health.
• If the user mentions spam complaints, note that Amplemarket analytics may not track spam complaint rates directly. Suggest checking Google Postmaster Tools or similar services for complaint data.

Examples

Example 1: Full Deliverability Health Check

User prompt: "Check my deliverability. I want to make sure we are not having any issues."

What the skill does:

1. Confirms the 30-day default timeframe.
2. Submits five analytics questions via mcp__claude_ai_Amplemarket__ask_analytics.
3. Waits 20 seconds, then polls each request_id with mcp__claude_ai_Amplemarket__get_analytics_result.
4. Compiles the health report with per-rep and per-sequence breakdowns.

Example output:

DELIVERABILITY HEALTH CHECK - Last 30 Days

Overall Health: WARNING (2.8% bounce rate)

Your overall bounce rate of 2.8% is above the healthy threshold of 2%. While not yet critical, this needs attention to prevent domain reputation damage.

Summary

Per-Rep Breakdown

Per-Sequence Breakdown

Critical Issues

1. Marcus Brown - 6.1% bounce rate. This is well above the 2% healthy threshold. His bounce rate alone is pulling the team average into warning territory. His lead sources need immediate review. His low open rate (38.4%) may also indicate some emails are landing in spam.
2. New Logo Prospecting sequence - 8.0% bounce rate. This is the primary source of bounce issues. The lead list feeding this sequence likely contains a high percentage of invalid or outdated email addresses. Pause this sequence immediately and validate the email list before resuming.

Recommendations

• Pause the "New Logo Prospecting" sequence until the lead list is cleaned and validated.
• Review Marcus Brown's lead import process. He may be adding contacts without email validation.
• Monitor David Kim's bounce rate (2.4%). It is borderline and could worsen.
• The rest of the team and sequences are healthy. No action needed for them.

Example 2: Investigating a Specific Rep

User prompt: "Marcus has been getting a lot of bounces. Can you check his deliverability?"

What the skill does:

1. Submits analytics questions focused on Marcus's sending activity.
2. Polls results and presents a detailed individual breakdown.

Example output:

DELIVERABILITY REPORT - Marcus Brown (Last 30 Days)

Marcus's bounce rate is 3x the team average. Combined with a below-average open rate, this suggests two issues:

1. Lead list quality - many email addresses are invalid or outdated.
2. Possible spam folder placement - the low open rate may indicate mailbox providers are flagging his sends.

Recommended actions:

• Immediately audit Marcus's recent lead imports for email validity.
• Pause any sequences Marcus is running until bounce rate is under control.
• Check if Marcus is using a different sending domain or email account that may have reputation issues.
• Run his upcoming lead lists through email validation before adding them to sequences.

Example 3: Clean Bill of Health

User prompt: "Quick deliverability check - are we in good shape?"

What the skill does:

1. Runs the standard five analytics queries for the last 30 days.
2. Finds all metrics within healthy thresholds.
3. Returns a brief, positive report.

Example output:

DELIVERABILITY HEALTH CHECK - Last 30 Days

Overall Health: HEALTHY (1.4% bounce rate)

Everything looks good. Your bounce rate is well below the 2% threshold, and open rates are healthy across all reps and sequences.

No action needed. Continue monitoring monthly to maintain this healthy status. Keep validating lead lists before import to prevent future issues.

Troubleshooting