Client Support Retainer Email Deliverability & Bounce Pack (With AI)

When invoices, password resets, or order confirmations suddenly bounce or land in spam, the client hears “email is broken” long before anyone proves why. On an active support retainer, you need a calm, billable path: separate facts from guesses, document DNS and header evidence, get written approval before touching production DNS, and keep a steady Gmail cadence until delivery is verified—not a vague “we’re looking into it.”

This pack fits reactive deliverability incidents while the retainer is live: rising bounces, spam-folder reports, or “nothing arrived” after a domain or ESP change. It is not a marketing warm-up guide, a one-off SSL renewal (you have a separate playbook for certs), or a full P1 outage war-room when the whole site is down.

What you need before you start

• Access: DNS read (or change) for the sending domain, ESP or Google Workspace admin if applicable, one sample failed message with SMTP bounce text or “Message-ID” from headers.
• Retainer rules: Who may approve DNS edits, whether after-hours DNS is in scope, and your hour-burn policy while waiting on the client’s registrar.
• Tools: ChatGPT for structured triage, Google Docs for the Deliverability Brief, Gmail for the client-facing sequence (you send; AI drafts).

End-to-end flow

Step 1 — Classify severity in five minutes

Paste into ChatGPT (replace bracketed fields):

You are a senior email ops consultant. FACTS ONLY vs ASSUMPTIONS.

FACTS I have:
- Domain: [example.com]
- Symptom: [hard bounces / spam folder / delayed only to Gmail]
- Started: [date/time timezone]
- Recent change: [new ESP / DNS provider move / none known]
- Sample bounce or header snippet: [paste]

Output:
1) Likely category: DNS auth / reputation / content / recipient-side / unknown
2) Three highest-probability causes ranked
3) Five concrete data points to collect next (with where to click)
4) What NOT to change yet

Label the ticket P2 deliverability unless business email is fully down for all users (then use your P1 pack). Deliverability work is often waiting on DNS propagation—say that explicitly in your first client email.

Step 2 — Build the one-page Deliverability Brief (Google Docs)

Use this outline; keep it to one screen for executives:

Section	Content
Summary	One sentence symptom + business impact (e.g., “40% of receipt emails soft-bouncing since Tuesday”).
Timeline	Last known good send, changes (ESP, website form, marketing tool).
DNS auth table	SPF, DKIM selector, DMARC policy—expected vs live (from dig or DNS UI screenshots).
Evidence	Bounce codes (550, 554), one redacted header, mail-tester or ESP dashboard link.
Plan A	Minimal DNS fixes with rollback values copied verbatim.
Plan B	If Plan A fails: open ESP ticket, temporary sending subdomain, or CR for migration.
Verification	Three test recipients (Gmail, Outlook, client domain) + success criteria.
Comms	Link to Gmail cadence below; owner on client side for DNS approval.

Ask ChatGPT to turn your FACT list into crisp table rows—then you verify every DNS value against the live zone.

Step 3 — Three-email Gmail cadence (copy-ready)

Email 1 — Acknowledge + data request (within 2 business hours)

Subject: [Retainer] Email delivery issue — facts we need today

Body bullets: symptom in client words, what is not broken (website up, etc.), link to Doc, three asks (DNS admin contact, sample bounce, approval window for DNS edits).

Email 2 — Plan + approval (after Brief is ready)

Subject: [Retainer] Proposed DNS fix for bounces — approve window [date]

Attach or link the Brief; state rollback TXT records; estimated propagation time (often 1–24h); pause on other marketing sends if recommended.

Email 3 — All-clear or pivot

Subject: [Retainer] Email delivery — verified / next step

If verified: test matrix results, what changed, monitor for 72h. If not: vendor ticket ID, hours burned, CR link for scoped migration work.

Draft each email in ChatGPT with tone “firm, non-alarmist, retainer-appropriate”; send from your Gmail after editing.

Step 4 — Fix and verify (hands-on checklist)

Work in this order to avoid making reputation worse:

1. Stop bulk/marketing sends from the broken domain until auth is correct (transactional only if urgent—and note risk).
2. SPF: One clear SPF TXT; remove duplicate or broken includes; document old value in the Brief.
3. DKIM: Confirm selector CNAME/TXT matches ESP; re-enable signing in ESP if toggled off.
4. DMARC: Start with p=none + rua if policy was missing; do not jump to reject during an incident.
5. Alignment: From domain matches DKIM d= and SPF bounce domain where possible.
6. Test matrix: Send plain-text tests to Gmail, Microsoft 365, and the client’s own domain; save “Show original” authentication results.
7. Form/API paths: If bounces come from website forms, check the envelope sender—not just the visible From address.

Log each change with timestamp in the Doc. If the client lacks DNS access, switch to your vendor escalation playbook with the ESP case number in the Brief.

Step 5 — Retainer billing and scope guardrails

• In scope: Triage, documentation, guided DNS fixes, test sends, comms cadence.
• Often out of scope (route to CR): Full ESP migration, cold-list rehabilitation, custom SMTP code, large-scale template rewrites for spam scoring.
• Hour cap while blocked: If DNS approval slips past 48h, send a short “paused pending client action” note and stop the clock per your retainer terms.

Step 6 — Monthly review log row

Add one line to the recurring retainer review doc:

Deliverability incident [date]: root cause [DNS/auth/reputation], fix [summary], residual risk [monitor DMARC reports / ESP score], owner [client DNS admin].

That row prevents the same SPF duplicate from being reintroduced when someone adds a new SaaS tool six months later.

When to use a different playbook

• Credential phishing or account takeover → security incident pack.
• Certificate expiry or TLS browser warnings → SSL renewal pack.
• Whole platform unreachable → urgent P1 pack.
• Third-party ESP outage with green DNS → vendor escalation pack.

Deliverability scares are reputation scares. A written Brief, explicit rollback DNS, and a three-email cadence keep trust—and your retainer hours—under control while inboxes start accepting mail again.