The Email Deliverability Runbook

Every brand blames the creative when email revenue drops. The subject line must have missed. The offer must have been weak.

The truth is usually more expensive. The campaign never reached the inbox.

Validity’s 2025 benchmark report put global average deliverability at 83.1%.¹ That means roughly one in six legitimate marketing emails never lands where a customer can see it. If your lifecycle model assumes those sends reached real people, the forecast is wrong before finance opens the sheet. At scale, a 10-point inbox gap can erase tens of thousands of dollars of retention revenue from a healthy list.

Treat that gap as an operating failure, not a copy problem. Deliverability slips when authentication breaks, DNS records drift, low-engagement segments stay on the list too long, or sender reputation gets dragged down by bad links and tired audiences. If you want lifecycle to behave like a revenue system, you need a standing runbook for inbox placement.

You can see the failure pattern in the weekly review. Clicks fall, the team debates creative, and nobody checks whether Gmail started routing the campaign somewhere customers never saw it. By the time finance asks why retention underperformed, the operator is already working with the wrong diagnosis.

This article gives you that runbook in five parts:

1. Measure the problem with the right metric.
2. Triage the technical and behavioral causes.
3. Run a maintenance rhythm that catches drift early.
4. Recover incidents with a defined sequence.
5. Harden the system with automation, identity, and reporting.

The Rule: Deliverability Is a Revenue Control, Not a Creative Variable

Deliverability belongs with lifecycle operations, not just campaign execution.

Delivery only tells you the receiving server accepted the message. Deliverability tells you whether the message landed where a customer could act on it: the inbox, not spam, and not a dead corner of the mailbox. That is the metric that protects retention revenue.

Open rates no longer solve this. Apple Mail Privacy Protection (MPP), Gmail image handling, and other privacy features made opens too noisy to run the system. Server acceptance and clicks tell you more. Inbox placement tells you whether the campaign had a chance to work at all.

That is why a deliverability runbook has to start with operator logic. If inbox placement drops, creative performance, offer performance, and even retention forecasting get harder to trust. You do not have an email problem at that point. You have a visibility failure inside a revenue channel.

What the Runbook Has to Catch Early

The first job is detection. You need to know when deliverability drift starts before it turns into a quarter-end revenue surprise.

Watch for the early symptoms:

• Sudden click-rate declines without a matching change in send volume or offer strategy
• Unusual spikes in bounces or complaints
• Gmail throttling or greylisting delays
• Falling server acceptance rates
• Lower inbox placement in seed-list testing

Greylisting is a temporary rejection tactic mailbox providers use to slow spam bots. Legitimate servers retry. Low-quality senders often do not.

Then look at the core operating metrics:

The cross-provider spread is wide enough to matter operationally. Microsoft inboxes often underperform Gmail. Promotions-tab placement is common in ecommerce. Mailgun reported that only 8.78% of ecommerce marketing emails landed in the Gmail Primary tab, with the rest routed elsewhere.² That does not mean the send failed. It does mean lifecycle teams need tighter operating assumptions about visibility.

Phase 1: Run Triage in the Right Order

When inbox placement drops, do not start by rewriting subject lines. Start with the infrastructure and the sender signals that mailbox providers evaluate first.

1. Authentication

Check SPF, DKIM, and DMARC before anything else.

Use MXToolbox or DMARCian to validate the setup.

Example SPF record:

v=spf1 include:_spf.google.com include:sendgrid.net ~all

Know the operating differences:

• ~all is a soft fail
• -all is a hard fail
• DMARC policy options are p=none, p=quarantine, and p=reject

Start with p=none if you need monitoring. Move to p=quarantine once alignment is clean and the reporting is stable. Use reporting from Cloudflare or Postmark so the domain team sees failures before a mailbox provider makes the decision for them.

2. DNS Integrity

Authentication cannot hold if the DNS foundation is weak.

Confirm that the domain has the right A records and MX records, even if the domain does not actively receive mail. Some lightweight hosting setups rely too heavily on CNAMEs. That can break routing, weaken trust, or create enough ambiguity that mailbox providers score the sender more aggressively. Missing MX records are one of the simplest ways to make a legitimate sender look careless.

3. Sender Reputation and IP Health

Check Google Postmaster Tools for domain and IP reputation.

If the domain drops from High to Medium or Low, slow down. Pause low-engagement sends for 48 hours. Concentrate volume on the most active subscribers so mailbox providers see cleaner behavior while you fix the issue.

4. Engagement Quality

Mailbox providers now rank senders by what recipients do after delivery.

Sort the audience by last click date. If fewer than 30% of subscribers clicked in the past 90 days, segment aggressively and re-permission the rest. Clicks are the cleanest engagement signal left. If you keep mailing subscribers who never respond, the mailbox provider learns the same lesson the operator should have learned first.

See also: Attribution Triage for Operators: Make Marketing Data Useful Without Breaking Your P&L

5. Content and URL Reputation

Mailbox providers do not just read the sender. They read the links.

They watch clicks, deletes, replies, and scroll-depth proxies where available. They also score the domains inside the email. One bad redirect, questionable link shortener, or recycled promotional domain can push a healthy sender into a weaker placement band.

6. Placement Testing

Run seed-list testing before you call the system healthy.

Use GlockApps or Validity Everest. Send to a seed list of at least 100 addresses spread across Gmail, Outlook, Yahoo, and Apple Mail. A healthy sender should see at least 85% reach inbox placement across the mix. If the seeds go sideways while campaign creative remains stable, you have a deliverability problem, not a messaging problem.

Phase 2: Hold a Standing Maintenance Rhythm

Once triage is clean, the real work is keeping it clean.

Deliverability drift is rarely dramatic at first. It usually shows up as a small complaint increase, a weaker click cohort, a throttled domain, or a template pattern that starts losing trust. The maintenance rhythm is what catches that drift while the fix is still cheap.

Daily

• Review bounces and complaints
• Confirm suppression automations are firing
• Check Postmaster dashboards for reputation movement

Weekly

• Run inbox placement tests
• Clean unengaged segments
• Rotate templates and offers to reduce filter fatigue

Monthly

• Audit SPF, DKIM, and DMARC
• Verify DNS records
• Review click-rate trends by segment and provider

Only about 60% of senders clean their lists regularly.¹ The rest let engagement decay until inbox performance slips and revenue follows it down. A disciplined QA rhythm can recover 10% to 15% of lost lifecycle revenue in a quarter if the list and infrastructure are still repairable.

Phase 3: Recover Incidents Without Making Them Worse

When deliverability is already impaired, the operator needs sequence more than theory.

Use this recovery order:

1. Remove bad volume

Suppress low-engagement subscribers first. If the sender is mailing people who have not clicked in 90 days, cut that volume before doing anything else. You need the mailbox providers to see cleaner response signals fast.

2. Check list quality

Avoid purchased or scraped lists. Validate addresses quarterly with NeverBounce or ZeroBounce. If spam traps are in play, the recovery window gets longer and the reputation damage becomes harder to reverse.

3. Slow down dedicated infrastructure

If you use a dedicated IP, warm it with discipline. Increase send volume gradually over the first two weeks. Shared pools such as Klaviyo’s can absorb some of that complexity, but high-volume brands still need to respect the warm-up curve.

4. Audit the recent change set

Do not treat the issue like bad luck. Look at what changed in the last two weeks:

• New sending domain
• Authentication edit
• Larger campaign volume
• New list source
• New redirect or link domain
• Sudden increase in promotional frequency

Incidents usually follow a concrete change. If you cannot name that change, you are not done diagnosing the problem.

The Behavioral Layer Mailbox Providers Actually Score

The recovery sequence above works because mailbox providers are scoring more than your records.

A lot of deliverability guidance stops at DNS. That is only part of the system now.

Mailbox providers also rank senders on recipient behavior:

Positive signals:

• Clicks
• Replies
• Adding the sender to contacts
• Pulling a message out of spam

Negative signals:

• Ignoring repeated messages
• Quick deletions
• Spam reports

That is why send-time discipline, audience segmentation, and content quality still belong inside the runbook. They do not sit next to deliverability. They help determine it.

To improve those signals:

• Send at consistent times
• Use a recognizable From address tied to your domain
• Ask for low-friction engagement such as replies or polls
• Mix HTML and plain-text formats where appropriate
• Keep link domains clean and reputable

Hardening the System: Identity, Automation, and Economics

Once the runbook works, strengthen the sender in three ways. First, make the sender more trustworthy in the inbox. Next, remove manual failure points from the maintenance rhythm. Then show finance what the gain is worth.

Identity Layers

Authentication gets you to the baseline. Visual identity layers help reinforce trust after that.

BIMI (Brand Indicators for Message Identification)

BIMI displays a verified logo in supported inboxes such as Gmail and Yahoo.

Requirements:

• SPF, DKIM, and DMARC alignment
• DMARC policy of p=quarantine or p=reject
• Verified SVG logo hosted under HTTPS
• Optional VMC certificate for added validation

Apple Business Connect Branded Mail

Apple Mail’s Branded Mail lets verified businesses display their logo, domain, and a signed identity label directly in the inbox.

Neither tool fixes a weak sender. Both help a strong sender convert trust into engagement.

Automation

Manual checks do not scale.

Automate the recurring tasks:

• Klaviyo: auto-suppress inactive subscribers
• Zapier: trigger Slack alerts when bounce rate exceeds 2%
• Cloudflare or Postmark: collect DMARC reports automatically
• Google Sheets or Data Studio: log daily deliverability metrics through an API
• Validity Everest: schedule weekly inbox placement tests

Automation does not replace judgment. It keeps the operator from relying on memory.

Economics

Deliverability changes cash flow fast enough to deserve executive attention.

If you send 100,000 emails a month and revenue per delivered email is $1.50:

• At 90% deliverability, you earn $135,000
• At 97% deliverability, you earn $145,500

That is a $10,500 gain from deliverability alone. No change to offer. No change to creative. No extra send volume.

Case Study: Dr. Brighten’s Lifecycle Stack

Before the fix, lifecycle performance fell 22% in open rate and 17% in click rate while Gmail throttled sends.

The team changed the operating system, not the headline. They ran daily quality assurance (QA), automated suppression, completed a full DNS audit, and added DMARC reporting through Cloudflare.

Six weeks later, deliverability rose from 87% to 97%. Lifecycle revenue increased 11% quarter over quarter.

That is the pattern to notice. The recovery came from infrastructure discipline and audience control, not from more creative pressure.

Build the Runbook Into the Team’s Weekly Operating Model

The runbook only matters if the team can run it on schedule and see who owns the next check.

Put deliverability next to revenue in the lifecycle dashboard. Track deliverability rate next to revenue per send. When one slips, the other usually follows. That pairing keeps the team from treating inbox placement like a technical side metric instead of a control on retention revenue.

Use benchmark ranges as reference points, not as excuses:

Regional averages vary too: Europe 89%, North America 83%, Asia-Pacific (APAC) 85%.³ Use those numbers to calibrate expectations, not to excuse a weak sender.

Your Next Move This Week

Do not start with a redesign of the email calendar.

Start with one audit block:

1. Pull your current deliverability rate, bounce rate, complaint rate, and click rate.
2. Validate SPF, DKIM, and DMARC.
3. Review the last 90-day click cohort and suppress the coldest segment.
4. Run one seed-list inbox placement test.
5. Put the daily, weekly, and monthly tasks on the calendar with owners.

Once those five moves are scheduled, put the review on repeat. Deliverability stays healthy when someone checks the sender every day, someone tests placement every week, and someone audits the foundations every month. That is when it stops acting like an occasional fire drill and starts behaving like a system the team actually runs.

Key Takeaways

• Deliverability is the visibility control for lifecycle revenue, not a creative side issue.
• Delivery and deliverability are different metrics; inbox placement is the one that governs revenue potential.
• The real runbook sequence is detect, triage, maintain, recover, and harden.
• Authentication, DNS integrity, sender reputation, engagement quality, and testing form the core of diagnosis.
• A standing QA rhythm and clean suppression policy protect revenue better than reactive campaign tweaks.