Newsletter audit checklist: 47 checks across 6 layers

What a newsletter audit actually is

Most "newsletter audit checklist" posts on page one of Google read like a feature tour of a generic ESP. They mix subject lines and DMARC records on the same line as if both live at the same altitude.

An audit is a specification check. It asks one question per layer: is the system in spec, yes or no? That is different from "did the last send do better than the one before" (analytics review) or "does variant A out-perform B" (A/B test). The audit catches the part of the program that has been broken for six weeks and nobody noticed.

The 47 checks below sit on six layers. Work bottom up. A deliverability fault makes every layer above it look like a content problem when it is not.

• Layer 1 - Deliverability (8 checks). Does the message arrive?
• Layer 2 - List health (6 checks). Is the audience real?
• Layer 3 - Content (10 checks). Is the message worth reading?
• Layer 4 - Render and design (7 checks). Does it look right where it lands?
• Layer 5 - Conversion (8 checks). Does it move someone to do the thing?
• Layer 6 - Competitive position (8 checks). Is the program competitive against 3-5 named peers?

Run the full pass quarterly. Run layers 1 and 2 monthly because authentication records and IP reputation drift weekly. Run layers 3 and 5 on every send. Re-run the whole thing after any ESP migration, IP change, or unexplained deliverability drop.

Layer 1 - Deliverability (8 checks)

If the message lands in spam, nothing else you do matters. Start here.

1. SPF, DKIM, and DMARC are all configured and aligned. DMARC policy is at minimum p=quarantine. See our walkthrough on SPF, DKIM, and DMARC for newsletters.
2. BIMI record published with a valid VMC or CMC. Gmail and Apple Mail render the logo only when the certificate validates. See the BIMI setup guide.
3. Sending IP warmed at least 30 days at current send volume. If you migrated ESP this quarter, you reset to day zero.
4. Google Postmaster Tools and Microsoft SNDS both show green or high reputation on all sending domains in the last 14 days.
5. Spam-trigger word density in the body is under 1.5%. Run the last send through a checker before you assume the subject line is the problem.
6. Image-to-text bytes under 60%. The Newsletrix corpus median is 41% across 12,000 sends. Image-to-text ratio details here.
7. RFC 8058 one-click List-Unsubscribe header present. Gmail and Yahoo have required this for bulk senders since February 2024 and they enforce it.
8. No URL shorteners (bit.ly, t.co) in the body. No redirect chains longer than one hop. Shortened links are a top contributor to phishing-pattern scoring.

Layer 2 - List health (6 checks)

A clean list is a deliverability feature, not a vanity metric. The numbers below are corpus medians, not industry quotes.

9. Hard bounce rate under 2% on the last send. Above 2% ESPs start throttling. See bounce rate benchmarks by vertical.
10. Soft bounce rate under 5%. Sustained soft bounces above 5% usually mean a reputation issue at the receiver, not a list issue.
11. Annual churn under 30%. MarketingCharts cohort data shows ~34% of new subscribers lapse in the first 30 days and 56% are still active at 12 months. Above 50% annual churn means the welcome sequence and re-engagement playbook need work.
12. A re-engagement segment is defined and gets a campaign at least quarterly. If you cannot point to the SQL or the segment definition right now, this check fails.
13. Double opt-in is on for organic signups. Single opt-in is acceptable for paid-list acquisition only if you have a confirm-or-suppress sequence inside 48 hours.
14. A sunset policy is documented. Anyone with zero opens in 90 days gets a final re-engagement attempt, then moves to suppression. See unsubscribe rate benchmarks for context.

Layer 3 - Content (10 checks)

Most operators want to start here. Resist. Fix layers 1 and 2 first or you are tuning a radio with the antenna unplugged.

15. Subject line between 30 and 50 characters. The 7 subject-line factors piece has the full breakdown.
16. Preheader is written by hand, not auto-pulled from the first line of body. Preheaders that read "View in browser" are a content fail.
17. Flesch reading ease score between 60 and 70 on the body copy. The readability sweet spot data backs this band.
18. One idea per email. If you cannot name the single takeaway in one sentence, the email is two emails.
19. The hook lands in the first 100 words. Gmail's preview pane cuts off near that point and Apple Mail truncates earlier on iPhone.
20. Scannable structure. No prose blocks longer than 4 sentences without a heading, list, or visual break.
21. Named author byline. "From the team" is not a byline. Real names get opens, anonymous senders do not.
22. Dateline present (a real date, in the body, near the top). It anchors the email in time for the reader and for AI assistants that summarise it later.
23. Zero broken merge tags. A single {{first_name}} in the wild kills trust for the whole send.
24. Alt text on every image. This is also a Layer 1 deliverability win because filters score alt attributes as additional text content.

Layer 4 - Render and design (7 checks)

What renders in your test inbox is not what renders on an old Android in dark mode.

25. Single-column mobile-first layout. Over 60% of opens happen on mobile across most B2C verticals, and the gap widens in B2B once you correct for desktop client overrepresentation.
26. Dark mode tested in Gmail iOS, Gmail Android, Apple Mail, and Outlook for Windows. Each renders dark mode differently and Outlook does not honour prefers-color-scheme at all.
27. Fallback fonts declared. Web fonts fail silently on Outlook desktop and on most corporate firewalls.
28. Primary CTA button is at least 44 pixels tall, which is the Apple Human Interface Guidelines minimum tap target.
29. Footer compliance. Physical mailing address (CAN-SPAM), an unmissable unsubscribe link, and a GDPR-friendly notice if any subscriber is in the EU. The footer compliance checklist covers the full surface.
30. Total HTML body under 102KB. Gmail clips above that and clipped emails lose the tracking pixel along with the footer.
31. Plain-text alternative present. Some receivers score multipart/alternative as a deliverability signal and most B2B inboxes still strip HTML by policy.

Layer 5 - Conversion (8 checks)

An email that lands in the inbox, looks right, and reads well still has to do something for the business. Most audits stop before this layer.

32. One primary CTA. If there are two, one is decoration and the data will say so.
33. Primary CTA appears above the fold on a 6.1-inch phone in portrait. Test it on a real device, not a Litmus screenshot.
34. Primary CTA repeats in the closing paragraph as a text link. The repeat lifts click rate without lifting design weight.
35. UTM parameters consistent across the send. utm_source, utm_medium, and utm_campaign spelled the same way every time.
36. Landing page promise matches the email promise. If the subject says "free template" the landing page hero says "free template" in the same words. Mismatch kills conversion before the page paints.
37. Click-through rate inside one standard deviation of your vertical benchmark. See open-rate benchmarks and click-to-open benchmarks for vertical-specific numbers.
38. Click-to-open rate above 8% for content sends, above 15% for transactional.
39. Downstream conversion event fires in your analytics. If you cannot trace from the click to the signup or purchase, you are flying blind.

Layer 6 - Competitive position (8 checks)

This is the layer every other audit checklist skips, and it is the one that produces the biggest single deltas. Auditing your newsletter without comparing it to peer senders is like reviewing a salary without knowing market rate.

40. 3-5 named competitors actively tracked. Not "the industry", named senders with real archives. The competitor analysis template has the worksheet.
41. Send-time overlap mapped. If three of your five peers send Tuesday 9am Eastern, you are competing for the same first impression. See the competitor send-time analysis for the method.
42. Subject-line pattern drift logged quarterly. Pattern shifts (numbers, questions, lowercase) propagate across a vertical in 90-day waves.
43. ESP migrations on the peer set noted with date and likely reason. Migrations are leading indicators of deliverability problems or budget shifts.
44. Sponsorship inventory benchmarked. Slot count, slot type, and rate-card position against the peer set.
45. Content gap matrix maintained. What topics do peers cover that you do not, and which of those are worth covering?
46. Peer reading time tracked. If three peers ship 6-minute reads and you ship a 14-minute read, you are competing on a different axis than you think.
47. Peer image-to-text ratio tracked. The vertical norm sets the deliverability ceiling for the whole category.

Compare against a single peer or "the industry" and this layer scores zero. Tools like Newsletrix and competitor-tracking alternatives such as Mailcharts make Layer 6 a 30-minute job instead of a manual archive scrape. The SWOT playbook covers the operating routine.

How often to audit and who owns it

Frequency by layer, with realistic time costs:

• Quarterly full audit (all 47 checks). 90 minutes the first time, 45 minutes once instrumented.
• Monthly deliverability and list-health pass (layers 1 and 2). 15 minutes.
• Per-send content and conversion pre-flight (layers 3 and 5). 5 minutes next to the send button.

Ownership matters more than frequency. The lifecycle marketer owns layers 1 through 5. Competitive intelligence (or whoever already runs the peer-tracking dashboard) owns layer 6. If both jobs land on the same person, layer 6 is the one that gets skipped because it sits outside send-cycle muscle memory. That is the tradeoff: keep layer 6 with someone who has time to look outside the inbox, even if they touch the rest less often.

Get the checklist and automate where you can

The 47 checks above are the full inventory. Copy them into a Notion page or a Google Sheet with three columns: check, threshold, last-passed date. That is the entire template. Anyone selling you a 200-row tracker is selling theatre.

Newsletrix automates the parts that scale poorly by hand. Layer 1 image-to-text and spam-trigger scoring runs in the spam-score checker. Layer 3 content checks (subject length, Flesch, hook position) run in the subject-line and hook tools. Layer 6 rolls into the SWOT report against the 3-5 senders you pick. Layers 2, 4, and parts of 5 still belong in your ESP and your analytics platform.

Frequently asked questions

What is a newsletter audit?

A newsletter audit is a structured pass across six layers of a newsletter program: deliverability, list health, content, render, conversion, and competitive position. It checks the program against a fixed specification, not against last quarter's results. The point is to catch the parts that are out of spec before they cost you opens, clicks, or subscribers.

How often should you audit a newsletter?

Run a full 47-point audit quarterly. Run the deliverability and list-health layers monthly because authentication records, IP reputation, and bounce rates drift weekly. Run the content and conversion layers on every send through a short pre-flight checklist. Re-run the full audit after any ESP migration, IP change, or unexplained deliverability drop.

What is included in an email marketing audit?

A complete audit covers six layers and 47 individual checks. Layer 1 deliverability has 8 checks (SPF, DKIM, DMARC, BIMI, IP reputation, image-to-text ratio, list-unsubscribe headers, link hygiene). Layer 2 list health has 6, Layer 3 content has 10, Layer 4 render has 7, Layer 5 conversion has 8, and Layer 6 competitive position has 8. Each check has a measurable pass or fail threshold.

How long does a newsletter audit take?

A first full audit runs 90 to 120 minutes for one program. Repeat audits drop to 45 to 60 minutes once you have the deliverability and list-health checks instrumented. The monthly deliverability and list-health pass takes about 15 minutes. The per-send content and conversion check takes 5 minutes if it lives next to the send button.

What is the difference between a newsletter audit and an A/B test?

An A/B test compares two variants of one element on one send. An audit checks whether the program meets a fixed specification across all layers, regardless of any single send's result. You run tests to find a better variant. You run audits to find out where the program is broken. They answer different questions and you need both.

Related reading