Guides7 min read

The 10-minute weekly health check

Five items every Monday morning. Ten minutes. Catches most problems two weeks before they become fires.

The quarterly review catches drift. The annual auth renewal catches staleness. The weekly check is different: it catches the surprises. An ESP decided to rate-limit you overnight. A vendor started sending from your domain without telling you. Gmail changed their filter weights. Someone in sales imported a thousand purchased leads. You don't need a full audit to catch these — just ten disciplined minutes every Monday.

The ritual

Monday morning, coffee in hand, before Slack. Five items. Ten minutes. Same five things, same order, every week. The magic is the repetition — anomalies jump out because you know what normal looks like.

Minutes 1–2: Placement spot-check

Run a single inbox placement test from your main marketing domain. One test. Look at the Primary / Promotions / Spam split. Write the numbers next to last week's. If Primary dropped by more than 5pp, note it for item 5. If everything looks normal, move on.

Over time you learn what normal looks like for Monday morning on your domain. Some natural variance is fine. A consistent drop over 2–3 weeks isn't.

Minutes 3–4: Postmaster dashboard

Open Google Postmaster Tools. Look at four tiles: domain reputation, IP reputation, spam rate, authentication rate. You're not analysing — you're scanning for change.

  • Reputation dropped a tier (High to Medium, Medium to Low) — investigate
  • Spam rate crossed 0.2% — trending up, noise or problem?
  • Spam rate crossed 0.3% — action required this week
  • Auth rate under 99% — something is sending without DKIM/SPF

If nothing changed, close the tab and move on. If something did, note it for item 5.

Minutes 5–6: ESP dashboard numbers

Open your ESP. Look at the last 7 days:

  • Sent volume — reasonable compared to last week, or a big jump?
  • Bounce rate — under 2% still?
  • Complaint rate — under 0.1% still?
  • Unsubscribe rate — trending or stable?
  • Any error-state sends — rate-limited, deferred, blocked?

You're looking for anomalies, not perfection. A big volume jump last week might be legitimate (a launch campaign) or might be an import gone wrong. Know which.

Minutes 7–8: Support signals

Open your support inbox or ticketing tool. Search the last seven days for three phrases: "didn't receive", "in spam", "can't find the email". Count them. If the count is more than a typical week, you have a deliverability problem showing up in support before it shows up in dashboards.

Support tickets lead metrics by 3–7 days. A sudden increase in "didn't get the password reset" tickets is usually the first sign of a transactional problem. Pay attention.

Minutes 9–10: Weekly log entry

In a single shared doc titled "Email weekly log", write three lines:

  1. Date + one-line summary: green / yellow / red
  2. The placement and reputation numbers from items 1 and 2
  3. Any anomaly worth tracking, with a suggested owner

That's it. 20 weeks later, the doc is a chronological record of how your program evolved, which is invaluable when something goes wrong and someone asks "when did this start?"

Red weeks: what to do

A red week is any week where one of these triggers:

  • Placement drops 10pp+ week over week
  • Spam rate crosses 0.3%
  • Domain reputation drops a tier
  • Support ticket count for delivery issues doubles
  • New blocklist listing
  • Bounce rate crosses 3%

On a red week, the 10-minute ritual expands into a 60-minute investigation. The point of the weekly check is to know the difference between a normal Monday and an investigation Monday without having to think about it.

What you can automate

Most of this can be pushed to Slack with automation, and should be once the ritual becomes routine:

  • Weekly placement test via the Inbox Check API, output to Slack
  • Postmaster numbers via the Google Postmaster API
  • ESP metrics via webhook
  • Support ticket search via your helpdesk's search API

Automated delivery is better than manual, but the human ritual of looking at the numbers and writing the log still matters. Dashboards go unread. A 10-minute Monday habit doesn't.

Inbox Check

Wire the Monday placement check into an API call that posts results to Slack. Free tier covers weekly cadence; upgrade if you want daily. Or just run the free test in a tab and call it good.

Who owns the ritual

Whoever cares most about deliverability. On a small team, the marketing lead. On a growth team, the email owner. On a product-led team, the onboarding engineer. Not an agency — agencies bill for the time, and the point of the ritual is to be free and habitual. If it becomes a billable task, it stops being a habit.

FAQ

Is weekly too often?

For a stable, well-run program under 50k sends/month, bi-weekly is fine. For anything above that, or anything with dedicated IPs, weekly is the right cadence.

What if I miss a week?

Skip it and resume next Monday. Don't try to 'catch up' two weeks of history. The value is consistency, not completeness.

Should I do this on Monday specifically?

Monday works because the weekend lull gives you stable baselines and the week's campaigns haven't run yet. Tuesday is a reasonable alternative. Avoid Friday — the weekend distorts what you see.

How long until the pattern of normal emerges?

Four to six weeks. Before that every signal looks like an anomaly. After six weeks, real anomalies jump out and normal volatility fades into the background.
Related reading

Check your deliverability across 20+ providers

Gmail, Outlook, Yahoo, Mail.ru, Yandex, GMX, ProtonMail and more. Real inbox screenshots, SPF/DKIM/DMARC, spam engine verdicts. Free, no signup.

Run Free Test →

Unlimited tests · 20+ seed mailboxes · Live results · No account required