Review open and click rates monthly

Put the review on the same day every month, for example the third working day. Keep it short. If it takes two hours, it will be skipped by March.

You need one person who owns it, one sheet and one list of the sequences that are live. Names matter here. Without an owner, the review is everyone's job and so nobody's.

Export the numbers from your email tool, such as HubSpot, Customer.io or Mailchimp, into a sheet in Google Sheets. One row per email, with these columns:

• Sequence and position.
• Emails sent in the last 90 days.
• Delivered rate.
• Open rate.
• Click rate, as a share of delivered emails.
• Reply rate.
• Unsubscribe and spam complaint rate.
• Outcome: how many people booked or took the next step after this email.

The last column is the one most teams skip. It is also the one that matters, because an email with a high click rate and no outcomes is a distraction.

Opens have been unreliable since Apple Mail Privacy Protection started loading emails automatically. Use them to compare an email with its own history and with its neighbours in the sequence, not as a score. Click rate, reply rate and outcomes are the numbers I would trust.

Do not judge an email that went to fewer than about 200 people in the last 90 days. A rate on a small base jumps around for no reason. Mark the row "too early" and look again next month.

Industry benchmarks tell you little about your list. Compare each email with the median of its own sequence instead. Flag any email whose click rate is less than half that median. Flag any step where the number of people drops sharply compared with the step before.

This gives you a short list of problem emails, usually two to four. That is the right size for one month.

Each pattern points to a different cause:

• Low opens, normal clicks among openers: the subject line, the sender name or the delivery. Check inbox placement and your sender health with a tool like MailReach.
• Good opens, low clicks: the body or the call to action. The email may be too long, or the ask unclear.
• Good clicks, no outcomes: the page behind the link. Check it loads and says what the email promised.
• Rising unsubscribes: the email is too frequent, too salesy or sent to the wrong segment.

Write the diagnosis in the sheet. It stops you changing the subject line when the real problem is the landing page.

For each flagged email, choose one action:

1. Leave it, if the sample is small or the dip is a one-off.
2. Rewrite it, if the idea is right but the execution is weak.
3. Replace it with new content, using material from Add new content to sequences.
4. Remove it, if the sequence works as well without it.

Change one thing per email, so you can see what moved the number. If you want to test subject lines or send times in a structured way, use Test subject lines and send times.

Write the date and the change next to the email. Next month, compare the new numbers with the old. If you changed three emails and two improved, keep going. If none improved, your diagnosis was wrong, and that is useful to know.

Add UTM parameters to the links so Google Analytics shows what visitors from each email do on your site. It gives you a view past the click.

• Chasing open rates and ignoring outcomes.
• Judging emails with small samples.
• Comparing your numbers with a published industry average.
• Rewriting the whole sequence at once.
• Running the review weekly and reacting to noise.

After three months, you should have a log with at least six changes and their results. The weakest email in each sequence should be better than it was, and the click rate and outcomes across the sequence should be steady or rising. Most telling of all, you should be able to say in two minutes which email in each sequence you would cut first, and why.