Email and customers · Lesson 5.3
Email Automation Sequence Metrics That Actually Matter
After you switch on an email automation sequence, your tool will show you a dashboard full of numbers. Most of them are easy to read and not very useful. A few are worth checking every week. This post tells you which, how to read them and what to do when one looks wrong.
I am not giving benchmark numbers. Averages from other businesses depend on the industry, the list and how the figure was measured, so they make poor targets. Compare your sequence to its own past.
Lesson 5.3 includes the last step, reading the results and fixing the weakest email: An email journey that keeps paying.
Track the outcome (sales), the action that leads to it (clicks and replies) and list health (unsubscribes and spam complaints), and treat open rates as a weak hint.
Why activity numbers fool you
Sent counts, delivered counts and open rates are activity measures. They tell you the machine ran. They do not tell you whether anyone wanted what you sent.
Open rates in particular have a known problem. Mailchimp's FAQ on Apple Mail Privacy Protection says that with it enabled, emails in Apple Mail are reported as opened regardless of the contact's activity, which results in inflated and inaccurate open rates. Google, for its part, says it does not track open rates and cannot verify the accuracy of open rates reported by third parties. So treat opens as a rough sign at best.
The four numbers that matter
1. Sales from the sequence
The point of the sequence is the goal you set, such as a first order or a booked call. Count how many people who went through the sequence reached it. If your tool cannot attribute sales, a simple way is to ask at checkout or booking, "Where did you hear about us?" and include an option for your email list. It is imperfect but it is something.
2. Clicks on the offer link
Clicks show action. Mailchimp defines the click rate in its open and click rates help page; read that page to see exactly how your tool counts, since definitions vary between tools. Look at clicks on the offer email and the nudge. If the offer gets clicks but there are no sales, the problem is probably the page the link leads to, not the email.
3. Replies
A reply is the strongest sign in email. If you ended a Help email with a question, count the answers. Replies are rarely captured in dashboards, so keep a simple tally in a notes file.
4. List health
Watch unsubscribes per email and spam complaints. A spike after one email tells you that email missed. For Gmail, Google's sender guidelines say to keep the spam rate reported in Postmaster Tools below 0.10% and to avoid ever reaching 0.30%. Postmaster Tools is mostly useful if you send in volume, so a small sender may not see much data there. Even so, the principle holds: complaints are a number worth keeping low.
A weekly reading routine
- Open your report at the same time each weekPick a fixed slot, such as Monday morning, so you compare like with like.
- Write down five figuresNew joins, clicks on the offer link, replies, sales, unsubscribes.
- Find the email with the biggest dropCompare each email's clicks and unsubscribes with the one before. The email where people disappear is your weakest.
- Pick one changeOne subject line, one opening, one link, one call to action.
- Log it and waitNote the date and the change. Do not touch the sequence again until you have enough new people to judge.
How to read a drop off
Imagine, as an illustration only, that you have five emails and these made up click counts: 40, 30, 28, 12, 10. The big fall happens at email four, the offer. That suggests people were engaged until the ask. Possible causes: the offer is unclear, the link is buried or the price is surprising. A small fall from 40 to 30 between emails one and two is normal, since some people read the freebie and drift.
Now imagine 40, 8, 7, 6, 5. The fall at email two points to the Help email, or to the gap before it, or to a subject line that did not earn the click. These numbers are invented to show how to read the pattern, not to suggest targets.
What to measure and what to skip
| Number | Useful? | Why |
|---|---|---|
| Sales from the sequence | Yes | It is the goal |
| Clicks on the offer link | Yes | Shows action |
| Replies | Yes | Shows real interest |
| Unsubscribes per email | Yes | Finds the email that misses |
| Spam complaints | Yes | Protects delivery |
| Open rate | Weak | Inflated by Apple Mail Privacy Protection |
| Emails sent | No | Shows only that the tool ran |
| List size alone | No | Big and silent is not an asset |
How long to wait before judging
Small lists produce noisy numbers. If only ten people have gone through, one person changes a percentage dramatically. Wait until you have a reasonable number of people through the whole sequence, and look for big differences rather than small ones. There is no magic number. Use your judgment, and favour patience over reacting to every dip.
Keep a simple log
A notes file with these columns is enough: date, change made, people through since, clicks, replies, sales, unsubscribes. After a few months you will know what moves your numbers, and that knowledge is worth more than any average.
See our post on common sequence mistakes for what to look for when a number looks off.
Questions people ask
Are open rates useless?
Not useless, but unreliable. With Apple Mail Privacy Protection, Apple Mail opens show as opened whatever the reader did. Use opens only as a loose signal for subject lines.
What is a good click rate?
It depends on your list and how your tool counts. Compare against your own earlier results rather than a number from an article.
My tool does not track sales. What now?
Ask new customers how they found you, or use a unique link or code in the offer email, and tally by hand.
Where to go next
The last part of Lesson 5.3, An email journey that keeps paying, shows you how to read the results and fix the weakest email. If you need the whole sequence first, see the complete guide.