Email Automation, Segments and Deliverability
Split your list by behaviour, compare revenue per email, plan an abandoned-basket flow, and understand the sender rules that decide whether email reaches the inbox, with a worked example.
In this lesson you will
- Segment a list by behaviour
- Compare revenue per email
- Plan an automation flow
- Know the main sender requirements
About 25 minutes
One email to everyone
Noor sends the same newsletter to all 1,000 people on her list. Some of them bought last week, some joined yesterday and some have not opened anything in months. She suspects one message cannot suit all of them. She is right. This lesson covers two ways to make email work harder: segmentation, which means sending different messages to different groups, and automation, which means sending messages triggered by what people do. It also covers the technical rules that decide whether your email reaches the inbox at all.
Segments by behaviour
The easiest and most useful segments are based on what people have done: joined recently, bought, bought more than once, left something in a basket, or gone quiet. Behaviour tells you more than a guess about a person’s tastes. The program below compares three made-up segments. For each it works out the click rate and the revenue per email sent, which is a fairer measure than clicks alone, because it counts what the email earned.
# MADE-UP results by segment: (emails sent, clicks, orders, revenue in pounds)
seg = {"new subscribers": (400, 28, 9, 162.0), "buyers": (300, 36, 14, 266.0), "lapsed 60+ days": (300, 6, 1, 18.0)}
total_sent = sum(v[0] for v in seg.values())
for name, (sent, clicks, orders, rev) in seg.items():
print(f"{name}:")
print(f" click rate {100 * clicks / sent:.1f}%")
print(f" revenue per email {rev / sent:.2f}")
tot_rev = sum(v[3] for v in seg.values())
print("all together:", f"{tot_rev / total_sent:.2f}", "per email")
best = max(seg, key=lambda k: seg[k][3] / seg[k][0])
print("best segment:", best)
# sender set-up check, MADE-UP flags
setup = {"SPF": True, "DKIM": True, "DMARC": False, "one-click unsubscribe": False}
spam_rate, daily = 0.12, 800
print("missing:", ", ".join(k for k, v in setup.items() if not v))
print("spam rate under 0.3%:", spam_rate < 0.3)
print("above 5,000 a day:", daily > 5000)
# abandoned-basket flow: hours after the basket was left
for step, h in enumerate([1, 24, 72], 1):
print(f"email {step}: after {h} h = day {h // 24} hour {h % 24}")
new subscribers:
click rate 7.0%
revenue per email 0.41
buyers:
click rate 12.0%
revenue per email 0.89
lapsed 60+ days:
click rate 2.0%
revenue per email 0.06
all together: 0.45 per email
best segment: buyers
missing: DMARC, one-click unsubscribe
spam rate under 0.3%: True
above 5,000 a day: False
email 1: after 1 h = day 0 hour 1
email 2: after 24 h = day 1 hour 0
email 3: after 72 h = day 3 hour 0
Buyers click far more, at 12.0 per cent, and earn 0.89 pounds per email sent. New subscribers earn 0.41 and the lapsed segment only 0.06. Across everyone, the average is 0.45 per email, which hides these big differences. That is the case for segments: the buyers want new scents and early access, the new subscribers need a welcome and some trust, and the lapsed segment needs either a reason to return or a polite goodbye. Sending the same message to all three wastes the best segment’s attention and annoys the least engaged.
Automation flows
A flow is a short series of emails that starts when someone does something. The last lines of the output show an abandoned-basket flow: a first nudge after one hour, a second after a day and a third after three days. Each should add something, such as an answer to a common worry, a review or a delivery reminder, and the last may offer a small incentive. Stop the flow as soon as the person buys. Also limit how often anyone can be emailed.
Will it reach the inbox?
Mailbox providers have tightened their rules. Google and Yahoo published requirements, in force since 2024, for people who send in bulk, which they define as about 5,000 or more messages a day to their users: authenticate your domain with SPF, DKIM and DMARC, make marketing emails easy to leave with one-click unsubscribe, honour opt-outs within two days, and keep spam complaints below 0.3 per cent. Microsoft announced similar rules for Outlook addresses from 2025. Small senders are not exempt from the sense of it: authentication and easy unsubscribing help everyone. In the program, Noor’s setup is missing: DMARC, one-click unsubscribe. Her spam rate is under the limit and she is far below the bulk threshold, but it is cheap to fix the gaps now. Ask your email tool’s help pages how to set up each item for your domain, and check each provider’s current page, because details change.
- Split your list by behaviour into at least three groups.
- Write a welcome flow, a basket flow and a win-back message.
- Compare revenue per email for each, not just clicks.
- Set up SPF, DKIM and DMARC, and a one-click unsubscribe, through your email tool.
- Remove people who have not engaged for a long time, after asking once if they want to stay.
Keeping everyone on the list forever. People who never engage drag down your results and can push you towards the spam folder. A smaller, interested list is worth more than a large, sleepy one.
Rules on consent and unsubscribing from lesson 6 still apply to every automated email. This course is not legal advice, and providers’ requirements change, so check the official pages.
Practice for this lesson
Check what you learned
Five quick questions. Pick an answer and you will see straight away whether it is right and why.
What is segmentation?
- Deleting the list
- Sending more emails
- Sending different messages to different groups
- Changing the logo
Segments let each group get a message that fits them.
Which measure is fairer than clicks alone?
- Revenue per email sent
- List size
- Subject length
- Image count
It counts what each email earned.
When should an abandoned-basket flow stop?
- After a year
- As soon as the person buys
- Never
- After the first click
Continuing after a purchase annoys people.
What do bulk-sender rules ask marketers to set up?
- More images
- Bigger lists
- Longer subjects
- SPF, DKIM, DMARC and easy unsubscribe
Providers require authentication and easy opt-out.
Which segment earned the most per email in the sample?
- Lapsed
- Buyers
- New subscribers
- All equal
Buyers earned the most per email sent.
Summary
- Segments are best made from what people have done.
- Compare revenue per email, not just clicks.
- A flow starts from an action and stops when the goal is met.
- Authenticate your domain and make unsubscribing easy.
Questions
Do the bulk sender rules apply to me if I send fewer emails?
The formal thresholds apply to large senders, but authentication and easy unsubscribing help any sender reach the inbox. Check each provider's current page for details.
Are the segment results real?
No. They are made-up samples to show the method.
Who sets up SPF, DKIM and DMARC?
Usually your email tool and your domain provider give you steps. They involve adding records to your domain settings. Follow their current help pages.
