You are sending the same email to everyone on your list
And that is the problem. Half your subscribers opened your last campaign. The other half ignored it. Not because your subject line was bad. Because half of them got an email about a product they already bought, and the other half got an email about a product they would never buy.
AI email personalization is not about putting a first name in the subject line. That stopped working in 2019. Real personalization means sending different content to different people based on what they have actually done. What they bought. What they browsed. What they ignored.
And the tools to do this are already inside the email platform you are using. You just have not turned them on yet.
What AI email personalization actually does
When most people hear AI email personalization, they think of a chatbot writing emails. That is not what this is. This is about using the data your email platform already collects (purchase history, browse behaviour, email engagement, demographic signals) to automatically sort your audience into segments and deliver different content to each one.
Here is what that looks like in practice:
- A customer who bought a face cleanser two weeks ago gets an email about the matching moisturiser. Not a generic sale announcement.
- A subscriber who clicked on candles three times but never bought gets a 10% off candles subject line. Not a newsletter about new arrivals.
- A customer who has not opened an email in 60 days gets a re-engagement subject line. Not the same broadcast everyone else received.
Same campaign. Three different versions. Three different results.
The three personalization levers you can pull today
Lever 1: Purchase-based segments
This is the most powerful personalization signal because it is based on actual money spent, not vague intent.
Set up three segments in your email platform:
- First-time buyers (purchased once, 0 to 30 days ago): Send a post-purchase series that thanks them, shows complementary products, and asks for a review. This is your highest-value segment because first-time buyers who buy a second time within 30 days are 3 times more likely to become long-term customers.
- Repeat buyers (purchased 2 or more times): Send early access to new products and loyalty rewards. These customers already trust you. Give them a reason to stay.
- Lapsed buyers (last purchase 90+ days ago): Send a win-back campaign with a specific product recommendation based on their last purchase category, not a generic discount.
Most email platforms let you create these segments with drag-and-drop rules. No code needed.
Lever 2: Behaviour-based content blocks
Instead of writing three separate emails, write one email with dynamic content blocks that change based on who is reading it.
Here is how that works:
- The header stays the same for everyone.
- The product recommendation block changes based on what the subscriber has browsed or bought recently.
- The call-to-action changes based on their purchase frequency.
For example, a homewares store sends a weekly email. The header is the same for all 5,000 subscribers. But the featured product section shows candles to people who browsed candles, planters to people who browsed planters, and a bestseller to people who have not browsed anything recently.
This is not science fiction. Klaviyo calls them dynamic blocks. Mailchimp calls them conditional content. Omnisend calls them smart products. All three platforms have had this feature for years.
Lever 3: Send-time optimisation
AI personalization is not just about what you send. It is about when you send it.
Most small businesses send their newsletter on Tuesday at 10am because a blog post told them that was the best time. But your audience is not everyone. Your audience is a specific group of people with specific habits.
Send-time optimisation uses your subscribers’ past open times to calculate the best delivery window for each person. Klaviyo and Mailchimp both have this built in. You click a toggle that says send at the best time for each recipient and the platform figures out the rest.
The impact is not subtle. Brands that switch from fixed send times to AI-optimised send times typically see a 15 to 25% increase in open rates within the first month. That is not because the email got better. It is because it arrived when the person was actually looking at their inbox.
Setting it up: the 30-minute version
If you have Klaviyo, Mailchimp, or Omnisend, here is the fastest path to personalised emails:
- Create your three purchase-based segments (first-time, repeat, lapsed). This takes 10 minutes with drag-and-drop filters.
- Turn on send-time optimisation for your next weekly campaign. This is one click in most platforms.
- Add one dynamic product block to your next newsletter. Set it to show the most-browsed category for each subscriber. This takes 15 minutes to set up.
Do those three things and your next campaign will outperform your last one. Not because the content got better. Because the right people got the right message at the right time.
What to measure
After your first personalised campaign, compare these metrics against your last unsegmented broadcast:
- Click-through rate by segment. First-time buyers should click at a higher rate than your average. If they do not, the product recommendations are not relevant enough.
- Revenue per recipient. This is the number that matters most. A personalised campaign that generates $0.15 per recipient is better than a generic one that generates $0.08.
- Unsubscribe rate. If your personalised campaigns have a higher unsubscribe rate than your generic ones, the personalization is off. People can tell when an email is creepy rather than helpful.
Give it three campaigns before you judge. The first one will always underperform because your segments are new and your data is thin. By campaign three, you should see a clear lift in revenue per recipient.
The one thing to avoid
Do not over-personalise. Showing someone the exact product they browsed yesterday feels helpful. Showing them every product they have ever looked at in a single email feels invasive. The line between this brand knows me and this brand is watching me is thinner than you think.
Stick to one personalised element per email. A product recommendation. A send time. A subject line variant. Not all three at once. One change per campaign is enough to see results without crossing into uncomfortable territory.
Want more like this? Join MOKU Club for free. Weekly resources, early access to new guides, and occasional templates you can actually use. Join below.
