"Hi {FirstName}" was personalization once. Now it's just the baseline every inbox expects, and on its own it does nothing to make an email more relevant. Genuine personalization comes from what a contact has actually done — what they've browsed, bought, clicked, and ignored — and from adapting the message and its timing to that behaviour rather than to a static field in a contact record.
Merge tags are easy to set up and easy to spot. Recipients have learned to see through them, and a first-name greeting on an otherwise generic broadcast doesn't change whether the content is relevant to them. Personalization that actually moves the needle changes what's in the email, not just the greeting at the top.
A behavioural trigger sends an email in response to something a contact did, rather than on a fixed schedule. This is the difference between a monthly newsletter that goes to everyone regardless of relevance, and an email that only goes out because a specific person did a specific thing.
Behaviours worth triggering on
Start with one trigger, not ten
It's tempting to build every behavioural automation at once. Start with the single highest-value trigger for your business — usually browse abandonment or a post-purchase follow-up — get it working well, then expand.
If you know what someone bought, you know a great deal about what to send them next. A customer who bought running shoes is a far better candidate for an email about running socks than your entire list is. This kind of recommendation doesn't require complex machine learning — simple rules based on product category, purchase date and typical reorder cycle cover most of the value.
A simple purchase-history segmentation approach
Group products into a small number of categories
You don't need granular SKU-level rules. Broad categories (skincare, running gear, coffee) are usually enough to make recommendation emails feel relevant.
Set a reorder or complement window per category
Consumable products suit a reorder reminder at a sensible interval; durable products suit a complementary product suggestion instead.
Exclude anyone who's already bought the recommended item
Nothing undermines personalization faster than recommending something a customer already owns.
The same email sent at 6am and at 6pm can get meaningfully different open rates, and the best time varies by contact, not just by industry. Send-time optimization looks at when each individual contact has historically opened email and times future sends to match, rather than sending your entire list at whatever hour is convenient for you.
None of this works without decent segmentation. Behavioural triggers, purchase-based recommendations and send-time optimization all depend on knowing enough about a contact to act on it. In Havari, contact activity across email, SMS and WhatsApp feeds into the same profile, so a segment can be built on real behaviour — opened the last three emails, clicked a specific link, replied on WhatsApp — rather than a static list someone manually tagged months ago.
“Personalization isn't the name at the top of the email. It's whether the content would be different for someone else.”
Is send-time optimization worth setting up for a small list?
It's most valuable once you have enough send history per contact to detect a pattern. For a very small or new list, focus on behavioural triggers first and revisit send-time optimization once you have more data.
How much personalization is too much?
If a recommendation feels surprising in a good way, it's working. If it feels like being watched, it's gone too far — stick to what a reasonable customer would expect a business to know.
Do I need separate tools for email personalization and SMS or WhatsApp?
Not necessarily. Platforms like Havari keep one contact profile across channels, so behavioural data gathered from an email click can inform an SMS or WhatsApp message later, and vice versa.
Real personalization is quieter than a first-name merge tag and does more work — it's the difference between an email that happens to be addressed to someone, and one that was actually built around what they've done.