There was a time when putting "Hi, {Name}" in the subject line boosted open rates. That time is over. Today everyone does it, people notice, and a greeting with a first name doesn't impress anyone anymore — sometimes it even sounds like a template. The good news is that real personalization was never about the name. It's something deeper and far more useful: using what you know about each person so the content speaks to their situation.
Personalizing well isn't about dropping a variable into the greeting. It's about deciding what you show each person. Let's look at the levels of personalization, which data is actually worth collecting, and where the line sits between relevant and invasive.
The three levels of personalization
Not all personalization is created equal. Think of it as a ladder:
- Basic: the greeting and loose data points. "Hi, John." Using custom fields to drop in a name, a company, or a city. It's fine, it doesn't hurt, but it's no longer a differentiator. It's the floor, not the ceiling.
- By segment: the content changes based on who you are. This is where things start to move the needle. Someone who bought indoor plants sees indoor-plant care tips; someone who never bought anything sees a getting-started guide. Not one word changes — the whole message does. This is segmentation put to work.
- Dynamic: the content responds to behavior. The highest level: "The monstera you were looking at is back in stock." You use what the person actually did — what they viewed, what they bought, what stage they're at — to build a message that feels written for them. Not because it says their name, but because it understands their context.
The gap between levels 1 and 3 is huge. Look:
- Level 1: "Hi John, check out this week's deals."
- Level 3: "John, the monstera you left in your cart is back in stock — and this week it ships free."
The second one isn't "more personal" because of the name. It's more personal because it knows something real and uses it to actually help you.
Which data is worth collecting
It's not about gathering every data point you can — it's about the ones that let you be relevant. In order of power:
- Behavior. What someone opened, clicked, bought, left in a cart. It's the best predictor of what they want right now.
- Purchase history. What they bought before tells you what to offer next (repurchase, add-on, next tier up).
- Declared preferences. What the person told you they're interested in when they subscribed. It's no accident that, according to a Sinch survey cited by eMarketer, people prefer personalization based on their stated preferences and history over personalization based on their name.
- Context. Location, lifecycle stage, recency. Useful as a filter, not as the star of the show.
The practical rule: in your forms, ask only for the data you're actually going to use. Every extra field lowers your signup rate, and a data point you never act on is just friction.
The line you don't want to cross
There's a point where personalization stops being useful and starts being creepy. "We noticed you were looking at this product at 11:14 PM from your house" is technically personalized and deeply uncomfortable. The goal is for the person to feel understood, not watched.
The guiding principle is simple: use data to be relevant, not to show off how much you know. "The thing you liked is back" adds value; "we know everything about you" scares people away. When in doubt, err on the side of discretion.
What changed in 2026: your edge is first-party data
Here's a new angle worth leaning into. In an inbox increasingly crowded with generic, AI-generated content — messages that are correct but impersonal, all starting to look alike — what stands out is exactly what someone else's AI can't know: the behavior and history of your customers with your brand. That's first-party data, yours alone, that no model trained on the open internet has access to.
And AI, on your side of the equation, makes personalizing at scale stop being a luxury. Amanda IA can build campaign variants by segment, adapt tone to your Brand Kit, and help you decide what to show each group — the work that used to make personalization unworkable once you had thousands of contacts. The combination is powerful: your first-party data gives the AI the raw material that makes it genuinely relevant.
How arrobaMail approaches it
In arrobaMail you have custom fields for your data, behavioral segmentation (opens, clicks, purchases) to change content based on who's receiving it, and Amanda to produce those variants without eating up your whole day. It's all connected to automation, so personalization happens on its own the moment someone enters a segment.
The mindset shift that matters: stop thinking of personalization as "dropping in the name" and start thinking of it as "showing each person what's useful to them." That's the kind people actually notice — and the kind that converts.
Want to try it? Create a free account and build two versions of your next campaign: one for people who've already bought from you, and one for people who haven't. You'll see the difference in your clicks.