Last updated October 2026 by Lily Newman
Personalization that works uses what a customer did, not who a data set says they are. The strongest versions leverage information about how someone engaged with your brand, what (or if) they purchased, how long ago they interacted, and what people like them usually do next.
At this point customers just expect it. McKinsey's 2021 research found that 71% of consumers expect companies to deliver personalized interactions and 76% get frustrated when brands don’t deliver (McKinsey).
Four levels of personalization
There are 4 basic buckets that personalization lives under, each requiring different types of data to effectively power.
| Level | What it uses | Example | Data you need |
|---|---|---|---|
| 1. Merge fields | Name, location | "Hi Sam" | Signup form |
| 2. Behavioral triggers | What the customer just did | Abandoned checkout showing the items left behind | Store and site events |
| 3. Purchase history and stage | What they bought and when | A care guide for the product they bought, a reorder reminder | Order data |
| 4. Predictive | Modeled next steps | A message timed to a predicted next-order date | Enough order history for the model |
Most of the revenue sits in levels 2 and 3. Level 4 needs volume: email providers like Klaviyo have native predictive features which require at least 500 customers with orders and 180 days of history (Klaviyo).
What it looks like in each flow
| Flow | Generic version | Personalized version |
|---|---|---|
| Welcome | The same offer for every subscriber | Content and products based on the signup source or a quiz answer |
| Abandoned checkout | "You forgot something" | The items left behind, with any shipping or stock detail that matters |
| Post-purchase | A thank-you | How to use or care for the product they bought |
| Replenishment | A calendar reminder | Timed to when that product usually runs out |
| Winback | A blanket discount | The category they bought, with a discount only where margin allows |
Collect data the right way
Personalization is only as good as the events you capture. Before you build a dynamic block, check that the foundation is in place.
Store and email platform connected natively. Use the integration for your store (Shopify, BigCommerce, etc) so orders, products, refunds, and customer records sync automatically. Confirm that product IDs, categories, and images come through, not just order totals. Level 3 personalization depends on them.
Onsite tracking installed and tested. Viewed product, added to cart, and started checkout should each fire as events with the item, price, image, and URL attached. Test with your own email and confirm the events show up in the profile’s activity feed.
Visitors tied to a person. An event is only usable once it is attached to a profile, usually when someone enters an email or phone number in a form or at checkout, logs in, or clicks through from one of your emails. Anonymous browsing does nothing for you until then, so make identification points easy to find. In a world where cookies are becoming less common, consider bringing in a 3rd party provider like customers.io, LittleData, or Audiense to help.
Key events sent server-side. Browsers and ad blockers cut off client-side tracking, and Safari limits how long cookies set by scripts survive (WebKit). Purchase and checkout events should come from your store backend or the platform’s native integration, not only from a browser tag.
Consistent names and properties. Pick one naming convention for events and properties, and record the signup source (form, popup, checkout, or UTM) on every profile. Without that, segments can quietly break.
Consent stored with the profile. Keep the email and SMS consent status, timestamp, and source on each record so you can suppress correctly and show where consent came from.
Once the tracking is solid, ask customers for what it cannot see. A short quiz, a preference center, or one question at signup gives you data they chose to share. Save the answers as profile properties so you can segment on them. Ask for one thing at a time, and use only what the customer would expect you to know: what they bought, what they asked for, what they clicked.
Privacy rules differ by state and country, so have counsel review how you collect and use data.
Where personalization goes wrong
Thin or wrong data. A gift purchase tells you little about the recipient, and a one-time purchase may not be a real preference.
Segments too small to read. If a segment has a few dozen people, you cannot tell whether the message worked.
Too clever. If the customer would be surprised how you know something, drop it. There’s a fine line between engaging and creepy.
Breaking the brand voice. Dynamic content still has to sound like you.
How to test it
Send the personalized version to most of the group and the generic one to a small random holdout. Compare revenue per recipient and unsubscribe rate over at least two send cycles.
Frequently asked questions
Does personalization increase revenue? McKinsey has estimated that personalization can lift revenue by 5% to 15%, but that is a 2016 estimate for "can," not a promise. Test it on your own list to establish what baseline looks like for you.
Is putting a name in the subject line worth it? Test it. It’s one of the easiest forms of personalization to execute.
Do I need a special tool? Not to start. Your email platform should handle levels 1 to 3. Predictive features come built into some platforms (but not all).
What if I have little data? Start with behavioral triggers (abandoned checkout, post-purchase) and ask customers what they want.
Want thoughtful personalization built around your audience and business? Talk to Newly Marketing.
Sources
About the Author
Lily Newman is the Founder and Principal of Newly Marketing, a digital marketing consultancy specializing in lifecycle, email, mobile, CRM, and retention marketing. A fractional DTC e-commerce marketing leader, Lily advises high-growth startups and Fortune 500 brands in retail, CPG, health and wellness, and technology, drawing on 15+ years leading acquisition and retention for Levi's, Peet's Coffee, Mighty Leaf Tea, Four Sigmatic Foods, and more.
