Crawl, Walk, Run: How to Add Personalization into Marketing Communications

by | Jun 8, 2026 | Email Marketing | 0 comments

Marketing personalization is one of the highest-ROI moves in marketing, and most brands are underutilizing the opportunity. The question isn’t whether to personalize. It’s what’s next.

The stats on why this matters are overwhelming. 71% of consumers expect personalized experiences and 76% get frustrated when brands don’t deliver them (McKinsey). 80% of consumers are more likely to buy from a company offering personalized experiences (Epsilon). 62% say a brand will lose their loyalty if the experience feels generic (Twilio). And fast-growing companies earn 40% more of their revenue from personalization than slower competitors (McKinsey). Marketing personalization isn’t a bonus anymore, it’s the baseline expectation that customers are comparing your brand against. It taps into the same behavioral economics that drives every other buying decision.

To get personalization right, it’s all about the data. There are three main types of data: zero-party, first-party, and third-party. 

Zero-party data is what customers voluntarily share through forms, surveys, quizzes, and preference centers. It’s gold because customers gave it to you on purpose, and they expect you to use it.

First-party data is collected from how customers interact with your brand. Browsing behavior, purchases, email engagement, account info, etc. It’s the largest data set most brands have, and the most underutilized.

Third-party data is purchased or licensed from outside sources, like data brokers or ad platforms. It’s how brands target people who aren’t in their database yet, but it’s also getting harder to access as privacy regulations tighten and cookies disappear.

All three can fuel personalization, but the most reliable foundation is the first two. They’re collected directly from customers, more accurate, and you own them.

Think of marketing personalization as crawl, walk, run. Three ways to turn data into results.

Crawl: Foundational Marketing Personalization

Crawl is straightforward, and you’re probably already doing it. Segmentation is the simplest type of personalization based on zero- and first-party data you probably already have about everyone: name, email, location, purchase history, etc. That’s enough data for foundational personalization too.

Crawl doesn’t require new data infrastructure, just dynamic text fields that every major platform supports out of the box. Pair this with strong email design best practices and you have a foundation that performs. First name in a subject line. A reference to a recent purchase or category browsed. Body copy that acknowledges they’re a loyal customer, so the email feels like it was written for them. A gentle reminder about items left in their cart. The same first name, location, and purchase data can fuel personalized SMS, push notifications, and even printed inserts in direct mail. 

Crawl works because the bar is low. Most marketing emails feel generic, written for no one in particular. “Hi there, check out our latest arrivals…” with some colorful emojis. This easily becomes “Hi Blake, those running shoes you’ve been eyeing are back in stock…” I know which one I’m responding to.

E.g. take your next promo email and add first name to the subject line and a recent browse or purchase category reference in the opening line. Leave it generic for customers you don’t have data for, it’s not going to hurt. Run this personalized version as an A/B test against the standard version. Measure open and click rate. If you see lift, you have proof that foundational personalization works, and you can roll the same approach into your SMS and push programs using the exact same data fields.

Walk: Contextual Marketing Personalization

Walk is where you start using what you know about someone, not just who they are. Engagement level, pages visited, content consumed, lifecycle stage, category affinity. That data is already sitting there, walk is about activating it across all channels that the customer sees.

Category-specific messaging means a running shoes buyer sees different hero content in email, different product modules on the homepage, and different retargeting creative in their feed. Behavior-based variation means a loyal repeat customer gets a different offer than a one-time buyer, whether that shows up in their inbox (a smart use of post-purchase email flows), on your site, or in a direct mail postcard. A product recommendation block built from someone’s recently viewed items makes the email feel built for them, because it was.

This stage usually requires a bit more work, integrating customer data with content creation. It means utilizing dynamic content blocks where entire sections move in and out based on category, behavior, or lifecycle. Dynamic CTAs which offer a discount to a lapsing customer and a new product launch to a recent buyer. Ads which show different things to non-purchasers compared to purchasers. Done right, the customer sees a consistent experience whether they’re opening an email, landing on your site, or seeing a paid ad.

E.g. segment your next nurture send by top purchase category. For an apparel brand, write one hero for activewear buyers, one for outerwear buyers, one for accessories buyers. Keep a generic fallback for everyone else. Compare click-to-open rates in an A/B test for personalized versus generic and see if there’s a lift. Personalized content almost always outperforms generic, and that makes the case for further personalizations. Those same segments power contextual website modules, retargeted ads showing the actual products viewed, and direct mail variations. Once built, you can use them everywhere, same audience logic, different channel.

Run: Adaptive Marketing Personalization

Run is where personalization stops being manual and starts being system driven. AI-driven content recommendations, dynamic email modules that change based on real-time behavior, predictive messaging that anticipates where someone is moving next. This is an advanced state, not the state most brands assume is the start.

When done well, adaptive personalization compounds over time. Like the ubiquitous “algorithm” that we pretend to dislike, but feeds us content we love, the recommendations get better with time. YouTube knows what you want to watch before you do. Amazon predicts what you’re going to buy next Tuesday and feeds you ads this week. The more data you have about someone, the better the personalization gets, and the more channels you can apply it to. The same data that surfaces a product recommendation in an email can serve it onsite, in an app push, and in a paid ad.

Done poorly, it can be creepy. “Hey [BOB?], we noticed you looked at our website 15 times in the past 546 days!” Too much personalization destroys trust faster than generic emails.

Getting to run is tough. It requires a clean data foundation, and successful crawl & walk personalizations are evidence you’re ready. It requires advanced tooling and ongoing optimization. Luckily, the AI tools are already there and are readily accessible. But even the smartest AI can’t fix bad data, so make sure the foundation and context are strong first.

E.g. before investing in an AI recommendation engine, audit two things: the data feeding into it, and the platforms running it. Pull a sample of customer profiles and check how complete they are. Do you have purchase history, browse behavior, and engagement data for most of them, or are large chunks of your database missing key fields? Then look at your stack. Can your ESP, CDP, and ad platforms actually talk to each other, or is your customer data siloed across tools that don’t share? If either is the problem, fixing that is a higher-ROI move than adding a recommendation layer on top. Clean data on the right platforms is what lets the same recommendation logic work across email, onsite, and ads. Without it, you’re just surfacing bad recommendations faster, in more places.

The Part Most People Skip

These stages don’t end just because you’ve added more. Crawling and walking are still useful, and often more efficient than running. You can use all of them, forever. A brand sending adaptive AI email content can still utilize first names in subject lines, and send different copy based on engagement level. Heck, they can still send generic emails too. A brand just starting out doesn’t need to run immediately, they can just focus on using what they have.

It’s not crawl, walk, then run. It’s crawl, walk, and run. Ask yourself: what quality data do you already have that you’re not using? Start with the quickest wins you can ship, prove the lift, and let the results fund the next step. Foundational personalization is where most brands find immediate value, and the case for contextual and adaptive builds from there.

Personalized communications feel more human. Customers will notice. Your metrics will too.

 

Sources

    Piqued your curiosity?

    Let’s talk about how we can grow your business.