Revisiting Personalization: Why Exactly Should We Expect Strong Returns?

The biggest barrier to more sophisticated segmentation and personalization is better data.  If everyone can do the same thing, is there a path to long-term upside?

I continue to read articles almost every day promising AI-driven precision targeting, whether that is segmentation or personalization, will deliver outsize returns for anyone who does it right.  Category is never mentioned, oddly.  So it is worth revisiting the question, can it really be that easy?  (See this recent MediaPost piece: Supercharged Segmentation: The Secret Behind 3x Growth.  3x growth from segmentation? Hmm.)

If “better data is the biggest barrier” to sophisticated personalization, and every major brand has invested heavily in Customer Data Platforms (CDPs) and AI tools, why do most of our inboxes and ad feeds still feel like mass marketing with a name tag?

The premise of the sub-head is both true and misleading. Yes, data is the engine, but when every major retailer uses their own narrow view of a consumer’s total transactional history, the same third-party demographic tags, and the same off-the-shelf AI models from the same handful of vendors, the "personalized" experience rapidly commoditizes and any competitive advantage fades into background noise.  Remember Marc Pritchard’s famous quote from 10 years ago:

“We targeted too much, and we went too narrow. And now we’re looking at: What is the best way to get the most reach but also the right precision?” 

This is still the core problem we must solve to justify the continued investment. As I wrote in the not-so-distant past (May 2025, What Do Marketers Think Is the Point of Personalization?):

“…we need to explain why personalization will deliver all these wonderful benefits, different than anything that has been done in the past, how we know it will work, and how brands and retailers should tailor their investments accordingly. Until we can do that, it is not enough to just hope it works, or that AI is some magic tool that will make it work better than in the past.”

So, is anyone winning this arms race? Absolutely, according to McKinsey and BCG, although I have some trouble with the source of their widely quoted data points.  I might suggest that the answer is a qualified yes, with the key difference being the desired outcome.

Consider other tactics marketers use, including price promotion and product innovation, that are critical to delivering incremental revenue, generally through short-term spikes in demand. A great sale drives a single transaction, and a new innovation is eventually copied or eclipsed by a competitor. They are extrinsic forces applied to the customer.

Personalization, when done right, is an intrinsic force—it's about creating an experience that makes the customer feel understood and valued, which builds loyalty and emotional connection. This is why personalization, not promotion, has the potential to drive long-term gains. Why? Personalization is fundamentally about increasing Customer Lifetime Value (CLV) by increasing purchase frequency and boosting Average Order Value (AOV) through relevant up-sells and cross-sells over the long-term, while reducing churn and price sensitivity along the way.

For example, Starbucks has consistently reported that their personalized mobile offers generate significant incremental sales. Their success isn't just about giving a discount, but about building a digital relationship where personalized messages and product suggestions boost order frequency and drive members toward high-margin items. The ability of their app to push an offer like, “Happy Birthday, dessert's on us” or a tailored recommendation brings people into the store and transforms a series of transactional events into a loyal brand relationship.

The long-term upside is out there, but only for brands willing to focus on three critical areas that move personalization beyond a mere tactic:

1. Orchestrate for CLV, Not CTR: Stop optimizing personalization solely for short-term Click-Through Rates (CTR) or non-revenue metrics. The goal should be to reduce friction and deliver genuine value that increases Customer Lifetime Value (CLV). This often means prioritizing relevant recommendations over immediate discounts.

2. Focus on Unique Data: If you use the same data inputs as your competitor, you get similar outputs. The long-term advantage lies in creating proprietary data (e.g., through unique loyalty program mechanics, community engagement, or in-store interactions) that no one else can replicate.

3. Personalization as Product: Treat the personalized experience itself as a core product feature. The success stories are from brands that see their app, their emails, or their in-store recommendations as an integral part of what they sell and how they engage, not just how they sell.

Do you agree or disagree? Want to chat? Find a convenient time here.