Personalization Has Changed. Has Your Strategy?

LinkedIn
Twitter

Explore AI Summary

By US Sales Director, Aaron Waters

Ten years ago, putting a first name in a subject line felt innovative. Today it’s just mail with a friendly face.

That’s the line from our recent joint webinar with Bloomreach, hosted with Bloomreach’s Simon Mills. The topic: why relevance, not recognition, is what actually moves customers now. Here’s what came out of it.

The bar moved

You’ve got a dozen emails in your inbox right now that just say “Hi, [Name].” You scroll past almost all of them.

Consumers don’t compare your emails to your competitor’s emails. They compare them to Netflix’s recommendations and to Spotify building a playlist around their mood on the fly. That’s the bar now, whether a brand realizes it or not.

Case in point:  a grocery chain in central Europe, Turno, found its average customer shops every 3.5 days. Instead of a generic weekly newsletter, they built an “empty fridge” campaign: a reminder featuring each customer’s most-purchased items, timed to when the fridge was likely running low.

The customers and the products stayed the same, but the timing improved and conversion went up nearly 30%

Personalizing the campaign isn’t the same as personalizing the journey

A lot of retailers personalize one email, then send every customer to the same generic homepage. 

Take product search. Search “cheese” on most retail sites, and you get 200 SKUs back. Overwhelming. Jumbo, a Dutch supermarket, used Bloomreach’s AI to rank cheese results by relevance per shopper instead of dumping everything on them. Add-to-cart and revenue per visitor went up. They applied the same logic to fresh produce next, and a small tweak turned into genuine growth. 

Five data points that actually matter

You don’t need 500 customer attributes. We found that just a few can move the needle: 

  • Product-level purchase history. Not just what someone bought, but which brands, and how often. This is the foundation under replenishment reminders and substitution recommendations. It’s what powered Turno’s empty fridge campaign.
  • Frequency and timing. How often does this specific customer shop? A wine merchant in Switzerland used purchase timing to figure out the right moment to send a cart recovery email. Too early, and you interrupt someone still researching. Too late, and the intent’s gone cold. Get the timing right and that campaign converted over 40%.
  • Browse and search behavior. It tells you what people want. It also tells you what you don’t have and maybe should.
  • Channel preference. In the US, I don’t really get Whatsapp messages. But work in the UAE or Africa, and SMS and WhatsApp are basically the only channels that matter. There’s no universal answer here.

Put those together and you can answer three questions: what does the customer want, when will they need it, and how should you reach them.

Testing is the gap

Most retailers have no shortage of data or tools, but they don’t have the time or opportunity to keep testing. 

A team launches one welcome email, one cart-abandon flow, and then call it a day. Meanwhile, a design furniture retailer tested something as small as a heart icon on a product page to let customers favorite an item. Conversion on favorited items went up 95%, and the bounce rate dropped 34%. One tiny change, properly tested, beat many larger campaigns.

What the numbers actually show

We see the same pattern across grocery, beauty, travel, and finance. Revolution Beauty hit a 71% open rate on AI-personalized reactivation emails, aimed at customers who were about to leave. Charles Tyrwhitt grew engagement and revenue by 20% after consolidating its tech and personalizing across channels. Coop in Denmark bumped loyalty offer redemption by 4.3 points just by pinging shoppers about an active offer right as they scanned in at checkout. A Dutch vitamin retailer saw a 73% jump in revenue on branded products after personalizing that specific product line.

Forrester’s research puts the average three-year ROI for Bloomreach customers at 251%, with payback in under six months. Personalized audiences convert 27% more than batch sends.

One thing to do tomorrow

You don’t need to overhaul everything at once. Pick one dimension of real-time context and build it into a campaign you’re already sending.

Stock availability is a good place to start. If you’re promoting a product across email and WhatsApp, make sure it’s actually in stock by the time the customer opens the message. Nothing kills trust faster than clicking through to “sold out.”

Simon mentioned a nice example after moving to Amsterdam: a grocery app there, Albert Heijn, ran a barbecue push. His banner showed sausages, in English. His Dutch, vegetarian colleague opened the same app and got vegetarian sausages, in Dutch. Same campaign, same day, completely different experience. 

That’s the level of contextualization worth aiming for.

Where this goes next

Our prediction: agentic AI. AI that builds the campaign, writes the copy in your brand voice, picks the channel and timing, launches it, then uses the results to tune itself. Bloomreach’s own agent, Lumi, already does a version of this from a plain-text brief.

Pair that with real-time contextual decisioning, sunglasses for the customer in the sun, a warm hat for the one somewhere cold, and a marketer’s job suddenly becomes easier. 

The point of all of it

Personalization is about making customers feel understood.

That’s what creates a brand someone comes back to. Bloomreach brings the customer intelligence, and Publitas turns it into shopping experiences people actually want to click through. Together, that gap gets a lot smaller.

Watch the full webinar here

Want to see what this looks like for your catalog? Get in touch with our team.

Search:

Search

Subscribe:

Tags