Don’t see this when you visit Barneys’s homepage? That’s because the Web site is now personalized. (Barneys New York)

After years of using customer data to fine-tune their marketing efforts to smaller and smaller groups, retailers are now making a massive and expensive effort to tailor their Web sites and promotions for an even narrower target: an audience of one.

Retailers and experts say the industry is in the early stages of a push toward personalization, in which individual shoppers have different experiences on retailers’ Web sites and receive highly customized e-mails, coupons and special offers.  CVS pharmacy, for example, might e-mail a shopper about allergy medication based on weather conditions in their geographic location.  At luxe department store Barneys New York, shoppers who browsed a collection from Italian designer Marcelo Burlon are likely to see a version of the Barneys Web site that prominently features clothes from like-minded designer Russell Westbrook. At Blue Nile, the online diamond retailer, the company is timing its customer e-mails about wedding bands based on when it estimates the customer got engaged.

Personalization may not sound cutting edge if you have ever browsed for a sundress on your favorite retailer’s Web site, only to have ads for that same sundress chase you around the Internet for days. But the technology retailers are using now is much more sophisticated: They are studying a wider range of activities — not just the last Web sites you visited — but also whether you opened or clicked on their emails, read a company blog post or previously redeemed a certain kind of coupon.

“For one million users, we want to have one million different site experiences,” said Matthew Woolsey, Barneys’s executive vice president for digital.

In March, Barneys rolled out a new Web site in which nearly every page —the homepage, category pages and individual product pages — feature personalized content that is served up based on data from both a shopper’s in-store purchasing and online browsing behavior.

This combined analysis of in-store and online patterns is particularly leading edge, industry experts said, as many retailers are still struggling to funnel these data sets together.

Woolsey said the unified data can have powerful results. For example, Barneys has learned that many of the women who buy fine jewelry in its stores have previously browsed for it online. If Barneys purely looked at these shoppers’ Web browsing history, Woolsey said they might deduce, “She’s never buying anything, so let’s try something else.”

In fact, by looking at this shopper’s behavior across channels, Barneys learns it is indeed valuable to keep showing her digital jewelry lookbooks: She’s interested in the products, she’s just closing the deal in person.

Zulily, the flash sales site, said the key to its personalization algorithms is learning about the customer over time, rather than simply looking at recent activity.

“We don’t like to shift preferences too quickly,” said Mike Errecart, the company’s director of software engineering, since perhaps a click or two on an item outside your comfort zone might not be a good reflection of your brand and price preferences.

The exception to that rule, Errecart said, is on maternity purchases. Since pregnancy is inherently short-term, the algorithm gets tweaked.

“We weight your most recent interactions more heavily and decrease weight of older interactions,” Errecart said, to determine what to serve you.

As retailers invest in personalization, they are closely watching how consumers respond to these new tactics, which experts say can quickly go from cool to creepy. The results of a recent study from consulting firm Accenture captures the challenge retailers face: Though 60 percent of shoppers said they want retailers to serve them real-time promotions and offers, only 20 percent of customers want those stores to know their current location. In other words, it appears shoppers are uncertain about exactly what kind of data they want to turn over to retailers.

Experts say that consumers, especially millennials, are generally willing to provide information to retailers if it is being used in some way that generates obvious value for them. (Think of this as the Google Maps principle:  People are typically willing to give up their location to Google when using this app, because, in exchange, they know they’ll get accurate directions that simplify their travel.)

Marcie Merriman, a consumer-engagement consultant at Ernst & Young, said some nascent personalization strategies have struggled because they’re too simplistic.

“It’s not based on understanding what’s missing in the consumer’s life,” Merriman said.

And that’s why retailers such as CVS are also paying attention to what you’re not buying at their stores. CVS’s personalization efforts are centered around its ExtraCare program, which was used by more than 90.8 million households in the last year. Even if you haven’t bought vitamins or toothpaste there in a long time, their data might still determine you’re a good candidate for a coupon for those products.

“Sometimes they may not realize we carry some of their other favorite items or that we may have a special that week on a type of product they would normally pick up elsewhere,” said Melissa Studzinski, CVS’s vice president of customer relationship management, in an e-mail.

Plenty of other retailers are experimenting with some form of personalized digital experience, including Best Buy, Nordstrom and Urban Outfitters.

Experts said that many of the personalization efforts seen across the industry so far do not fully unlock the potential of this technology. For example, some retailers’ idea of personalization is simply to serve up lots of red cocktail dresses to someone who bought a red cocktail dress. But, more than likely, that customer doesn’t want two red cocktail dresses — she’d be more likely to spring for a gold clutch and a sultry lipstick to complete the look.

Errecart, of Zulily, said another key challenge in designing a personalization algorithm is figuring out not just the brands and price point that someone finds appealing, but their actual style preferences.

“Are you preppy? Are you trendy? Do you tend to like traditional garb or pretty pink princess stuff for your daughter?” Errecart said. “There’s a lot to fashion and shopping that’s about stylistic stuff.”

Even with reams of big data, that is still a tough equation to get right.