Retailers collect a lot of consumer data online that could help store associates sell more effectively. When combined with machine learning technology, that digital data becomes even more powerful.

Shai Cohen, vice president of sales,

Shai Cohen, vice president of sales,

With direct-to-consumer brands including Allbirds, Bonobos, Casper, Glossier, Warby Parker and Zappos all opening physical stores, it’s pretty clear that online-only isn’t the answer for retail segments where touching, feeling and experiencing the brand provides value.

But what about the other way around?

Can traditional retailers benefit from the digital extensions of their brands to improve their in-store customer experience?

What if the salesperson who helped me that day could have had access to my wife’s browsing and shopping data on Sephora’s website and app?

To obtain a better understanding of the possibilities of offline customer experience powered by online engagement, we explored the opportunities available to modern retailers with both a digital and multichannel presence.


Privacy versus convenience

In an informal survey of customers, more than 80% were happy to enable in-store sales staff to access their digital engagement in order to provide a better shopping experience.

I know that my wife likes shopping at Sephora, so when I’m away for longer business trips, I look for a Sephora store to buy her a present. On my last trip, I purchased Philosophy Renewed Hope in A Jar Refreshing & Refining Moisturizer, but what my wife really wanted was Philosophy Renewed Hope in A Jar Overnight Recharging & Refining Moisturizer.

Maybe I’m not the most attentive husband, but I’m certain that I’m not the first person to confuse ‘refreshing’ with ‘overnight recharging’.

Well, what if the salesperson who helped me that day could have had access to my wife’s browsing and shopping data on Sephora’s website and app? Then she would have known that my wife has purchased ‘Overnight Recharging’ three times, the last time six months ago, meaning that she should be ready for a new moisturizer.


In fact, based on my wife’s ‘Love List’ on Sephora, the salesperson could have made five different product recommendations which would have made my wife happy.

The value of machine learning

But the real value of accessing my wife’s shopping data comes by apply machine learning technology. What product recommendations could predictive technology make based on the items my wife browsed, put in her shopping cart, or bought?  A complementary shade of matte powder foundation (that’s face makeup for the uninitiated) to one that my wife bought last year at a 30% discount would be an easy up-sell opportunity for Sephora.

And in-store up-sell and cross-sell opportunities could be easy and instantaneous when utilizing predictive technologies. Because all of these product recommendations will be run by machine learning technology and appear automatically on the salesperson’s tablet or phone based on in-store availability, the salesperson will be free to focus on servicing the customer.

And based on the shopping patterns exhibited by my wife’s browsing, she would be placed into a segment of Sephora shoppers, enabling the predictive machine learning technology to make additional product recommendations based on the products purchased by similar shoppers, either to me or to my wife when she visits a Sephora store.


So if I decide to buy my wife Algenist’s eye serum brightener, the salesperson could then suggest that I purchase Algenist’s eye serum concealer or eye renewal balm based on the purchase patterns of shoppers similar to my wife—all data which Sephora’s data scientists already have.

Think of this technology-driven but human-enabled experience as the ultimate personal shopper. And it’s within reach for retailers in 2019.

Competitive advantage

Physical retailers rolled out digital commerce sites and apps in order to offer their customers convenience and to compete with digital-only retailers.

Now, those retailers have an opportunity to utilize customer engagement data—products viewed, placed in the shopping cart and wish lists and products purchased—in order to empower in-store sales reps with this data while also utilizing machine learning and predictive technologies in order to make better in-store product recommendations as well as up-sell and cross-sell opportunities.


By utilizing digital shopping data, traditional retailers have the opportunity to regain the upper hand versus their digital-only competitors. provides software to help retailers predict sales and other product-related requirements.