Personalization and analysis of data in offline retail

We translated an interesting overview article on the prospects for personalization in offline retail. We recommend reading what the leaders of the user experience personalization industry think about.

Imagine, when you enter the store, everyone knows everything about you: name, size, purchase history, even your views on life, on the world around, on everything.
')
How will you feel? As a celebrity or as a victim of manic persecution?

How you answer will affect and already affect the future of retail, which is currently undergoing a radical transformation driven by real-time big data analysis.

Barbecue and beer

We are accustomed to the fact that retailers constantly collect data about us since the company Dunnhumby helped Tesco supermarkets to set up their cardcard loyalty system in 1994.

But today, a simple shopping history is complemented by a number of new data. For example, the weather is an extremely important parameter for understanding how much you need to put on the shelves of umbrellas, barbecue sets or beer. This also includes data from social networks and mobile devices.

The ability to analyze this data in real time provides an unprecedented opportunity for sellers to offer us specialized services both online and offline.

“If you know what your customers are buying and what you have in stock, then you can create the right special offers to meet their needs, and at the same time do it in real time,” says Klaus (Klaus Boeckle), a software representative. SAP, whose data analysis platform “Hana” is used by such trade monsters as eBay and B & Q.

Created to measure

Merchants, using the power of powerful analysis systems, will be able to determine by the posts in social networks how we plan to spend the weekend or that we are going to buy new clothes. Then they will encourage the acquisition of relevant products - either those that will be of interest to us now, or those that we bought earlier.

“Apple’s iBeacon technology (indor navigation) designed to interact with smartphones will soon allow retailers and mobile app developers to identify us when they enter the store,” says Coin-sized Owen Geddes, Appflare, a company specializing in deploying and managing devices. .

Special offers will then be displayed in our smartphones and will vary depending on where we are in the store.

However, we will always ask the client’s permissions for this, says Geddes, warning critics about the violation of the user's privacy.

“A lot of data improves service,” says Scott Silverthorn, head of data for cosmetics retailer Lush.

Companies today have the opportunity to use big data in working with their staff. And not only in the sales area, but also in stock, which allows you to control sales statistics literally at your fingertips.

This not only helps maintain competitive spirit in employees to improve sales and productivity, but also gives them information to improve customer service.

“For example, if employees noticed that bath bombs are a good buy along with a certain shampoo, then they can change the layout in the store so that these products are next to each other,” he says.

The bigger, the better

Amazon’s success, which is close to 240 million customers worldwide and annual revenues of $ 75 billion, can be traced to its ability to analyze customer data and adapt its service to them, says David Selinger, CEO, RichRelevance. specializing in content personalization solutions.

“Already in 2004, Amazon had more data processing capabilities than most retail chains today,” says David.

Werner Vogels, director of technology for Amazon, advises: “There is never too much data — the more, the better. The more you collect, the clearer the results you get. ”

The heyday of cloud computing and real-time data processing allows store owners to create targeted offers for their customers much more accurately, Fogels continues to speculate.

“For example, on a particularly cold day of winter in your city, a retailer will be able to recommend a coat from the collection, items from which you purchased earlier. In the future, when you add other data sources, voice or video, your possibilities will become much more interesting. ”

Amazon’s purchase of a referral engine that offered products based on behavior and ratings was “not entirely successful,” Vogels says. However, thanks to "machine learning" he began to work better.

“You are looking for a kettle and we recommend a kettle that is better suited to the things you have already bought for your kitchen from us.”

Resistance

Traditional retailers resist the attack of Amazon, owning their own big data arsenal, claims Selinger.

His company, RichRelevance, with clients such as Marks and Spencer, Boots, John Lewis, Argos, Dixons and Ann Summers, specializes in processing arrays of data that retailers already accumulate about their customers and use them to personalize the customer experience.

RichRelevance software processes this data through the open-source Apache Hadoop framework, and then applies 125 different algorithms that try to predict which products the client is likely to buy depending on its previous and current behavior.

And all this is done in 20 milliseconds, says David.

Each of the algorithms is evaluated for accuracy and its predictive ability, and these ratings affect which images and offers will be shown to customers the next time they visit the retailer's website or any other site, for example, the same Pinterest.

Selinger calls this process "multi-threaded learning" (ensemble learning).

“By helping consumers find products that are most relevant to them, we increase sales by 3-10%,” he says.

“And this is not only about website content that can be adapted to consumer preferences and consumer behavior,” emphasizes David Brussin, executive director of Monetate (e-commerce personalization), whose system is used by 400 brands around the world.

Even the content of marketing letters can be adapted based on how you click on them, he says.

Well ... It seems that we are already inside this new wondrous world - whether we like it or not.