Tanusha - Fotolia
The holiday shopping season is in full swing, and retailers are doing all they can to make the most of these make-or-break business weeks. For many, retail analytics tools are a key component of that effort, providing a way to identify potential online customers and target them with ads, recommendations and special offers tailored to their interests.
For example, The Grommet, an online retailer that specializes in small-run consumer products, is using analytics to help devise strategies for promoting items that are likely to appeal to shoppers on its website and through its social media accounts. "This time of year, it's all about making it as easy as we can to help customers find that perfect gift," said Tori Tait, senior community manager at the Somerville, Mass.-based company.
The Grommet primarily analyzes social media data from the website Pinterest through a managed service run by marketing and analytics technology vendor Curalate, which it started working with in 2012. The retailer tracks how many times images of items it sells or webpages about them are shared on Pinterest to gauge consumer interest in specific products.
Working faster -- and smarter
And the approach is paying off, according to Tait. Banner ads placed on the site featuring products that have proven to perform well on social media have a 50% higher click-through rate than ads based solely on the marketing team's instincts, she said. And by basing decisions on what customers have demonstrated an interest in, and reaching people where they're actively engaging with products, the company has been able to move more quickly to capitalize on business opportunities during this critical time of the year. "It takes out a lot of the subjectivity that slows down the workflow," Tait said. "We're just able to work a little smarter."
Delivering timely and relevant marketing content to customers online has become even more important for retailers because of the growth in shopping via the Web. Marc Hayem, vice president of platform transformation at RichRelevance Inc., which provides an online recommendation engine service for retailers including Kohl's and Target, said consumers today have high expectations for online shopping and want to be able to find products as intuitively as they can in stores.
"People don't understand why it's different than in the store," Hayem said. "For them, it should all be the same. This is where the market is going."
Multiple tools tackle analytics challenge
But even as the need to quickly deliver relevant information has grown, so too has the difficulty of doing so. For example, in order to develop recommendation engines for retailers' sites, RichRelevance first has to build online catalogues. The problem is that some retailers constantly add new products without ever discontinuing older ones; as a result, their product databases grow substantially every month. To keep up, RichRelevance uses an ETL data integration tool from Pentaho to load daily updates into a Hadoop cluster and then employs a variety of other open source tools, including the Hive data warehouse software, HBase database and Kafka messaging technology, to conduct its analyses and create purchase recommendations for online shoppers.
Hayem said that during peak shopping hours on Black Friday, RichRelevance's data center was handling 17,000 requests per second, whereas on a typical day that number would be 3,500. Building out the data center with an eye toward scalability was a primary objective from the start, which is what has allowed the company to keep up with the increasing demands, he added.
Terri Albert, an associate professor of marketing at Northwestern University's Kellogg School of Management, said in an interview at the 2014 SAS Premier Business Leadership Series conference in September that the best way to inspire loyalty in customers is to customize their experience. By using retail analytics technology to better figure out who their customers are, retailers can more effectively target products and marketing campaigns to the preferences of individual shoppers, she noted.
And this time of year, with all the money that's at stake, it would be a big blunder for retailers to miss out on any opportunity to boost sales. "You can't afford to make stupid mistakes," Albert said.
How to develop an analytics strategy as a retailer
Text mining has great potential to boost retail analytics
Tips on using BI and analytics tools in retail