E-retail is without a doubt growing rapidly in the United States as well as on a global level. E-commerce sales in 2013 accounted for 5.8 percent of total sales in the US, according to the US Census Bureau. And worldwide business-to-consumer (B2C) ecommerce sales will increase by 20.1% in 2014 to reach $1.500 trillion.
With this tremendous growth in online shopping across the world, retailers need Big Data analytics to stay competitive. With the help of data analytics, companies are able to gain descriptive and prescriptive insights. This means that retailers can now not only find out what happened in a particular sales period, but also what they can do in the future to increase sales.
For companies looking to make those investments, here are the three key areas within their organizations that can benefit from data analytics insights:
Marketing
Marketing is one area where analytics can add tremendous value. Exorbitant marketing spends entailing huge discounts coupled with predatory pricing has become the norm of the industry. However, despite the high marketing spends, metrics such as open and click-through rates of email and banner ads portray a different picture. Our pick for three analyses e-retailers should have in their repertoire are:
- Running targeted campaigns Blasting emails to your entire customer base is a waste of money and might lead to your customers unsubscribing altogether besides resulting in your emails ending up in the Spam folder. We recommend using targeting models to identify the most relevant customers for each campaign.
- Predicting churn and attrition Retaining customers is easier (and more profitable) than getting new ones. Churner prediction models built on demographic, transaction and engagement history can help you identify customers likely to churn so that preventive steps can be taken to retain them.
- Understanding customer behavior Segmenting and understanding your customers in detail will help you understand what they are looking for and enable you to formulate the prefect strategy. For example, Bill Clinton won the 1996 U.S Presidential Elections when his advisor Mark Penn successfully identified and influenced the micro-segment Soccer Mom to vote for him. Penn’s 1996 Neuro Poll helped him identify a new swing voter: the “soccer mom.” Penn urged Clinton to focus on policies that appealed to suburban parents and to speak about these policies in terms of values rather than economics. He subsequently became famous for focusing on the “soccer mom”, cited as the key swing vote that helped President
Fraud and Risk
Three analyses to prevent fraud and mitigate risk are:
- Identifying fraudulent transactions Every transaction needs to be scrutinized in real time, and given a risk score based on the customer and product history and demographics. Riskier transactions should be validated thoroughly before shipment.
- Preventing illegal practices Marketplace model of e-retail gives customers a wider variety of products but it presents a unique set of challenges too. Companies should use fraud detection models on buyer-seller-product-time panel data to curb money laundering and prevent sale of illegal products.
- Identifying bad sellers or suppliers To mitigate the risk of losing customers because of bad sellers or suppliers not fulfilling their commitments, retailers must build scorecards that can help identify bad partners.
Social Media
72 percent of Internet users are active on various social networks and 70 percent of active online adult social networkers shop online. This makes social media analytics an essential element to keep companies competitive. Keeping followers and likers engaged is only the first part. There are several other insights that retailers can glean from social media analyses.
- Understanding consumer perception Understanding the competitive landscape is imperative for any business and this can be done through social media analyses of what people are saying about various brands. Have a look at the Twitter analysis we did of the top four U.S retailers here.
- Preventing PR mishaps Given the viral nature of social media, a single bad review can precipitate into a colossal PR mishap. Take for example the case of Dave Caroll vs. United Airlines. Canadian musician Dave Carrol and his band, Sons of Maxwell wrote a protest song titled “United Breaks Guitars” based on his real-life experience of how his guitar was broken during a trip on United Airlines in 2008, and the subsequent reaction from the airline. The song became an immediate YouTube and iTunes hit upon its release in July 2009 and a public relations embarrassment for the airline. The only way to stay ahead of mishaps like this is to build and keep a regular check on social media tracking dashboards and to build models to predict viral content.
- Identifying and engaging social influencers Social influencers are at the center of social networks. They are the ones helping create trends and opinions on social media. Identifying and engaging key social influencers can multiply your reach and help you reach your target consumers.
Analytics can also help various other areas of business from pricing and supply chain optimization to demand forecasting and building accurate recommendation engines. We will tackle this in the second part of this article.
