With the advent of eCommerce, online shopping, and fierce competition for customer loyalty, retailers are increasingly using new sources of data and big data analytics to stay relevant or stay afloat. Every retailer is looking for the answer to the question: What message do you send to which customer and what is the most favourable moment? The answer in a nutshell: By applying profiling, predicting behaviour and personalizing one-to-one contact moments.
Nowadays, consumers are becoming less sensitive to mass communication. They want to get the feeling that the assortment has been created for them. This can be with personal approach and services and services tailored to their specific needs. Yet there are not many (web) retailers who use their database in combination with predictive algorithms. An e-mail message sent to a large number of recipients, often a newsletter or publicity, are still every day’s business. But someone who recently bought a product, for example a gas barbeque, does not want to receive an offer to buy coals the next week. How do you switch from mass campaigns to one-to-one marketing?
The first step is profiling, to create a 360-degree customer image by analysing the behaviour patterns of the customers. In other words, you will collect as much information as possible about your customer. When did he last bought something, what did he buy, how often does he buy with you and how much does he spend at once, et cetera. (Marr, 2015)
The second step implies that algorithms look for patterns in behaviour, related items, and similar attributes among other customers to predict customers ‘ interests and buying habits. As well as demand, inventory levels and competitor activity. (Marr, 2015) This is actually nothing else that what the staff in the shops do on a daily basis. They advise the customer and think online with them, just as it happens offline. The advantage of the algorithm is that the process is automated.
Step three is to use these insights to set up personalised email marketing. For example: If two people from a particular audience have bought a specific trouser and jacket and a third only the pants, then it makes sense to bring that matching coat to his attention. This results in one-to-one communication with high relevance to the receiver and thus a high conversion ratio. Another smart way to use the available data is to look at the customer’s buying history. Machine learning models are trained in historical data which allows the retailer to generate accurate recommendations. (Virmani, 2017)
So in conclusion, it comes down to the three components. First of all to the company should create correct profiling of customers. Furthermore, the company should collect enough information about the specific customer to sort him or her in a in a so called ‘profile’. In addition an automatic algorithm looks for opportunities in the database to advise customers on a more personal basis. Finally with more personalised email marketing, the one-to-one communication between company and customer, is used to advise customers in a more efficient matter.
Virmani, A. (2017), ‘How Big Data is Transforming Retail Industry’, 23 February 2017 [online], https://www.simplilearn.com/big-data-transforming-retail-industry-article [Accessed on 09 October 2017]
Marr, B. (2015), ‘Big Data: A game changer in the retail sector’, 10 November 2015 [online] , https://www.forbes.com/sites/bernardmarr/2015/11/10/big-data-a-game-changer-in-the-retail-sector/#188df1b59f37 [Accessed on 09 October 2017]
