Unlocking the Value of Data: How Analytics Can Offer Customer Insights to Boost Your Business

Find out how one retailer determined what customers would buy next and when using predictive analytics.

With businesses across all sectors undergoing digital transformation at an extraordinary rate, huge amounts of data are being generated. Collecting and storing this data securely is a challenge for all businesses, with information being created in various formats and coming from a number of sources including sales demographics, social media engagement and IoT sensors. However, this data is a huge asset for any organisation and can no longer be ignored, especially as tech-savvy start-ups continue to disrupt entire industries with data-driven strategies.

Unlocking the value of this data can be difficult, but if handled correctly, the insights gained can have a significant impact on organisations of all sizes. Moving away from antiquated business practices and adopting new data analytics procedures can help companies to work faster and smarter, boost productivity, improve customer loyalty and ultimately increase profits. Once a data analytics strategy is in place, businesses can discover new ways of finding value in the metrics they already hold, and those that they will gather in future. The insights that can be gained from analysing this data are varied, depending on the sector, and include determining when factory floor machinery is likely to need maintenance, foreseeing retail stock requirements and gaining clearer visibility of customers and their purchasing habits.

“Moving away from antiquated business practices and adopting new data analytics procedures can help companies to work faster and smarter, boosting productivity, improving customer loyalty and ultimately having a positive impact on profits”

In a blog post about turning data into customer insight, Bob Rogers, Chief Data Scientist for Big Data Solutions at Intel, gives the example of how a French household electronics company is using predictive analytics to foresee future customer buying behaviour. The company had a huge amount of raw data on its customers, including previous buying behaviour, and maintenance requests, but it didn’t have the expertise to use this to predict what these customers might do next. In particular, the retailer wanted to identify the 500 households most likely to make a purchase within the next year, and which appliances they might buy. Armed with this information, the company would be able to form a detailed business strategy for the purchasing trends of their customers.

“In order to gain this insight, the company decided to use the Cloudera* distribution of Hadoop* (CDH*) and model codes developed by Intel to process large amounts of raw customer data from nine different sources,” explains Rogers. Software company Cloudera’s open source CDH platform is the most popular version of the Hadoop analytics framework, which is used for processing large datasets.

“This data was merged and aggregated to create 106 variables covering basic customer information, household information and segmentation, previous purchase behaviour, previous service and maintenance requests and events triggered by them, and government demographic information,” said Rogers.

Next, to work out which element of the customer data had the strongest impact on their purchasing habits, a ‘random forest’ algorithm was applied to each one. A random forest algorithm comprises multiple decision trees and is used for accurately classifying large amounts of data in order to make predictions. Based on this process, each customer was given a ‘buy’ or ‘no buy’ flag to indicate whether each of the variables had a positive impact on the likelihood of them making an imminent purchase.

Once this was established, the customer households were placed in order of the likelihood of them making a purchase, from the highest to the lowest. Using customer data from 2005, Intel tested the model using it to predict buying behaviour then comparing the results against the actual purchasing data for 2005. The result was an impressive accuracy rate of 68 percent.

“With this predictive analytics model in place, the retailer now has a reliable method of predicting customer behaviour, which can be adapted to take into account different variables as needed,” explains Rogers. “With this additional insight, it can offer much more timely and targeted communications to its most valuable customers, increasing loyalty and boosting revenue opportunities.” Using this kind of data-driven insight, the retailer can identify when a customer is likely to buy a particular product soon and send them targeted marketing material including details of products that are likely to fit their needs, and suggestions for trading up to the next model.

This example shows how taking a customer-centric approach to data analytics can help businesses to develop a clear strategy for the future. Data scientists can create predictive analytics models that enable businesses to operate more efficiently, highlighting the importance of investing in analytics skills within the company. While the case study here concentrates on retail, unlocking the value of raw data can benefit companies across all sectors from healthcare and manufacturing to entertainment and financial services.

*Other names and brands may be claimed as the property of others

For more information:

Stay Connected

Stay connected to technologies, trends, and ideas that are shaping the future of the workplace with the Intel IT Center.

By submitting this form, you are confirming you are an adult 18 years or older and you agree to Intel contacting you with marketing-related emails or by telephone. You may unsubscribe at any time. Intel’s web sites and communications are subject to our Privacy Notice and Terms of Use.