The predictable consumer: 4 case studies
Nike is selling its fuelband data to marketing companies as a product. Consumers with a Nike fuelband are gifting Nike and its partner companies with a wealth of data. Nike knows exactly when someone needs a new pair of trainers and that is obviously invaluable information.
This is just one of the many examples that demonstrate what modern marketing is all about: getting your hands on relevant consumer behavior data and using it to best advantage. Back in the day marketing was based on understanding the average consumer; nowadays marketing tries to understand every individual customer.
It’s been said that consumer behavior can no longer be predicted. Nothing could be further from the truth. Consumer behavior has never been more predictable. Companies like Netflix only launch new TV shows when the available data predict success. Analysis of Twitter conversations can predict the following day’s stock movements with 95% accuracy. Analysis of our calendars gives a highly accurate picture of what our following day will look like, and so on and so forth.
We live in a world where consumer behavior is more predictable than ever. Every individual consumer feeds the corporate world tips as to his of her current expectations. In reality, though, only a handful of companies are making the most of these opportunities. The concept â€œbig dataâ€ alone deters a lot of companies. Let’s be honest: most companies can’t even handle â€œsmall dataâ€. After all, how many organizations have a database that’s truly useful and up to date?
In the future, big data management will become a necessary competency if companies are to be highly customer-centric. Consumer expectations evolve. Customers will be less tolerant towards wrong messages at the wrong time. Apart from this low tolerance, a lack of interest will automatically result in a limited impact. On the other hand, customers display a large degree of openness towards the right (commercial) messages delivered at the right time. My study shows that 1 in 3 customers have a very positive attitude towards personalized ads. Customers like to buy products. They also like information on products but only if it’s the right info at the right time. The pointers (content) provided by consumers hand us the ammunition we need to meet this challenge. The modern marketer should turn his attention away from the average consumer and toward the individual consumer.
Case 1: The Weather Channel
The weather channel can predict what products its customers will buy based on local weather predictions. This knowledge is then used to sell very specific ads to local companies. The first warm day in spring in Chicago is good news for the local airco manufacturers. This may be an obvious example, but just like the daily weather forecast, they can predict consumer behavior every day of the year. Digital ads account for half of the weather channel’s advertisement revenue. By linking their big data story to the fast adoption of smartphones they are expecting to generate a marked increase in digital revenue. Their goal is to create the perfect ad for every individual consumer.
Case 2: Taco Bell predicts success of a new product
Thanks to their social command center, the popular fast food chain Taco Bell is able to predict the success of product innovations with 90% accuracy. Every year, some 18 million messages on Taco Bell are posted online and the real-time analysis of these data helps them predict success or failure with eerie precision. Thanks to the use of big data they haven’t had a failed product launch in over 15 years.
Case 3: Starbucks can predict the ideal location to open up a new coffee shop
Opening up a new coffee shop is always a calculated risk. In one location the shop may be buzzing with customers whereas 100 yards down the road, that same new Starbucks may be forced to close shop six months later.
Researchers have used Foursquare data to determine which locations are most suited for a new Starbucks. For Starbucks, the degree of competition turned out to be the deciding factor in customer frequency. The study revealed that the Foursquare data in itself are not enough to make the right choice. The prediction becomes more accurate when the Foursquare data are added to existing socio-demographic data. The combination of offline and online data improves the odds of success for a new Starbucks outlet.
Case 4: Wonga.com evaluates creditworthiness
Wonga.com is a financial player in London. The company gives payday loans to consumers and is one of the fastest growing players on the British financial market. They discovered that evaluating creditworthiness based on the classic data and bank blacklists was not enough to make an accurate assessment.
Wonga created its own algorithm to evaluate its clients correctly. They use classic data supplemented with data taken from social media. Every new customer brings new data and makes the assessment more accurate. Adding data from social media gives a clearer picture of consumer spending behavior than impersonal financial data. This has proven particularly useful for the evaluation of creditworthiness and it has also made the organization more profitable.