Читать книгу Predictive Marketing - Levin Dominique - Страница 9

Predictive marketing is the evolution of relationship marketing defined and practiced by many direct marketers in the last few decades. Predictive marketing is not a technology, but an approach or a philosophy. Predictive marketing uses predictive analytics as a way to deliver more relevant and meaningful customer experiences, at all customer touch points, throughout the customer life cycle, boosting customer loyalty and revenues.

The rise of predictive marketing is fueled by three factors: (1) customers are demanding a more personal, integrated approach as they interact with marketing and sales through many channels, (2) early adopters show that predictive marketing delivers enormous value, and (3) new technologies are available to capture new and existing sources of customer data, to recognize patterns, and to make it easier than ever to use customer data at the intersection of the physical and digital worlds.

Predictive analytics is a set of tools and algorithms used to make predictive marketing possible. It is an umbrella term that covers a variety of mathematical and statistical techniques to recognize patterns in data or make predictions about the future. When applied to marketing, predictive analytics can predict future customer behavior, classify customers into clusters among other use cases. Other terms you might hear in the media to describe this process include machine learning, pattern recognition, artificial intelligence, and data mining. Predictive analytics and machine learning are used interchangeably in this book.

Predictive marketing is fundamentally changing both business and consumer marketing across the customer life cycle. It is transforming the focus from products and channels to a focus on the customer. Predictive analytics is used to improve strategies to acquire new customers, to grow customer lifetime value, and to retain more customers over time.

Innovative, technology driven companies like Netflix and Amazon have been using predictive analytics for years, and so have others like many in the telecommunications, financial services, and gaming industries, such as Harrah's Entertainment. The row of movies and TV shows “you might like” that appear when you curl up on the couch and turn on Netflix is a driving force of the company's success. It's all made possible by the translation of customer data with smart analytics. In fact, “75 % of what people watch [on Netflix] is from some sort of recommendation,” Netflix's Research Director Xavier Amatriain wrote on the company's tech blog in 2012.

Amazon has been using predictive analytics to drive success since the very beginning of the company. Recommendations that appear under a product you are thinking of adding into your cart is part of what makes Amazon such an e-commerce powerhouse today. The company has stated publicly that 35 percent of its sales comes from recommendations made by their predictive engines. That would equate to $26 billion of revenue in 2013. The company is using predictive analytics in many other ways too, such as predicting which email newsletter to send you, or to nudge you at the right times to reorder an item.

In the gaming industry predictive models can set budgets and calendars for the casino's gamblers, calculating their predicted lifetime value in the process. If a gambler wagers less than usual because they may have skipped a monthly visit, the casino can intervene with a letter or phone call offering a free meal, a show ticket, or gaming comps. Without this type of customer analytics, casino operators might not notice what could be a slight, almost imperceptible change in customer behavior that might portend future problems with that patron. For example, if a long-time customer decides to cash in all their player card points, perhaps it's because they are dissatisfied with their last experience at the casino property. Predictive analytics can quickly spot these trends and alert casino management to the issue so that they can approach the individual to find out if there is a problem. This kind of personalization can go a long way in appeasing a disgruntled customer, which might be the difference between retaining or losing them as a customer.

Harrah's Entertainment's Total Rewards, which was rolled out as Total Gold in 1997 and renamed Total Rewards a year later, is heralded by many as the gold standard of customer-relationship programs and is powered heavily by predictive analytics algorithms. The company's belief in its loyalty program grew so strong that it cut its traditional ad spending from 2008 and 2009 more than 50 %. The company spent $106 million on measured media in 2008; for the first half of last year it spent $52 million and in this year's first half $20 million. (Source: http://adage.com/article/news/harrah-s-loyalty-program-industry-s-gold-standard/139424/.)

Although some large brands have been using predictive analytics for many years now, it is not too late for other brands, large and small. In fact, predictive marketing is only now finding widespread adoption in medium and small organizations. A good example of a company that has achieved significant success with predictive marketing is Mavi, a high-fashion clothing manufacturer and retailer based in Istanbul, Turkey. Mavi is known for its organic denim favored by celebrities and supermodels. Mavi operates over 350 multinational stores and sales channels in the United States, Canada, Australia, Turkey, and 10 European countries.

Mavi started with a single predictive marketing campaign six years ago. When Mavi first got started, each department, including marketing and IT, used its own set of marketing reports and customer data, including key performance indicators. This led to cumbersome cross-referencing and impeded important decision making. Like many companies, the Mavi marketing team initially did not have access to customer data without relying on IT resources. This was the first problem that the team tackled. Mavi deployed a modern, cloud-based predictive marketing solution in 2009. This allowed the company to consolidate, cleanse, and de-dupe their customer data on a daily basis. They were then ready to start using data in hyperpersonalized campaigns.

One of the first predictive marketing programs that Mavi tested was a program around specific buying personas. Mavi used predictive analytics to find groups of people with distinct product preferences. In predictive lingo these are called product-based clusters. Mavi found at least three very different groups of shoppers: customers who favored mostly woven shirts, others who favored beachwear, whereas a third persona mostly shopped for new season high fashion and accessories. Mavi started to use these personas to implement more targeted marketing campaigns via email and short message service (SMS). Specifically, it implemented a reengagement campaign for lapsed customers that featured the right types of products with the right customers. Using these clusters, Mavi was able to reactivate 20 percent of lapsed customers. This was a big breakthrough because every customer saved or reactivated reduces Mavi's need to acquire new customers.

Mavi today is running more than 80 different predictive marketing programs in a year. Collectively, these campaigns helped add 7 percentage points to Mavi's overall revenues in the first few years, which is a huge sum on a dollars and cents basis. Wikipedia reports that Mavi revenues in 2014 were $747 million, so that would be an incremental $52 million. Mavi is still finding new ways to increase customer lifetime value, and with every campaign launched this number is pushed up higher.

Elif Oner, Mavi's head of customer relationship management, recommends all marketers get started with predictive marketing. She says: “Start small and pick just one program and build on that success.” Elif is also the CFO's favorite marketer. Every dollar she spends in marketing, every discount she gives, is accounted for, tested, and optimized. The CIO Bulent Dursun also played an important role in realizing the potential of analytics and was a key supporter, which made the approach successful.