Like to welcome everyone to the guts and heart of our lecture on applications of business analytics in marketing. I always like to teach stuff right from the front slide. And if you notice here, the slide says underneath the future of marketing and then in parenthesis, science. And the reason I say this is because to me, the future of marketing is business analytics. There's no firm today that should be thinking purely mass marketing, there should be no firm today that isn't thinking about individual level customers. There isn't any firm today that shouldn't be using technology to measure their customers better and the reason in parentheses science is now that you can measure stuff, marketing really has become a science. Meaning, we have data. We talked about that in my introduction, that's kind of the first step, what data have you captured? We have ways to explore that data. There's ways to explore massive datasets today. Third thing, we have models that we can build. We have ways we can optimize and then we can make business decisions. To me, that sounds like a science. And so again, the focus of this lecture will be on applications of business analytics in marketing. The first, if you like question I always like to ask is how do you compute corporate profits? Now you might say, well, that's not a hard question. Revenues minus cost, that's corporate profits. And I say, sort of, I mean, yes from a finance perspective that might be right, but not from a marketing analytics perspective. From a marketing analytics perspective, you should be thinking not of, if you like, imagine a matrix where the rows are your customers and the columns are the products they buy and the entries of the matrix are how much did they spend on a given product. And of course, some of them can be zeroes, they didn't buy any of the products. But if you wanna think about it that way most people say, I'll just add up the columns. I'll see how much each product makes. I'll see what it costs to make that product, and I'll compute corporate profits and a marketing view of business analytics are just the opposite. The columns aren't your profit centers. The products aren't your profit centers, but the rows are your profit centers. You should be thinking of each customer. How much should I sell to each customer? What does it cost me to serve that customer? How much profit do I make from each customer? And then I add up the rows. And hopefully, all of you guys remember, it was probably eighth grade math for most of us. Whether you add up a table of numbers from a columns or the rows, the sum is still the same. So the classic finance way of thinking about profits is revenue minus cost. The classic business analytics and marketing way of thinking about profits is how much revenue minus cost do I get from each customer and let's add it up over all the customers. Now the advantage of thinking of marketing analytics that way is that some customers have a negative value. What should you do? You fire those customers. Some customers have a lot of value to the firm, those are the ones you have to not only cultivate and retain, but you have to find tons more of them. So even just on this first substantive slide, you can see in a way how the mindset of marketing analytics works. It works one customer at a time. So again, make profit one customer at a time. And that really is gonna be the underlying theme for the rest of these slides, I'm gonna tell you how leading firms like Google and Amazon, etc. Think about making one profit one customer at a time and how really technology has changed that. So if you'd like, it's really the intersection of technology, data and statistics that have made business analytics and marketing kind of the science that it is today. So let me just start with what is customer analytics? And it might seem like an obvious thing for many of you, it took us actually seven years to write this definition. And you might say, wait, you've been doing this for 20 years and it took you 7 years to write what this definition is. And the answer is yes, that is how long it took. But if you look at what we've said, customer analytics refers to the collection. So first of all, you have to have data and you have to collect it. As I said in my introduction, if you don't measure it, it's almost like it didn't happen. Second is the management of the data. So there has to be someone that can collect these data into large databases. And whether you store it in the cloud or through to a server or locally, someone has to manage the data. You have to analyze the data. Data in itself is data. Data doesn't give you direct insights, you have to analyze the data. And then finally, it's the strategic leverage of that data. So customer analytics refers tot he collection, management, analysis and strategic leverage of an organization's granular data about the behavior of its customers. The other part that's key is the granular nature of what I just talked about, which is again, going back to the though about how you compute profits analytics. Business analytics. Customer analytics from a marketing perspective is about granular data. Track every customer at the most granular level. Next, you can see, again, inherently granular. It focuses on individual level behavior, not aggregate patterns. So the example that I love to use all the time is if a firm wants to predict sales in some state, very interesting, very important business problem. But that's not customer analytics, you wanna compute it one customer at a time. Secondly, it's behavioral. The thing that's really changed 20 years ago when I started at DuPont and I was working in their customer marketing division. When we wanted to know something about customers, we surveyed them, we asked them what they did. But now, you can measure what they did. And the old marketing expression is what predicts behavior in the future really well? Behavior in the past. Behavior predicts behavior a lot more than what is called stated intention data fits behavior. Third, it's forward looking. I spoke about this in the introduction when I gave that old marketing joke. Prediction's only hard when it's about the future, but business analytics for marketing is a forward looking science. What are my customers going to do next? Next, it's multi-platform. So it's not just about what you can measure on the web, it's integration of online, offline, survey data, etc. If you'd like, the language you'll hear all the time out there is the problem of data fusion. How do I fuse together data from different platforms? All of that is part of marketing customer analytics. It's broadly applicable. So notice the word customer is in quotes. So for example, if I'm Expedia, then a customer could be someone that goes to my website and is looking for a trip. If I'm the American Red Cross, my customer might be someone that donates to either a charity or someone that donates blood. If I'm a Pfizer like a pharmaceutical company, my customer might be a physician or might be a consumer, direct to consumer. So think of customer in the most broad way. And finally, it's multidisciplinary. So if you're watching this Coursera course and you're in a field of marketing, statistics, computer science, information science, operations research. It's applicable to every industry vertical. It's applicable to retail and pharmaceutical, telecom, etc. The nice thing about the application of customer analytics is that it knows no industry boundaries and it really is cross-disciplinary. Just to say a few things for those of you who wanna learn more about customer analytics in marketing besides just it's application in this course, you can go to our website, wcai.wharton.upenn.edu. This is a center my colleague Pete Fader and I co-director on customer analytics and you can take a look at how we've applied customer analytics to many different industries and problems
