Data about our browsing and buying patterns are everywhere. From credit card transactions and online shopping carts, to customer loyalty programs and user-generated ratings/reviews, there is a staggering amount of data that can be used to describe our past buying behaviors, predict future ones, and prescribe new ways to influence future purchasing decisions. In this brand new course, four of Wharton’s top marketing professors will dive deeper into the key areas of customer analytics: descriptive analytics, predictive analytics, prescriptive analytics, and their application to real-world business practices including Amazon, Google, and Starbucks to name a few. By the end of this course, you’ll understand how data is used to describe, explain, and predict customer behavior; best practices for using this data to prescribe more effective business strategies; and how successful companies are currently using such data to meet (and shape) customer needs, and to increase their bottom line in the process. This course has been designed to help you make better business decisions about emerging data structures, so that you can apply what you’ve learned to your own company or business today.
Na lição
Descriptive Analytics
In this module, you’ll learn what data can and can’t describe about customer behavior as well as the most effective methods for collecting data and deciding what it means. You’ll understand the critical difference between data which describes a causal relationship and data which describes a correlative one as you explore the synergy between data and decisions, including the principles for systematically collecting and interpreting data to make better business decisions. You’ll also learn how data is used to explore a problem or question, and how to use that data to create products, marketing campaigns, and other strategies. By the end of this module, you’ll have a solid understanding of effective data collection and interpretation so that you can use the right data to make the right decision for your company or business.
Professor of Marketing, Statistics, and Education, Chairperson, Wharton Marketing Department, Vice Dean and Director, Wharton Doctoral Program, Co-Director, Wharton Customer Analytics Initiative The Wharton School
Peter Fader
Professor of Marketing and Co-Director of the Wharton Customer Analytics Initiative The Wharton School
Raghu Iyengar
Associate Professor of Marketing The Wharton School
Ron Berman
Assistant Professor of Marketing The Wharton School
So now from Point of Sales data, let's go to Media Planning.
Now I'm going to give you some examples of how companies can go ahead and collect
some data, sometimes themselves and sometimes by the aid of other companies.
To get a sense of audience engagement.
So, in radio for example,
there are companies out there that can collect this type of data.
So, Kantar Media, for example,
collects a lot of data in terms of, who's listening to different radio shows,
how is the popularity of different stations, and so on.
There are other companies as well that collect this type of data, for
example Nielson Audio.
Again, many companies out there try to get a sense of
audience engagement at the radio stations.
You can think about these kinds of engagement, even at the TV level.
So, here, Nielsen collects a lot of data which is from the TV level.
How do they do that?
With a set-top box.
So, this is a box which will go along with your cable box.
What are they trying to do here?
They're trying to basically record who are the people who are watching a particular
TV station, who's in the family, and
then they aggregate different households to get a sense of
how popular is a particular TV show, how popular different channels, and so on.
So again, Nielsen is not the only company doing this,
there are many other companies as well.
And other companies are following the example.
Again, Grand Track is a company that basically collects information, again,
about TV engagement, with slightly different ways of doling it,
but the overall idea is exactly the same.
What would we want to do, in these kinds of issues, TV or
radio, get a sense of how popular is a particular show, or particular channel.
What are the kinds of questions that can be answered
with these kinds of data that's available?
So for example, who is watching what show?
As a brand manager, or
in fact as a TV station manager, we would like to know which shows are popular.
Which shows potentially can be funded next year?
How is the viewership pattern changing over time?
And for example, if you start thinking about viewership pattern, how should I be
spending different kinds of money on different types of ads in these TV shows.
So, broadly, these kinds of questions can be answered by getting information about
audience engagement, and that's what companies like Nielsen,
companies like Kantar Media, that's what they're doing when they collecting
the different types of data, either through radios shows or for TV shows.
Now, we'll talk about social media analytics.
So social media is becoming extremely popular.
There are many, many people on Facebook, Twitter, you name the different formats,
people are there talking about brands, talking to each other and so on.
Clearly, when you think about it yourself as a brand manager, you start
thinking about what are customers talking about You need data about social media.
And there are companies out there that will help you get that data.
For example Hoodsuite is a company that does a lot of information and
data collection on things going on Facebook or Twitter.
So it's a platform that collects that data.
Sprout Social is another example.
So there are many companies out there that are collecting information on what
customers are saying and doing on social media.
So what are the kinds of questions that a manager can answer
using this type of data?
The first one is just audience engagement.
If I'm running a campaign on Facebook.
If I'm running a campaign on Twitter.
How many people are responding?
That gives me a sense on how many people are responding.
That gives me a sense on how good that campaign is.
Another example might be more along the lines of thinking about brand mentions.
Is our branding mentioned more times than our competitors?
What's our shared voice?
So for example, as we just saw, we could track, for instance,
how many times Wharton came up in the last month on Twitter.
You can start doing the search for
your own brand and see how it compares to your competitors.
You could also start thinking of doing sentiment analysis.
So it's not just about mentions and how is it being mentioned.
Is it being mentioned in a positive way?
Is it being mentioned in a negative way?
How is it being mentioned?
And then you can start thinking about thinking, for
instance, what might be going on in the economy, or different kinds of issues
together with dimensions to see how they might change over time.
So social media data is very very powerful.
There are companies out there that can help you collect this type of data to see
how your company's trending on social media as compared to your competitors.
So now we'll talk about another type of data.
This is broadly categorized as web data.
This will be basically all of the different kind of searches you're doing on
the internet, all the different web sites that people might be visiting, and so on.
Of course there's a of information out there.
First thing is you can start looking at their own company website and
seeing what customers are looking at, but at the same time you want to learn what is
happening in the marketplace and that's where these other companies are coming at.
For example Compete.com is a website that will allow you to look at what
people are looking at at different websites.
There are of course many other companies that do the same thing.
For instance, Comscore is another company that collects this type of data as well.
So are many other companies.
This is another example of lots of other companies that
basically collect information on what are people searching for
with websites they're visiting and so on.
What does a data steam look like?
Again, I went to compete.com and looked for
the type of searches and type of people who are coming in on the Wharton website.
That's what it looks like in the last month.
So it gives you information, on for instance, unique visitors and so on.
So besides of course, going out and looking at company that
collect relative information, you can have data from your own website.
For instance, as an example, etsy.com is a company where you can get a lot of
information, for instance, on what are people looking at.
So Etsy has this data, we can look at what people are searching for and
get a sense of what products are becoming popular, what websites, for example,
are becoming popular, and so on.
So electively, web data gives you a lot of information
in terms of what your customers are looking at, either on your own website or
a competitor's website and so on.
Broadly, this type of data helps you answer
questions which are very managerially oriented.
You can start thinking about any kind of campaign.
You can broadly categorize the type of spending that you're doing in terms of
earned media, paid media, and owned media.
Owned media basically is your own website.
Paid media is any kind of engagement that you're doing in terms of
paying for that media.
And earned media are people who are organically coming to you.
If you start thinking about collecting this type of web data, it
gives you a sense of what are the ROI for many kinds of campaigns that you're doing.
So again, the type of data and
the type of questions you are trying to answer have to go hand and hand.
Finally of course, when you start thinking about the Holy Grail, it's mobile data.
That's where many, many companies are making Androids.
Think about Facebook.
Of course, it started off as a desktop, but
many more people are accessing Facebook using their mobile phones.
Facebook has this data.
They know where people are accessing Facebook, and
they know the location as well.
That's a wealth of information.
As another example, there is Foursquare.
Again, this is a website in which people check in,
usually through their mobile phone.
Foursquare, again has a lot of information on where people are,
where they're checking in, what else is around.
So, many, many such companies are basically, collecting this information.
On the flip side, there are companies such as, Flurry Analytics.
What do they do?
They get a sense of, hey, how's your app doing, in terms of overall customer
engagement and they can also start helping you monetize the app and make it better
in terms of people accessing it easily and trying to make more money out of that.
What are the type of managerial questions that can be answered?
For example,
is customer search on the mobile platform different from the desktop?
What information to show customers based on their location?
So think about Foursquare for instance.
They have information of where you are.
Should the information they show be different depending up on your location?
Probably yes.
And of course, another type of question would be,
are there coupons that can be sent based on your location?
All of these questions can be answered by looking at mobile data.
And again, there are companies out there that can help you collect this data.
Explore nosso catálogo
Registre-se gratuitamente e obtenha recomendações, atualizações e ofertas personalizadas.
Eric BradlowProfessor of Marketing, Statistics, and Education, Chairperson, Wharton Marketing Department, Vice Dean and Director, Wharton Doctoral Program, Co-Director, Wharton Customer Analytics Initiative
Peter FaderProfessor of Marketing and Co-Director of the Wharton Customer Analytics Initiative
Raghu IyengarAssociate Professor of Marketing
Ron BermanAssistant Professor of Marketing
