The explosion in digital media - web, social and now mobile - represents a departure from how things were like in the last century. This proliferation of digital media is both a threat and an opportunity for many businesses. Business Analytics can be leveraged to process data, sentiment, buzz, contacts, context and other aspects of business interest in real time, for business performance and impact. The course picks and uses use-cases from a variety of industries and geographies, to showcase the potential and impact that business analytics done properly (or not) can have on business performance.
The App Era in Digital Social Media
Loading...
来自 Indian School of Business 的课程
Business Analytics and Digital Media
91 个评分
从本节课中
Digital Media
An overview of the big questions, possibilities and challenges.
Learning outcomes by the end of week 4 should normatively be that
(a) Students have an understanding of the major types of digital media in use currently by people and firms, (b) and, develop an appreciation for the types of problem solving, data collection, prediction and optimization that can be enabled using digital media tools.
与讲师见面
Sudhir VoletiAssistant Professor and ISB Research Fellow
Marketing
[MUSIC]
Welcome back, folks.
Social Graphs and Hyper-Segmentation, let's jump right in.
Exploiting the social graph for reach.
Suppose, you want to reach in some sense gardening enthusiasts in India.
Bowling fans, hiking enthusiasts.
How would you go about it?
What would be the size of the segment?
How would you reach them?
Who would have such information?
Take a minute, think about it, type out.
Who do you think?
Which publisher would have such information?
Gardening enthusiast.
Can we go to maybe some gardening magazines?
What proportion of gardening enthusiasts would such a magazine reach?
Who would know how many there really are?
Well, virtual social networks, which didn't exist in the last century,
today basically loom large on our consciousness.
They are there. Who would know who
are the gardening enthusiasts in India?
Are polo fans in India are?
My guess is a Facebook would know.
The answers not just to the about questions, but
it could also leverage our social graph for so much more.
So, what is a social graph?
It's basically a social network.
Who do you know, how well and what do you talk about with them?
Enter hyper-segmentation, individual specific targeting and
there are business implications galore and we will see some of this right away.
Targeting a hyper-segment, why a social graph?
Let me basically show you before I get there my own Facebook network from
a couple of years ago.
Our connections with my engineering class and the ISB community here.
That is my MBA group.
Quite a lot of connection between the engineering group and the MBA group.
That is the PhD group, just a few isolated souls here and there.
That is my family in Uttar Pradesh and this is my wife's family in Sikkim.
You can actually see where my wife is.
There, that red one which connects to everybody else in Sikkim.
So, she's basically the gateway to all of them.
And even within her circle, there are her friends in blue and
then there is a family in red who don't connect on Facebook and
they know each other communities coming through.
That's my high school group.
That's my secondary school group and
there are no connections between them they go on move to city.
Question is you can map your network like this which each of these groups,
what I talk about?
What they will talk about to me?
What I can recommend to them?
What they can recommend to me in terms of products, preferences and so
on is going to be different imagine a publisher actually knowing that.
Imagine the effect it could have, the impact it could have.
Image how much advertisers would pay to know something like that.
Who would recommend a product like this to whom?
Who else is likely to buy and so on.
Targeting a hyper-segment.
An age old question the marketers face is how to get people to pay more for
the same old product?
People who pay x for a birthday cake are willing to pay x plus y for
a personalized birthday cake.
The same reason you can say, tailored suits are much more expensive,
always so than ready made one.
But can you apply something like this to let's say, beverages?
Take a look at that.
You think Christopher is going to be willing to pay a little more for
something like that?
Maybe true and it's not just an every day thing.
If Coke also knew a particular location, maybe a wedding anniversary or
maybe a birthday or maybe some celebration.
Revolving around [INAUDIBLE] code leverage and
get this to print and you would actually see more revenues coming through.
So imagine being able to know the right person, right time,
right place to be able to deploy.
For things in India, for instance.
I'm always on [INAUDIBLE].
If you can actually deploy personalized beverages, people would be willing to pay.
A baby is born in a house.
Hey, welcome.
You could actually have a lot of these services, but what price?
What about the supply side?
Will we be able to scale up and so on?
Will the economists walk out?
It's possible.
Now, we actually have the technology to customize at mass scale and
it's possible to do this.
And if there's profits to be made, then why not?
Who you know matter, let's see why.
Heard of FICO scores?
Basically, what are they?
They are your credit scores in the US.
The General inputs that go into calculating a credit score or
your credit history, or income and all of that education.
What does this have to do with network analytics?
How about your social network?
Will it also in some sense be able to give useful information to
calculating your credit score?
Turns out yes.
SNA, they're standing for Social Network Analysis.
It can actually give very important information.
There you go.
If you are friends with people who have good FICO scores,
the chance is that you also have a good one are going to be just that much higher.
It also means you access resources in case you get into a credit crunch.
It actually yields information that is not there and
other information that you should be going into credit school.
And so, this is going to be reality.
It's already reality in some sense.
The basic principle is homophily.
So, birds of a feather flock together and
high FICO score people are going to flock together?
Likewise, I guess not so high credit call people as well maybe.
I believe that's the assumption going there.
Mapping a social graph, what comes next?
Well, if you go back to the Facebook Social Graph example that I just showed
you a while ago, what was the space that we saw?
It wasn't geographic space, people are spread all over the globe there.
What space was it?
One may say in this kind of mapping, it's social,
personal interaction space, social interaction space.
Is it going to be interesting to marketers and brand managers?
Yes.
However, marketers are going to want to know something else.
People's relative locations and the space of common interest, and
the space of beliefs, and the space of behavior.
Consumption factors, psychographics, affiliations.
In other words, can we map people in affiliation space?
Regression is rhetorical and it is there, because yes, we can.
Let see this.
Mapping brand affiliation for the class of 2015 from the ISB.
So each of those shallow dots there, the pale ones are basically people and
each of the big green ones are brands.
And basically, brands are connected by different liking the same set of brands.
That's how you can see apple has been liked by a lot of people and
you could see it has already high centrality score there.
Yep and so on.
I can remove the pale ones and keep only the green ones,
which is what I do in this one and this is what I get.
What does this show me?
This shows me some degree of affiliation.
It shows me that people who like Coca-Cola also like Dairy Milk.
We don't know why the data are saying that.
People who like Pepsi also like Levi's.
Is there a chance of having some sort of an alliance?
Some sort of co-branding?
Yes and this would be the starting point for that.
People who like all Mongols, also like Nokia.
Social graph, let me quickly recap what we just saw.
We saw social graphs is basically social influence networks.
They include purchase decisions, brand preferences and all of that.
Social graphs plus hyper-segmentation can exploit customization based profit
premiums and these are more example with beverages.
The larger social network platforms are valued the way they are,
because they have as publishers this kind of information.
Right person, right time, right message.
They alone are able to provide it better for now.
There's a reason why Facebook acquired What's App for $19 billion.
They would know what the groupings are?
What the graph is What are the conversations within the graph?
We also wandered slightly into more interesting brand affiliation spaces, but
just briefly.
Scratched the surface of network analytics.
Admittedly we didn't go there at all, but at least got to see what it might be like.
[MUSIC]
