1.2 What is Big Data? - Video Tutorials & Practice Problems
Video duration:
15m
Play a video:
Video transcript
<v ->All right, so you've probably heard of big data.</v> Why are we talking about this in an AI class? Well, 'cause the fuel of AI is big data. And so I'm sure as marketers, when you're thinking about what big data means, you're thinking about email, about photos, Twitter, so these are all things you might be thinking about and say, boy, yeah, that is a lot of data. Well, I gotta tell you, that ain't nothing. What's really starting to happen is think about omnipresent cameras and sensors that are on every machine and every item in the supply chain. So here's what the interesting thing is to know about big data, whatever you think big data is now, when you look back in a few years, you say, oh, now we have big data. Back then, it was nothing. So what's happening is it's just increasing constantly. And so IBM came out with a statistic that said 98% of the world's data was created in the last two years. And so that seems amazing. That goes all the way back to Gutenberg's Bible. But 98% in the last two years, but the question is five years from now, will 98% of the world's data have been created in the last two months? That's really what we wanna ask ourselves because that's how fast it's increasing. The pace at which data is increasing is just unbelievable. Now, you might ask, what is big data? Well, there's really five characteristics. The five V's. See, I'm a consultant, so I have to come up with some catchy way to explain it. So let's look at the five V's. So volume, well, that's the big part. I bet you knew that one. Velocity is one of the ones we were just talking about, which is that the data is coming in so fast. It changes every minute. It's coming in second after second. Variety. Variety is actually one of the hard ones because the type of data keeps changing. A few years ago, you didn't care about having likes or shares 'cause those things didn't exist. And so every time the world changes, the data changes too. Variability, the numbers go up and down seemingly at random and what's happening is that the data keeps changing, not even in the type, but just in the statistics themselves. But the last one is actually the most important one for you, the marketer. It's veracity 'cause none of the rest of this makes any difference if you can't trust the accuracy of the data. And this is the thing that for the foreseeable future, I can't come up with any way that a computer is gonna know if the data has veracity, if the data is actually true. That's what people have to do. That's what you have to do. And so don't get lost in all of this computing stuff because human beings still have a very big role to play in knowing if the data is accurate, in knowing what you can decide based on that data. What can we safely conclude based on this data? That's where you come in. So marketing data has exploded. So if you go back to say 1995, there were maybe 1,000 people in the United States who figured out what was the news on television. And if you extrapolated it to the world, it wasn't that many more, maybe 5,000, 10,000. It was a pretty small number of people. And if there was a story that came out that was a bad story for your company, you had days to respond to it. Some of you might remember the Tylenol scare that happened back in the '90s where some madman was putting poison in Tylenol bottles and some people died. And what Tylenol did is they were lauded for their response it. Tylenol pulled all of the bottles off the shelves because they had no idea what was happening, but they said the safety of our customers is more important than our profits. And everybody said Tylenol, what a great product, what a wonderful company. Do you know how long it took for them to do that? A week. Can you imagine any company spending a week making that decision now? They would be pilloried as being like the most unfeeling, uncaring company ever because that's how fast people expect you to respond. If you fast forward only 10 years from 1995 when blogs came out in 2005, there were about 100,000 people who decided what was newsworthy and response time went to hours where you really had to answer things right away or else people thought you didn't care. What happened 10 years after that? Social networks. And now there were a billion people who were deciding what's newsworthy. In fact, if you turn on your local news, half the time, one of the stories is somebody who took a video on the subway of some crazy thing that happened. That was never gonna be news back in the 1980s, but now it is because somebody has a video of it. And if you do something wrong as a company, you better answer right away. And that's really what's going on. And so you might notice from this that you can't keep speeding up any more than this. This is as fast as we can go. We're gonna run out of people and we're gonna run out of time. So this is really what's happened in marketing over the last 20, 25 years. And so why is this happening? Why is there so much data? Why is it moving so fast? Well, it's because of something called Moore's Law. Now I have to tell you, this is the single most boring slide that will ever be presented to a marketing audience. Moore's Law says transistor density doubles every two years and costs go down by half. And so I'm sure as a marketer, you could immediately put that to work. All right, maybe you can't. All right, so let me explain what it really means. What it means is that technology is getting so much more powerful and so much cheaper, there's no sense in anybody throwing away data anymore. Anything you can collect, we will collect. And so if you could have amassed eight gigabytes of memory back in 1970, the same amount that's on a $2 memory stick today, it might've cost you $3 billion. That's three billion. And so if you're asking yourself, how come I'm not keeping up with all these trends? How come computers are moving so fast that I feel like I'm falling behind? It's because human evolution never prepared you for this type of change. This is crazy type of change. Three billion to two. Now, honestly, I did play a little trick on you there. It's actually not three billion. If you count inflation, it's actually 20 billion, so 20 billion to two. And so don't worry that you're not keeping up. Nobody's keeping up. This is a change that's very hard to really respond to. It's something that you shouldn't try to think about keeping up. What you wanna do is to just adapt as things change. And so what this might cause you to think, this Moore's Law, is that a $300 iPad will in five years cost $300. Well, that isn't actually what's going to happen. Instead, what's gonna happen is we're gonna have iPad power all over our bodies. We're gonna have all sorts of Fitbit stuff on our wrists. We're gonna have iWatches. We're gonna have these Google Glass stuff. And you might say to me, well, no, I'm never gonna be that guy. That guy who's taking videos all the time while walking around. He's just creepy. That's not gonna be me. And I'm gonna tell you, you know what? It really is gonna be you. You know how I know that? Because if I went back 20 years or 30 years and I said to you, I have this device that you're gonna carry with you all the time day or night and it's gonna be the first thing you look at when you wake up in the morning and the last thing you look at before you go to bed at night, and this device is gonna be so important to you that if you get halfway to work and you realize you forgot it, you're gonna go home and get it even if you might not have gone back for your wallet or your purse. And here's what this device does. Let me tell you. I'll explain it to you. It lets your boss call you any time of the day or night when you're on vacation and the boss knows that you can pick up the call. There's no answering machine. There's no I was out of my house. There's nothing. The boss knows they can reach you. And you are gonna carry this device all the time. And you would have said to me, "You're nuts. I'm never going to do that." But we all know that we do. And the reason is because we can all think of the horrible things that technology brings. We get worried about changes and what technology is gonna do, but we have a very bad imagination for what the good things are that technology can do. You might find that everybody wants to wear these Google Glass-type objects, and everybody is taking video of everything that's going on because it's actually considered rude that you would run into somebody that you met two years ago at the trade show and you didn't even care enough to have a device that would you remember their name. We don't know how society is going to evolve. We keep evolving, our societal norms based on what technology can do. And you can expect that if the technology can do it, we're gonna figure out a way to make something good come out of it. Now you can also be sure that we're gonna figure out bad things to come out of it too. I'm not sure who the person was who said it, but somebody said, when you invent the ship, you invent the shipwreck. And that is always true of technology. There's always gonna be bad stuff that comes out of it and we'll worry about, but good stuff comes out of it as well. So let's look at some examples of data that AI can work with in marketing. So what does Google knows about us? Well, they know who we are, what we're interested in, what we buy, who our friends are, where we go. And you might say, well, no, it doesn't know those things. Well, yeah, it really does 'cause you use Gmail so it knows who you are. Because you're signed in to Gmail all the time, it also knows what you're searching in when you search in Google 'cause you're signed in. It knows what you buy because maybe you're using Google Pay or maybe it knows because Google Analytics is on the website that you went to. It knows who your friends are and you would say no, how would it know who my friends are? It doesn't know what's in Facebook. No, but it knows your email. It knows all the people you send email to. And it knows where you go because when did you not use Google Maps? And so Google knows a lot about you. Amazon knows a lot about you too. It knows where you came from before you went to the Amazon site. Knows what products you looked at already, what ads you saw. It knows your previous purchases. Hey, if you've enabled 1-Click, it even knows your credit card. It knows your address. It knows all sorts of stuff about you. How about what the mall knows about us? And you're like, the mall? The mall doesn't know anything about me. I'm just walking through. It doesn't know anything. Well, it might know more than you think. You know when you're on Facebook and Facebook can tag you in a picture because it knows it's you? Well, guess what? That means computers know how to tag you and they know who you are. Now, the only reason that a mall doesn't do this yet is because it takes too much time. It's too slow. But remember what Moore's Law does? Moore's Law says that all these things that computers can do will be sped up faster and faster and faster. And so it can use facial recognition to say, hey, what's your gender? What's your age? What's your ethnicity? And maybe even your identity eventually. And so there are things in the mall right now that when you're walking by one of these ads, it might recognize that you are a 28-year-old woman and it's going to put up a Sephora ad. Now, it doesn't necessarily know that it's you yet, but remember the Facebook example, it's gonna know you're you pretty soon. Now, I wanna ask you, you start to get a little creeped out about this stuff 'cause most people would be creeped out at this point, but maybe not, maybe you're an intrepid soul, and so let's talk about what the store knows about you when you're in the mall and you walk into that store. So what happens when you go into the store? Well, you might ask yourself, hey, is your phone set to switch to the free wifi? And you might say, I don't know. Well, if you think you don't know, then the answer is probably yes because that's the default on most phones. And so if it switches to free wifi, you say, hey, cool. I wanna get off my data plan. I'm running out of data at the end of the month. Cool, this store has wifi. Let me switch to it. And what do you think is happening in the store? Whose wifi do you think it is? That's right. It's the store's. And so what's going on in the back of the store? Well, new customer in the TV aisle. Hmm. There's a person actually paying attention to where people are in the store and what they're looking at. And then you say, hey, wow, this is weird. This store just sent me a coupon. Why did they do that? Because they're overstocked. What were the odds that the store would send you a coupon? Let me tell you it was 100%. And so this is what's starting to happen now. These systems are in stores today and you might say to yourself, okay, well, I see the price here. I wonder if I can get it cheaper online. This does look like the one I want. And what do you think is happening in the store? Customer checking Amazon. Oh boy, now what's happening? The next thing that happens is someone comes over and says, "Can I help you with something?" Isn't 'cause she really cares about helping you. She's just trying to interrupt you from what you're doing with Amazon and she's gonna be there to offer you a deal. So these systems are in stores today. Here's a system that is not available today, but it's coming soon. So what about when you buy something like a carton of milk, what kind of technology could possibly be involved in buying a carton of milk? Well, there could be a lot. Now, if you're one of those people that loves to buy local so that you buy local produce, you kind of know where your food's coming from, here's another way of figuring out where your food is coming from. They can put something called an RFID tag. You can see that in the picture. See how small it is? What it does is it can store a little bit of information in and it can be picked up by scanners as it goes by. So when the cow is milked, they can put an RFID tag in the milk jug and then a new RFID tag when the milk is placed in the carton. And with sensors, they can actually, as the dairy ships the carton to the store, they can know the cow wasn't given antibiotics. The cow was milked 11 days ago and the highest temperature in transit was 42 degrees Fahrenheit. And so you can take your phone and you can hold it up to the carton and get all of this information. This doesn't exist now, but it will soon because right now, RFID tags cost about two cents, so they're a little too expensive to put on every milk carton just to get thrown away. But in a few years, they're gonna cost 0.2 cents. And guess what? That's what's going to happen. So this is your introduction to big data. I hope this was helpful for you. We'll see you back here soon.