Hello there, welcome back. Just to recap what we did before this, we went through a motivating example that somehow in some sense or to tie data and hence, analytics with enterprise value. And where we're going now, would we talk over some basic conceptual preliminaries. So, just to be clear on what I mean by a conceptual preliminaries, there are fundamental terms, things that you and I know about, but for completeness sake, we make sure that we are on the same page when we say these terms, that we both mean the same thing. Okay. Basic Conceptual Preliminaries. Now, this is a course in Business Analytics, sure we all know that. That brings up the question, "What is a business?" And the associated question, "What is Analytics?" What is the nature of analytics? Is it art? Is it science? And we will see all of that. You will see some of these things right now. The Anatomy of a Business which in some sense goes to answer the question, "What is a business in the first place?" There are two fundamental axioms in economics, in micro econ, microeconomics. And in case you are wondering what is an axiom, an axiom is a statement that can neither be proved or disproved. So, we take it to be true. So for instance, two parallel lines never meet, is an axiom. Taken to be true, there is no proof per say, all the proofs in some sense derived from it and Mathematics and Euclidean Geometry. In Economics, there are a couple of axioms which we take to be true. There are these assumptions which we take to be true and the rest of the entire foundation of micro econ is built on these axioms. The first axioms says that firms exist to maximize profits. There is no proof for it, we take it to be true. It is fundamental in the sense that if it were not true, everything else that we would be doing later in this course will not really matter. Okay. And in case you're wondering what the second axiom is, the second axiom says, that customers, consumers exist to maximize utility. Take these two together and you can build the entire edifice of economics, starting from there. Okay, now look at that word there, profits. What is profit? A simple one line definition of what is profit would be, profit is equal to revenue minus cost. Now, I want you to look at this cost piece here, right? What are the different cost heads in a business? What kind of costs could be there? For instance, you could to have operational costs, cost of materials and cost of maintenance, cost of labor. Minimizing operational costs and thereby maximizing firm profit is the domain of OM, operations management, that's basically what they do. Take a look at the cost of capital, minimizing in some sense capital costs, and thereby maximizing profit would be primarily the responsibility of the corporate finance people. Minimizing regulatory costs would be in the accounting domain. Then there are HR and the other functions that you have. If you look at the other side on revenue, well before I go there, taken together all these costs on the supply side, all these costs, functional costs on the supply side would be basically what economists call, the supply side of the profit equation. If you look at the other part revenue question then is. Well, this is the domain of market, maximizing profit by maximizing revenue would be the primary responsibility in some sense. So, this would be the demand side of the profit equation. Business functions represent a logical way, but only one way in which we can view the business enterprise. There are other ways to look at a business enterprise. We are going to follow this because basically, it helps in what we're going to do later on. Business functions would also yield analytics that are function-specific. So, you have people analytics, for instance, in the HR side. The marketing side, you have customer analytics, you have campaign analytics, and so on. So, you have all of these things also coming into the picture. The market power can derive from competencies either on the demand side or on the supply side. A good example is Apple. World class competency is on the demand side. People want to buy their products, which is why on the demand side, they are able to, in some sense, deploy competencies which are world class. On the supply side, a good example will be commodity companies. So, take a Reliance Industries Limited, which is basically the lowest cost crude producer, refined crude producer in the world. Take a Riotinto, take a Hindalco, the lowest cost aluminium producer in the world, one of the lowest cost producers. So, when you have competencies like that on the supply side, then well really, the demand side in some sense becomes secondary to your core competency to your focus. Which brings me to my second conceptual preliminary, "what is analytics?" Well, let me explain this through an example. Through basically a representation picture or a diagram. There is a real world system out there. Okay, there are real businesses and there are real problems in real businesses in there, and questions that need answering. Now, this real world system has, in some sense, or that is there, where you want to go, is a real world conclusion. This real world answers to your real world questions. Ideally, we'd be able to get from there to there in one shot, and it turns out there is a tool that allows you to do that. Experimentation, in some sense, within certain limitations and we will start the causal research experimentation later in the course, we actually see what it is. Allow you to get from there to there. However, there are instances where you can not go directly from a to b, and then we have to take or detour, things become a little complicated. They also become quite interesting. So, we have a mathematical system which we model based on the real world system. It's an abstraction, perhaps. So, we are modeling the real world inside a mathematical model. And from that mathematical system, I'm going to basically reach mathematical conclusions, answers to the real world questions that you had originally asked me. That pathway is what we could call analytics. This predictive analytics that would tell me how would the real world system behave if this were to happen. There's diagnostic analytics, which in some sense, would tell me, what it is that did go wrong in the past? And so on. So, all of that is possible. And once you reach the mathematical conclusion piece, from there we have to well close the loop and get to in some sense real world conclusions and that last part, as we what call interpretation, critically important. So, analytics sits between these two, abstraction and interpretation. Analytics brings the two together. It, in some sense, is going to answer real world questions by creating a simulation, a model of the real world within itself, and answer those questions. So, one way to look at analytics would fairly broad, in some sense, overarching way to do so, which brings up the third related question. Look scientific indeed, mathematical representation, and so on. Hold on to that thought. Is analytic scientific? Is it an art? Is it a science? What is going on here? Well, we can look at science, a broad classification of the sciences could be, in some sense, into the natural sciences to think about Physics, Chemistry, the Life Sciences, and so on, and the Social sciences. So, that would include things like Sociology, Anthropology, and yes, economics. Which, in some sense, is a parent of management, all of management. Management falls there in the Social Sciences pair. All right, my question to you, what is that fundamental difference between these two branches of science? What distinguishes Natural Sciences from the Social Sciences? If you think about it for a minute, the answer becomes apparent, the answer lies in the unit of analysis employed. The unit of analysis in Natural Sciences is in animate matter. Sure, in the Life Sciences, we're looking at cells and living tissue, but basically, inanimate matter, in the sense that those things don't have free will. Which basically means that an experiment conducted under identical conditions will produce an identical stimulus, will produce an identical response under identical conditions. Which basically means my experiments in the Natural Sciences should be perfectly replicable or close to that. Which means that what I get from there are actually laws of nature. On the other hand, when we look at the Social Sciences, the unit of analysis is the human being or some aggregation there of groups, and so on. And there is free will associated with human behavior. Which basically means that even under identical conditions given an identical stimulus, you may not get an identical response. Which basically means that our experiments will not be replicable. Which basically means that the best we can do here are weak theories of social organization. We don't get Laws of Nature in the Social Sciences. So, where am I going with this? A lot of the analytics that we will do in businesses fall under the social sciences domain, where we are dealing with human beings as the units of analysis, which basically means that there is a limit. There's only so much of precision that we can expect in our measurements, in our modeling, in our results, that we could rightfully expect within a given cost and time. That's basically. A couple of preliminaries more, one is about Market Intelligence versus Market Research. Market intelligence, also called business intelligence in a lot of context, and market research is basically a lot of business research done or research that businesses commission. This is a fundamental dichotomy and I'd like to, in some sense, talk about this. Market intelligence or business intelligence is an ongoing activity which is not tied to a specific project. Whereas market research is tied to a specific project and specific questions. What does that imply? It implies on the business intelligence side. This is about keeping eyes and ears open. This is about an open-ended exploration, whereas on this side, you have to be more deliberate in data gathering. It also means, that in the business intelligence side, what you actually need is the ability to shift through large amounts of data. Remember, this is an ongoing activity, data keeps flowing in, whether you are analyzing it or not. On the other side, it might need expert judgment, and the experts need not always be human as we will see. On the other side, what is actually needed, because it is tied to a specific project and there are specific questions and it is deliberate in data gathering, it is time bound implied. The competency is needed to pull off a good business intelligence, market intelligence, operation database management, and organizational leadership skills. The ability, in some sense, to stick with something that may not yield immediate returns. On the market research side, the core competency required is problem formulation skills. And from there, everything else follows. Why am I putting this up? Because what is going to follow would be problem formulation. Okay, where is this distinction, I mean, are the two really different? Well, that's a good question, partly, because we actually see that those boundaries are blurring. The chances either the two will converge in the near future. They probably have already in certain firms, in certain sectors in industries. What does that mean, driver of this trend? Remember what we did when we saw the anatomy of a business is the demand side and the supply side. The driver of this trend is technology, sure, and the way it is acting, there is acting both on the demand side and on the supply side. Here's a quick example. On the supply side, it acts by basically lowering the cost of analysis. What you see there in that graph is an exponentially declining graph of the cost of a PC. It's a good, but the cost of storage has crashed even more sharply. For instance, the cost of processing has crashed even more sharply than that. On the demand side, it works by basically speeding up diffusion and adoption. Here's an example. There is a time taken by each of these devices to reach 50 million users and it is clear that that time is shrinking the speed of adoption, the speed of diffusion is just going up. Angry Birds took 35 days to reach 50 million users. The telephone, the fixed line telephone took 75 years to reach there. Bottom line, owing to these two pressures, both on the demand and supply side, we may actually see both marketing intelligence, market intelligence, market research coalesce into a third new function, merge, basically. And analytics will be written all over it.
