free_LUNCH

When taking THE ‘Market Intelligence’ course here at Goizueta, we came across a data set that contained lots of missing values. Yes, I know that most of the data sets you see out there have some missing data or the other. But when I started looking for material on how to deal with these data points (other than simply discarding them) I was frustrated not to find anything for a manager-type audience. Most of the discussion was for statisticians. So, based on what I heard in class, here is a flow diagram that should help. Note that there are some acronyms used, but you can look them up and find out what they mean. They are not difficult.

1. Various Statistical packages have automated functionality to achieve this.
2. Various options available to impute data. Use business judgement.

Optional

• Run analysis with imputed data and dropped cases
• Check for any significant difference in the models
• Provide necessary details of imputation when presenting analysis

All “risks” have some “uncertainties”, but not the other way around. A good example from Wikipedia is:

We can be uncertain about the winner of a contest, but unless we have some personal stake in it, we have no risk. If we bet money on the outcome of the contest, then we have a risk. In both cases there are more than one outcome.

As I finish my “SDA Application Portfolio” series I think about the utility of the knowledge gained in this class. It was one of those classes where you are not taught how financial instruments work, how brand management is done, nor how profitability is calculated. There were no formulas to memorize, no equations to be remembered. It did not go into the world of marketers or traders, it did not made me analyze a country’s economic growth, it did not teach me how to increase ROI. Yet, it taught me all that and more.

You may ask – What was this course? How does it fit into the MBA curriculum? How will it help you in the long run? Answers will follow.

This course was called “Strategic Decision Analysis”. It was a course that harnessed the power of analysis and thinking in order to take decisions and increase your chances of success. The course introduced many concepts such as Negotiations, Prisoner’s Dilemma, Voting, and Auctions, among others. But more than that, this course provided insight into how a human mind works and what it thinks. You don’t need to be a CFA, a Six Sigma black belt holder, or a Statistics major to understand all this. All you need is common sense thinking and more importantly what the other person is thinking.

It fits well into an MBA curriculum because it is expected of us to go out and take decisions. Difficult decisions. Decisions that will affect companies, people, nations. Decisions that will alter the way you live, you eat, you sleep, and all that in between. How can you take such decisions and maximize your chances to take the right kind of decisions? Note that when you are taking the decisions there is always someone else taking a similar type of decision. You have one goal – to win in this duel of decision making. Because if you don’t win, somebody else takes all the glory or the pain associated with the decision. This course, in part, was about maximizing your chances to win.

With knowledge comes great power. But also comes with it the “curse of knowledge”. It is dangerous and can be applied dangerously. An incomplete assessment of your own understanding of the knowledge can actually lead to disaster. So it becomes important that we become competent in the use of this powerful knowledge and how we apply it.

The reason I am stressing this is because the knowledge is not just limited to the world of business. Your understanding of six sigma will not make you a great husband, a great father. But this knowledge is different. It spans our daily life and our relationships. Thus it becomes more important to understand the subject matter closely and intimately. The most important piece of this knowledge is – knowing what the other person is thinking. Not what you know and think. Bluntly speaking – you need to drop your ego from the equation to apply this knowledge. You have to stop force fitting because there are ample opportunities to do so. One formula to apply this knowledge is this –> First, you need to drop what you know. Second, know what the other person knows. Third, apply what you know and what the other person does not know. What you know is what you have learnt from this course, from your observations, and from the uncommon common sense that has been bestowed upon you.

Final thought – Always remember – Cooperation is better than defection! It will become all too apparent for you to defect with this knowledge. With this toolkit in hand, you will take decisions that go against the grain of this thought, that of cooperation. Recognize such tendencies, observe your thought process, re-evaluate the knowledge, and then apply it again.

I was hungry the other night and decided to order pizza. When I arrived at the Domino’s site, I was greeted by the graphic above. And a bulb went on.

Fairness principle applied at its best – Order any three (or more) items for $5.55 each. For a family or a group of friends this is a great deal provided we need fairness. Lets look at the distributive principles at work here:

• Parity – Every member is free to choose what they want. Difficult to be jealous of one another.
• Proportionality – 1 order per person or if more than more, proportionally higher price paid.
• Priority – Not applicable in this scenario. Any person can choose any item in any given order.

We, as in SDA graduates at Goizueta, have studied “negotiations” and “auctions” to a certain degree. Some of us have even had advanced training in negotiations. Now Harvard professor Guhan Subramanian has come up with Negotiauctions. According to him there is a strong interplay between negotiations and auctions in real life situations. I certainly felt the truth in his statement, based on my experience in the various assignments during the course duration, and am looking forward to get my hands on the book during my MBA so that I have the time to read and ponder over his thoughts.

On December 9, 2009 (i.e. today) Indian government agreed to create a new state. It will be called Telangana and will be carved out of the current state of Andhra Pradesh (see right image – Telangana is marked in white). The division had been called on by Telangana supporter for a very long time and most recently activities had become quite volatile. In fact, the leader of the movement has been on an indefinite hunger strike for the past 10 days. Finally the government had to give in and accept the demand.

But what is the reason for this uproar? Why do some people want a separate state and not be a part of the larger whole? Haven’t we read than co-operation is better for all players involved? The answer lies in the fairness principle.

Long before, Telangana used to be a separate state. In 1953 the government of India merged the then Andhra state and Telangana state into one big entity called Andhra Pradesh. The reason it was done was because then Prime Minister Pt. Jawaharlal Nehru wanted to create states based on linguistic lines. Both Andhra and Telangana had predominantly Telugu speaking population. He had also appointed a committee to oversee the matter. Even then Telangana did not want to be a part of Andhra. The issue was resources and revenue sharing. From Wikipedia:

The region had a less developed economy than Andhra, but with a larger revenue base (mostly because it taxed rather than prohibited alcoholic beverages), which Telanganas feared might be diverted for use in Andhra. They also feared that planned dam projects on the Krishna and Godavari rivers would not benefit Telangana proportionately even though Telanganas controlled the headwaters of the rivers. Telanganas feared too that the people of Andhra would have the advantage in jobs, particularly in government and education.

The committee hence, rightfully so, suggested not to merge the two states. The central government decided not to follow the suggestion and combined the states. They additionally negotiated an agreement that provided reassurances to the Telangana people as well to Andhra people in terms of power sharing as well as administrative domicile rules and distribution of expenses of various regions. This agreement is known as Gentlemen’s agreement of Andhra Pradesh (1956).

Well, it seems that the “Gentlemen’s agreement” did not live up to its nature and hence all problems began again. This time though, after careful deliberation and discussions, the government of India has decided to split the states back to their original forms.

As part of our “Pricing and Marketing Productivity” class, we had to prepare a reading list from the articles that we referenced in order to get a good understanding of the subject matter. Many of these articles have been immensely helpful in learning about the various material we have covered. Though this is a formal submission, I am also releasing this to the blog so that it benefits its readers also.

Marketing productivity analysis

• http://thefinancialbrand.com/2009/02/23/marketing-roi/
  • With social media exploding in the past few years, many companies are entering this channel in order to reach its consumers. But one complex issue that arises with this is how to measure your ROI. This article talks about one such idea.
• http://www.businessweek.com/magazine/content/04_50/b3912109.htm
  • Similar to the article above. It mentions that marketing is slowly transitioning from the domain of just art to a combination of art and science. Given this fact, the C-level executives are interested in knowing exactly how the ad revenue translates into tangible sales of their company’s products and services.
• http://www.emergencemarketing.com/2008/01/16/measuring-marketing-roi-can-be-harmful-to-your-marketing-department/
  • An interesting twist to the thought process. The author of this article argues that ROI is really a ‘trailing’ indicator and not a ‘leading’ indicator. So should new strategies be first justified and thus constrained in its creativity? Leave the marketers alone.

Market response models

• Market Response Models
  • I haven’t read this thesis completely because of it is actually a complete book in itself. But I have often referred to it in absence of a true textbook whenever I have to understand certain types of models. This document has helped me get familiar with many models not discussed in class but handy and useful.
• A Customized Market_Response Model
  • This article talks about building a market response model. The interesting thing about this was the ideas in this document matched very closely with the data we used most of the semester. Very helpful article. There are some portions that involve higher knowledge of mathematics but apart from that, very good.

Functional forms, Estimating competitive effects

• Mid semester module course material – Syndicated Data Analysis for Brand Scientists by Prof. Doug Bowman
  • This is not a link because the material was provided in hard copy by one of our other professor. I always have this deck handy because it is easy to read and clearly explains various functional forms and methods of estimating competitive effects. It starts from the very basic and slowly builds in the concepts.

Costs, Customers, Competition, and Pricing

• Games of Strategy – THE book on game theory
  • This book is a required material on one of my other subjects – Strategic Decision Analysis. A lot of the concepts we study in pricing involves understanding competition, their actions, and consumer’s willingness to pay. This book is a great resource to learn about concepts like the prisoner’s dilemma and other games that are helpful in maximizing marketing dollars.
• Evaluating Willingness to Pay
  • A basic presentation on willingness to pay – covers basic economic concepts and the ever elusive consumer surplus
• http://www.cval.com/pdfs/VBMarketingAndPricing.pdf
  • White paper on Value based pricing model. This model speaks about an approach a specific company takes in measuring the true value to consumer. Interesting approach.
• http://cauvin.blogspot.com/2009/01/value-based-versus-cost-based-pricing.html
  • Value based pricing vs. cost based pricing
• http://www.sciencedaily.com/releases/2008/01/080123125551.htm
  • An interesting insight from one of my favorite sites about what actually determines our willingness to pay. Are we objective or subjective? But the real question is how can marketers exploit this sweet spot to sell their product.

Psychological aspects

• Influence: Science and Practice – A seminal reading – Read review here.
• http://en.wikipedia.org/wiki/Psychological_pricing
  • Wikipedia provides this interesting article on psychological pricing for e.g. why do prices of most items end in 99? The article talks about similar pricing mechanism in order to lure the customer into purchasing.

Segmentation, and pricing

• http://www.mindofmarketing.net/2007/05/customer-segmentation-why-exactly-does.html
  • A fresher’s guide to the importance of customer segmentation – probably one of the most important step before any marketing can actually be done.
• The “Killer Application” of Revenue Management: Harrah’s Cherokee Casino & Hotel
  • Segementation and pricing at its best. This paper by a Goizueta professor was not only an interesting read but also very informative. Given the right strategy and right method to segment customers, Harrah’s Cherokee Casino truly knows which customer to give their rooms and what to charge them.
• Read Cluster Analysis concepts

We have learnt a number of bidding strategies in class and many a times the best strategy in most bidding scenarios is to know the value that you assign to the item being auctioned. It should be true when you are buying a house too. But, do you bid for a house even when there is no auction?

When purchasing a house, your real estate agent may show you a house and will tell you the list price. Typically, you will give a counter-offer which will be less than list price. You will hope that the  owner will either accept or negotiate some amount in between your amount and the list price. In this case you may be assuming that the owner understands prisoner’s dilemma and will act in the best interest of both parties (you and him) and that you will respond accordingly. But in our assumption we tend to forget that there is another player in the game who needs to maximize his returns as well – the realtor. There may be instances where he will drag you into a bidding game without you being aware of the situation. And then, you may get carried away.

One common strategy that realtors employ is that of finding another prospective buyer and them pitching you against them. Psychologically you will be expected to get desperate to win and get the house because you have by now developed an affiliation for it (To understand this concept better, do read my highly recommended Influence: Science and Practice). It is very important that you truly estimate what your ZOPA (Zone Of Possible Agreement) is and what your final limit is.

A good article that talks about such realtor strategies and how to defend yourself against being dragged into bidding, read this article by a real estate agent herself, Alison Rogers.

Number of articles have been written on Amazon and Wal-Mart’s war on book prices this holiday season. Let’s take a different look at the same story and how it can possibly help the economy and the consumers by – increasing the pie.

First, the story. Wal-Mart recently decided to start its own price war, taking on Amazon in the online book market. Wal-Mart began by marking down the prices of ten best-sellers—including the new Stephen King and the upcoming Sarah Palin—to ten bucks. When Amazon, predictably, matched that price, Wal-Mart went to nine dollars, and, when Amazon matched again, Wal-Mart went to $8.99, at which point Amazon rested…From a game-theory perspective, price wars are usually negative-sum games: everyone loses. A recent study found that, if competitors do match price cuts, industry profits can get cut almost in half. (Read more: here)

So the final prices were $9.00 (Amazon) and $8.99 (Wal-Mart). Where is the pie here?

I did some research to find out what exactly is the price elasticity of demand for books. Its a difficult number to calculate and I wasn’t very hopeful until I found this research article. This research intended to determine demand for books and was conducted in Norway. The research took into account number of demographic parameters into consideration. One of the numerous conclusion of the article was:

For our calculations, lets assume that demand is fairly elastic and it is –1.5.

So if the prices fell from $10.00 to $9.00 (say) then prices fell by 10%. Which means the books expected to be sold will increase by 15%. Which in turn means that the revenues will grow from 10X to 10.35X (9 x 1.15X). A value creation of 0.35X or 3.5%. While this number may look small in percentage terms, it is substantially big when we can find the real value of X (in 2006 3.13B books were sold) and substitute.

What Amazon and Wal-Mart have unintentionally done is to increase the total pie (the total revenues that can be earned by them). Yes, the margins may have taken a hit, but don’t we know that NPV is a better measure than IRR.

Do you think this research if conducted and analyzed in the US can change prices in the longer run? Thoughts?

A good analysis on this price war has also been done here.

Some time back I had written a piece about my experience with Atlanta’s transit system – Marta. Recently I had an opportunity (or should I say, misfortune) of travelling on Marta again. Because of current economic conditions Marta had decided to make some changes in its fare structure. And guess what, in my opinion, they took the wrong step. The fare has been increased from $1.75 to $2.00 and if you don’t have a breeze card (that costs $5.00) then you don’t get a transfer. C’mon!!!

But instead of complaining again, I want to look at Marta’s fare structure from a different viewpoint – fairness.

As I had spoken in my last writing, I do not find it correct that Marta charges everyone the same fare. If I have to travel 1.5 miles to school, I pay the same amount as someone who has to travel 15 miles to the other end of the town. It seems that because of simplicity of operational implementation, Marta decided to adopt this fare structure. But I feel that I am actually subsidizing the passenger who wants to travel longer distances. This assessment can be done by looking at some principles of fairness. Is the fair equitable or proportional? Definitely not. Is it efficient? Possible and will require further analysis. Is it envy-free? H**l no! So why does Marta have to adopt such a pricing structure?

Are there other solutions to this problem? Check out fare structure of Singapore’s transit system, one of the best in the world. It has a tiered pricing scheme that works well and hopefully caters to all the fairness requirements much more than that of Marta’s. It is important that Marta plays around with the supply demand curve much more intimately than just moving up the curves.