"Marketing Engineering" - Digest 5

Study of cases

• Chapter 3: Segmentation and targeting
• Case problem: 
• How to find key drivers of customers' choice on products based survey data of customer preference and their choice
• How to win loyalty of customer and how to reduce customer churn rate based on key driver found in above questions
• Case solution: 
• Use logit model to derive relationship between propensity to buy with their preference on features
• Cluster customer based on propensity to buy. Focus on competitive customer (slight prefer ABB product instead of competitors) and switchable customers (slight prefer ABB's competitor product instead of ABB's and ABB is the second)
• Based on the formula, find independent variable with statistic significance
• Investigate impact of these key drives to competitive customer and switchable customers and figure out how to win loyalty of customer and how to reduce customer churn rate
• Why the author put this case after the chapter
• Customer segmentation based on propensity to buy is very useful to identify customers that worth sales' efforts on them
• How can the case be implemented into real situation
• It might not be easy to identify a few variable that really represent customers' attitude toward the product. The starting point might a find key variable from lots of variables through ANOVA.

"Marketing Engineering" - Digest 4

Study of cases

• Chapter 3: Segmentation and targeting
• Case problem: 
• How to segment customers based on data of segmentation basis (data about customer preference on product features and price)
• How to categorize customer to different segment based on data of segmentation descriptor (data about customer demographics) 
• Case solution: 
• Cluster analysis (KNN maybe) to segment customers based on data of segmentation basis.
• Based on cluster information, use discriminant analysis (multi regression maybe) to derive formula to categorize cluster
• Why the author put this case after the chapter
• Emphasize the difference between two variables, basis and descriptor
• Emphasize segmentation should focus on customer needs instead of profiles
• How can the case be implemented into real situation
• Though data of segmentation descriptor can be get easily, data of segmentation basis designed and collected purposely.
• For different purpose of segmentation, different data of segmentation basis should be collected. Here lists some in the book
• for position studies
• product use
• product preference
• benefits sought
• a hybird of the variables above
• for price decisions
• price sensitivity
• deal proneness
• price sensitivity by purchase/use patterns
• for advertising decision
• benefit sought
• media use
• psycho-graphic/lifestyle
• a hybrid
• for distribution decision
• store loyalty and patronage
• benefits sought in store selection

"Marketing Engineering" - Digest 3

Topics:
Back to the book of "Marketing Engineering". Look at the book, I'm not sure if I will read the book from cover to cover. Case in the book seems practical. And algorithm looks useful as well, but time consuming to dive into since I don't have the software with the book. Maybe the easy thing to do is to go through cases at the end of each chapter and figure out the idea behind it.

Framework
The following framework is designed to construct the study of cases after each chapter.

• Chapter
• Case problem
• Case solution
• Why the author put this case after the chapter
• How can the case be implemented into real situation

Study of cases

• Chapter 2: Tools for marketing engineering: market response models
• Case problem: 
• How to spend promotion budget? 
• The relationship between promotional spending and sales units are given and the price and cost of each unit are provided
• Case solution: 

• Use Adbudg model to curve fit the relationship between promotional spending and sales units.
• Maximize profit based on the derived curve.
• Why the author put this case after the chapter
• Emphasize saturation of sales due to promotional spending
• How can the case be implemented into real situation
• One challenge to implement it is that relationship between sales units and promotional units. 
• What if it is a new products?
• What if it is a new promotional method?
• One possible answer for this might be to get the relationship through trials and errors
• The other challenge is that parameters in promotional spending might be more than one
• maybe it is the problem of strategy for advertising

Summary of my knowledge about statistics or hypothesis test in particular

Reason for summary
I feel it's time to stop looking into more statistics or hypothesis test in particular. Anyway I'm not going to work as a statistician. I should position myself as marketing analyst with solid knowledge of statistics and capable of conducting complicated hypothesis test for marketing test.

List of my knowledge about hypothesis test

• parametric testing
• one sample test (z test and t test)
• two sample test (mean test and standard deviation test)
• one way ANOVA
• two way ANOVA (full factorial design and fractal factorial design - Taguchi method)
• non parametric testing (normal distribution test)

The next question is what is the value to marketing. Maybe it is easy to explain it in design of experiment. hypothesis test is one step of design of experiment

Marketing campaign can be considered as one type of experiment with input as marketing mix (4P, price, place, promotion, and product), output as goals (revenue, profit, market share and etc) and market itself as black box. After you have insight about the market, you might find a few key factors. You will try to find the optimal combination to maximize your goal with minimum cost. That's where hypothesis test comes in and help it.

Application,
One interesting application might be in Google ads optimization. Here is an very interesting youtube video to find the best ads with different heading, first description line, second description line and url. One thing to note is that, in practice, people care about the optimal level of the factor, rather than know the factor really has impact on the experiment.

P.S. From tomorrow, I need to focus more on marketing strategy and cases instead of techniques in statistics. Don't be a nerd.

Buzz words in job requirement - Taguchi method

The first time I heard "Taguchi method" is in the six sigma course. It's not covered by the intense course. But the professor did mention it several times.

I kept on searching for details about the method. It seems not well defined even in Wikipedia. Maybe the term itself is not clear or accurate. The exact term might be "Taguchi method in design of experiments". If that is the case, I think "Taguchi method" is just orthogonal analysis in fractal factorial design.

the Idea
In simple English. "Taguchi method" is the method to use smallest runs of experiment, which neglects effect of combination of factors.

If effect of combination of factors is not our interest, we can use 8 runs of experiment to represent an experiment with up to 7 facors, which is far more effective than full factorial case, 2⁷=128

Design of experiment
I'm not sure if should use this as a heading. It covers lots of topics and Fisher even wrote a book about it in 1935. Here I will copy the rule of thumbs about options in design of experiments from one online document
as the summary of this blog.

When do we use which method?

• Option 1: Factorial Design
• Small numbers of variables with few states (1 to 3)
• Interactions between variables are strong and important
• Every variable contributes significantly
• Option 2: Taguchi Method
• Intermediate numbers of variables (3 to 50)
• Few interactions between variables
• Only a few variables contributes significantly
• Option 3: Random Design
• Many variables (50+)
• Few interactions between variables
• Very few variables contributes significantly

Buzz words in job requirement - fractal factorial design

Here is the buzz word I used in my experience as project officer in charge of an experiment. To relate the experience to marketing test, I have to use to some buzz words. "fractal factorial design" seems complicate, and the professor of six sigma mentioned the course. So, it must be useful somewhere in some way.

First of all, definition from Wikipedia.

In statistics, fractional factorial designs are experimental designs consisting of a carefully chosen subset (fraction) of the experimental runs of a full factorial design. The subset is chosen so as to exploit the sparsity-of-effects principle to expose information about the most important features of the problem studied, while using a fraction of the effort of a full factorial design in terms of experimental runs and resources.

There are key terms here, "fraction" and "full factorial design".

"full factorial design" is the way to list all the combination of change in each factors to take all factors and combination and factors into consideration.
For an experiment with 2 factors, source of errors/difference in observation comes from X₁, X₂ and X₁X₂

For an experiment with 3 factors, source of errors/difference in observation comes from X₁,X₂,X₃,X₁X₂,X₁X₃,X₂X₃,and X₁X₂X₃

It's simple. But when number of factors increases, runs of experiment increase exponentially. it's 2ⁿ. In some cases, it's extremely expensive to conduct the full factorial design.

That's where "factual" comes. Number of runs does not increase when new variable comes in, but some combination of variable need to be compromised. Here is the example.

Conclusion of fractal factorial design, full factorial design is the dumb but expensive method. fractal factorial design is the cheap method with the trade off of cost.

P.S. I need to related it to the context of marketing. The technique can be used in design of marketing campaign test. For traditional marketing campaign, fractal factorial design might be necessary because it's expensive. But for online marketing campaign test, it seems full factorial design is possible.

P.S.2 Here is a youtube video that might be useful to explain fractal factorial design.

Buzz words in job requirement - ANOVA part 2

I am afraid there has to be ANOVA part 2 if change of more than one factor is conduct simultaneously. Another reason is that the title with part 2 is so cool. It sounds like a Kill Bill 2, Harry Potter and deathly hallows part2.

1. What problem can cool ANOVA part 2 can solve?
In one-way ANOVA, samples are different due to change of one variable. What if there are more than one factor/variable? In this case, there are 2-by-2 matrix, and each cell contains one sample with several observations.
In the context of marketing, imagine the marketer design a campaign with input of ads spending and discount level. He/she is not sure about how to fine tune the parameter, so he/she set different levels for each factor. impact of market is recorded. The question is if market really respond differently as factors change.

2. Idea of two-way ANOVA
Assume there are only two factors. And assume means of these samples are identical. Errors of each observation comes from four sources, factor A, factor B, interaction between A and B, randomness.

SS_A, SS_B and SS_AXB are all compared with SS_within. If they are relatively large, they causes difference more than randomness.

3. Formula of two-way ANOVA
Of course, there should be formula when there is no analyst supporting you. You can find formula here.

Buzz words in job requirement - ANOVA

One of the popular buzz word is ANOVA, which is acronym of analysis of variance.

1. What type of problem can be solved by ANOVA?
There are several samples, which could be generated by different methods. ANOVA can test if means of these sample are different. In the context of marketing, several campaigns are run among different countries and impact of the campaigns (revenue maybe) is recorded. ANOVA can test if impact of campaigns differs among different countries.

2. ANOVA v.s. t test of means by two groups
If there are only two groups, t test can be used instead of ANOVA for simplicity.

3. The idea of ANOVA
If means of the sample groups are assumed to identical, then error of each observation is due to two reasons. One is the error within sample, the other is between samples.

when SS_A is relative larger than SS_withinwithin, we can say that means of different samples are different.

4. Formula of one-way ANOVA
Formula is always boring. You can find it here.

Buzz words in job requirement - A/B testing

Here are some other buzz words, A/B testing and multi variate testing.

This blog focuses on A/B testing because it's simple.

First of all, let's see the definition in Wikipedia.

A/B testing, split testing or bucket testing is a method of marketing testing by which a baseline control sample is compared to a variety of single-variable test samples in order to improve response rates. A classic direct mail tactic, this method has been recently adopted within the interactive space to test tactics such as banner ads, emails and landing pages.

If we can put the buzz word in the context of experiment instead of marketing or business, things are quite easy actually. The situation is that we need to find the optimal parameter for an experiment. The most simple thing we know is to change the parameter as many as time as we like, and compare the performance understand different value of the parameter. Of course, we need enough samples under each value of the parameter to see difference in a statistical significance way.

This is the dumbest way to optimize an experiment. But there are a few companies well know for A/B testing, listed by Wikipedia,

• Amazon.com pioneered its use within the web ecommerce space. 
• BBC.
• Google. One of their top designers, Douglas Bowman, left and spoke out against excessive use of the practice.
• Microsoft 
• Playdom (Disney Interactive) 
• Zynga 
• eBay

It's a really dumb method. That is my conclusion for this blog.

P.S. Here is product tour of a company who help to optimize visual website through A/B testing. It might be helpful to understand A/B testing.
http://visualwebsiteoptimizer.com/features.php

Buzz words in job requirement - Email marketing

After searching for job days and days online, I have to say some employers are really obsessed with certain buzz words. I really doubt these are fundamental skill sets. In the following blogger, I'm going to list some of them and show my thoughts about them.

buzz word 1: email marketing

First of all, I didn't know what email marketing is before job searching. Here is the paragraph to describe it in Wikipedia.

Email marketing is a form of direct marketing which uses electronic mail as a means of communicating commercial or fund-raising messages to an audience. In its broadest sense, every email sent to a potential or current customer could be considered email marketing. However, the term is usually used to refer to:

• sending email messages with the purpose of enhancing the relationship of a merchant with its current or previous customers, to encourage customer loyalty and repeat business,
• sending email messages with the purpose of acquiring new customers or convincing current customers to purchase something immediately,
• adding advertisements to email messages sent by other companies to their customers, and
• sending email messages over the Internet, as email did and does exist outside the Internet (e.g., network email and FIDO).

OK. it's just a method of direct marketing. We receive spam email everyday with unrelated products. It's simple. What should we care if we need to do email marketing? 1. Selected customer. 2. Proper contents.

For the first one, targeting the right customer with the right product is always the key to lots of marketing/sales approach. Sending an email to irrelevant customers might cost the company little, but it could damage brand/image of the company.

For the second one, contents could be proper products/promotion. It still relates to customer selection. In addition, if the link for purchasing is provided in the email, success rate of the email marketing can be evaluated easily.

In essence, email marketing reflects three aspects, customer selection, customized promotion and close loop marketing. I feel that's the fundamental of email marketing. That should be areas employers care.

One example I think about is Groupon, which might be considered as an successful example of email marketing. But uniqueness of Groupon is that it matches local demand (customers in certain city) and local supply (sellers in certain city). Geo info is a very good filter for customer selection. Now Groupon has lots of copycat. We can see daily special of Expedia, but I really doubt that would be successful.

"Marketing Engineering" - Digest 2

Still in Chapter 5: strategic market analysis: conceptual framework and tools

There are two important concepts: product life cycle and cost dynamics: scale and experience effects


digest 1: S shape in product life cycle

* the picture is not from the book

digest 2: existing other shapes in product life cycle

* the picture is not from the book

Cycle-recycle pattern has been observed in drug products by Cox in 1967 and in Boeing 727 and 747.

digest 3: experience effects
C_q=C_n(q/n)^-b
where q= cumulative production to date
          n= cumulative production at a particular, earlier time
          b=learning curve
in practice, experience curves are characterized by their learning rate. Suppose that each tme experience doubles, cost per unit drops to 80 percent of the original level, then the 80 percent is known as the learning rate.
r=2^-b*100

"Marketing Engineering" - Digest 1

Marketing Engineering, Gary L. Lilien (Author), Arvind Rangaswamy (Author)
Bought in Amazon to prepare myself for possible interview for an marketing analyst position

Chapters read today. Chapter 5- strategy market analysis: conceptual framework and tools.

Thought 1: In simple English, think about big pictures before diving into data.Simple truth, but are not always reflected in actions. Big picture (in nowadays, is buzz word most of the time) here means corporate strategy, which might cover product innovation, markets to serve, personnel, R&D, and corporate image.

Thought 2: Rest of the chapter talks market forecasting, which totally makes sense because any manager focuses on pursuing big opportunity with limit resources. Here is the nice summary of classification of market forecasting approaches.

Judgmental	Market and survey analysis	Time series	Causal or correlational analysis
Sales-force composite Jury of executive opinion Delphi methods	Buyer intentions Product tests	Native methods Moving averages Exponential smoothing Box-Jenkins method	Regression analysis Economics models Input-output analysis MARMA

Digest 1: what method to use

• use structured methods rather than unstructured methods
• use quantitative methods rather than judgmental methods if enough data exist
• use causal methods rather extrapolation methods, especially if changes are expected to be large
• use simple methods unless substantial evidence exists that complexity helps
• match the forecasting method to the situation

Day 1 - First day of unemployed

May 21st, the first day after MBA graduation and the first day of unemployed. I just can't believe the facts that I am unemployed. I have to write down something each to keep myself sane.
The topics can cover a wide range of subjects. Books, articles in journals and newspapers, movies and music. mainly, I guess.  And my thoughts of course. Things I'm willing to say.