mCommerce - Definition

Definition
Let's look at definition in wikipedia.

"Mobile Commerce is any transaction, involving the transfer of ownership or rights to use goods and services, which is initiated and/or completed by using mobile access to computer-mediated networks with the help of an electronic device."

 and Wikipedia also provides products and services available.

• Mobile ticketing
• Mobile vouchers, coupons and loyalty cards
• Content purchase and delivery
• Location-based services
• Information services
• Mobile banking
• Mobile StoreFront
• Mobile brokerage
• Auctions
• Mobile Browsing
• Mobile Purchase
• Mobile marketing and advertising

It's a very broad definitions. In different context, mobile commerce can mean different things. In my understanding, three products are mostly often referred as mobile commerce, mobile purchasing (in the context of retailing), location-based services (in the context of mapping of services) and mobile banking (in the context of payment).

So, it's important to specify the context when the term mCommerce is used.

Where will the jobs be?

Back to my interest, where will the job be? 

• mobile purchasing: Yes. If mCommerce is considered as a special case of eCommerce. There are much improvement to be done for mCommerce, such as improvement on webpage design for small screen. It seems all technologies are there, but you still need to tweak them. 
• location-based services. Yes. It's exciting to introduce one news important factor into traditional marketing. For example, product listing would be more specific when sellers know exactly where customers are.
• mobile banking. Maybe not. Definitely it represents huge business value. Like credit card, it will enable transactions that are previously time consuming, such as payment with touch of mobile in supermarket.

mCommerce

Introduction: why am I interest into eCommerce

I started to be fascinated about the word since yesterday night. It starts from a job posting by Toys'R'Us. One of the requirement is to manager update of iPad app. Then I checked the iPad app for Toys'R'Us. I felt impressed. Browsing of products was intuitive. Designing of website is kids friendly. And then I used "Discvr apps" (one of the amazing app to visualize connection between apps and find news apps based on known apps) to explore related mCommerce apps, I was surprised to see so many retails shops has built apps in iPad, Target, Stop'N'Shop and etc.

Categories of products that has built apps in iPad

I was thinking if there are still some products not covered, and that would be an opportunity. My starting points is fixed basket of CPI.

Housing: 

   Existing apps like Realtor.com that provides information of house for sale

Transportation: 

   Existing apps like CarZen that provides spec/price of cars

Foods and beverage: 

   all sorts of online grocery shops that sells foods 

   Yelp that provide feedback on restaurants

   McDonald's provide an app to indicate all its branches around location

Recreation

   Disney provides apps for waiting time in its Theme park.

Medical care

   MGH has its own to promote its brand. Not that commercialized, but it's a start.

Appreal

   Lots of catalogers now have their iPad version. Cheaper and nicer (including some video of products)

There are lots of space for improvement, but one fact is obvious that all sellers are taking channel of mobile seriously and they start to work on it.

What I can do for these sellers

• Develop apps: write process/documents to develop apps in screens of 4-5 inches instead of 17 inches. That's where MS fits in.
• Collect customers' needs and translate into apps: marketing stuff
• How eCommerce help to cut cost and drive revenue: strategy stuff

A few thoughts about CPI in China

After I've read the book of "Principle of macroeconomics", I started to look into problems of macroeconomics around me. For US, most import problems are unemployment and trade deficit. For China, most important problem is inflation.

When I looked into CPI is China, I found it tricky in several aspects.

1. Weight of Items in the fixed basket is not transparent.
People estimate from published monthly that weight of items might be

• Food  30.5%
• Tobacco, Liquor and Articles  3%
• Clothing 8.5%
• Household Facilities, Articles and Services 5.5%
• Health Care and Personal Articles 7.5%
• Transportation and Communication 12.3%
• Recreation, Education and Culture Articles 12.3%
• Residence 20%

Compare with US CPI weight, Food is the biggest components in CPI, while residence is relatively small. In US, Food takes around 15%, and residence around 40%. Basically it reflects reality in China. Normal people is not rich, so lots of expenditure are needed for foods. And most of people are living in rural areas so that they can enjoy low cost of residence. That's why high price of commercial residence property is not reflected in CPI yet. With the development of Chinese economy, CPI weight in China is expected to be close to that in US.

2. CPI is not taken serious in China

CPI is supposed to be an important indicator to reflect real interest, but in China few people cares about CPI because of

• Method to calculate CPI is not transparent. Weight, data collection and etc. Nothing is revealed on website
• Data manipulation by National Bureau of Statistics in China. Some National Bureau of Statistics in China is the tool of government to achieve some kind of target.

Currently, increment of CPI in China is around 5%. Government expect to lower CPI through policies. To be peculiar, the government set price ceiling to control price. It is totally against mechanism of market. The following situation is expected to be seen

• Short supply of goods. Shortage of power supply this year is due to that.
• Unreasonable distribution of resource/profit among industries. 

In the end, I just hope government could be wiser and follow the rules in economics.

Real Estate Bubble and when it's over

What does real estate bubble relate to me?
I don't plan to buy house in US in the near future. But yesterday's published Case Shiller showed home price has dropped last quarter and invests are afraid of real estate bubble will impact whole economics so bad that there would be another round of recession. Then private sector will be very cautious to hire people and i will have small chance to be employed. It means I need to investigate into it.

What is real estate bubble and how this real estate bubble is triggered?
let's look at definition of real estate bubble in Wikipedia.

A real estate bubble or property bubble (or housing bubble for residential markets) is a type of economic bubble that occurs periodically in local or global real estate markets. It is characterized by rapid increases in valuations of real property such as housing until they reach unsustainable levels relative to incomes and other economic elements, followed by a slide in price levels over a number of years.

The figure of price index since 1988 explains how it increase rapidly and fall due to unsustainability. (dot line for index, while solid line for growth)

Khan explain the reason for bubble well in his video "The housing price conundrum". Summary of his video is that, price is based on supply and demand. From 2000 to 2006, demand effected by population and income almost stagnate, while  supply of new house increase slightly. Real reason behind hiking of previous home price is constraint for mortgage is loosen, while stimulates demand.

Which price level is normal level? or when home price will stop dropping?
From 2000 to 2006, there are lots of people who were not qualified to buy house actually bought house. Now these houses face foreclosure and these excessive supply lower current prices.
We can derive that, when these house which were bought by unqualified owners are all bought by current qualified house buyers, price will be back to normal. That's the base of the calculation. Also we need some addtional assumption,

• 1990-1999 is baseline. Sales unit increase proportionally to population
• 2000-2006 is the up hill of bubble time. Rational (or expected) sales unit is supposed to increase proportionally to population, but actually buyers bought more. The difference is excessively bought house.
• 2007-Near future is the down hill of bubble time. Level sales unit stays at the level of 2008-2010, which is below rational (or expected) sales units. Now buyers are buying houses accumulated in 2000-2006.

From the calculation, in 2011, all accumulated excessive house will be bought by current buyers so price is supposed to be stable possibly Q2 of 2011. Radical reaction of stock market (2.34% of Dow) reflects only emotions of panic investors.

What does it relate to me?
Things will be better tomorrow. I will find a job. I have faith for that.
Maybe I should long stock index for a few days. Investors will realize bubble is almost over soon.

P.S. Risk the calculation
Whole calculation is sensitive to slope of sales unit. Several facts could affect the calculated slope.

• income increase in 1990-1999, while it decrease in 2000-2009. So calculated slope might be larger than real one
• 1990 and 1991, US just recovered from last recession, sales unit might be lower than expected. So calculated slope might be larger than real one.

When calculated slope is adjusted lower due to these facts, time need to be back to normal level might be longer. If calculated slope is only half original one, we need to wait till beginning of 2013 to return to back-to-normal level.

Correlate of Google Lab

Google can still be innovative. That's the one piece of proof.

What is Google correlate?
The story starts from Google trends. Google trends lists pattern of key words in time series.

It's natural to think it in a reverse way. If we know pattern, can we also get key word from the pattern? The answer is Yes with conditions. These patterns are supposed to be existing in Google database.

Next, if both are achievable, we correlate keys words that share same or similar patterns.

Features
This product provides two interesting features.

• One is to search for similar pattern with input key-words/drawing/user-data. With it, we can find associated rule for products. (just one of many applications)
  Here is one test i've done. I inputed "macbook". it lists bunch of garbage related key words, like "buy macbook", "apple macbook". There are also useful info, like "magsafe" (accessory), "macbook pro keyboard" (accessory), "macbook battery" (accessory), "logmein" (access software for different OS?), "freenas" (more storage on cloud?), "lenovo laptop" (comparison before purchase?) and some unexplainable words.
• The other is to search trends distribution on geo. It's a function should included in trends, but not.
  Here is one test i've done. I inputed "tornado". The area impacted by tornado is clearly shown.

How does it related to my job searching

It's fun, but I should not spend too much time on that. Here is the area it might help me on job searching, distribution of job across different states in US.

Key words "finance job": more finance job in NYC and around NYC

Key words "marketing job": almost everywhere

Key words "IT job": it's a little bit surprise. I thought CA might have lots of IT job. Maybe it's due to two reasons, one is CA is weak in hiring, the other is IT is a support function now. it could hire people anywhere. and some states show growth in hiring.

"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