Name: Thai Hoang Nguyen (Lisa)

ID student: M0945019

CASE STUDY
Case One (Starbucks) assignment questions:
1. What factors accounted for the extraordinary success of Starbucks in the early
1990s? What was so compelling about the Starbucks value proposition? What brand
image did Starbucks develop during this period?
In the early 1990s, Starbucks was nearly became the leading branding company for the coffee
over the world. Nowadays, the company continues to expand rapidly. There are some factors that
accounted for the extraordinary success of Starbucks:
 Quality of coffee beans: Starbucks is considered themselves for the highest-quality
coffee. By controlling as much of the supply chain as possible, this helps them to
maintain control over their products and services.
 Atmosphere and services: Starbucks has trained their partners in both hard and soft
skills as how to mix drinks, interact with the customer by a smile, and eyes contact, etc.
Their goal is to create an uplifting experience where people would find a "third place"
between work and home.
 Employer satisfaction: Starbucks had one of the lowest employee turnover rates in the
industry. This is because Starbucks partners received the highest paid hourly workers,
health benefits, and stock options.
 Customer’s demands are there commands: In 1990s, Starbucks's targeted customers
are primarily the affluent, well‐educated, and white‐collar people. For those customers,
the heaviest parts are to provide the service at a consistent level and keep the customers
satisfied. Thus, the company always seems customer's demands as their commands by
flexible drinks customization, etc.

Starbucks value proposition is compelling because it places the customer on top. The initial
value proposition consisted of three elements:
a. Coffee itself (a physical good):
- Offering the highest-quality coffee in the world.
 - Maintaining and controlling of as much of the supply chain as possible.
b. Service (“customer intimacy”):
- Making the experience pleasant for the customer.
- Considering customer's demands are their commands.
c. Atmosphere (“People come for the coffee”):
- Providing comfortable seating areas that encouraged people to linger.

By the early 1990s, the brand image for Starbucks was:

 The other place between home and work, where people could go to relax and enjoy
themselves.
 The stores sold whole beans and premium-priced coffee beverages by catering primarily
to affluent, well-educated, and white-collar patrons, who were willing to pay the higher
prices for the coffee.
 In any Starbuck stores, customized service and a pleasing environment are helping them
to define and raise brand awareness.

2. How does the Starbucks of 2002 differ from the Starbucks of 1992?
In 1992 In 2002
Store distribution
- After the company went public, - In 2002, Starbucks had over 5000
Starbucks had 140 stores located in stores scattered the globe and was
the Northwest and Chicago. opening on average three new store a
day.
Company sales
- In 1992, half of the company’s - In 2002, about 77% of the sales came
revenue came from sales of whole- from beverages.
bean coffees.
Range of products
- The stores sold whole beans and - The company had develop new
coffee beverages by the cup. product innovation such as food items
and new beverages in its menu.
 Demographic profile of the customer base
- The historical customer profile of - The customer base of Starbucks in
Starbucks consisted of affluent, well- 2002 was changing to a younger, less
educated , and white-collar patrons. educated, and in a lower income
bracket.
The brand image
- The image of 1992 consisted of a - In 2002, the image had replaced to
place which you can call “third place”, convenient place, where customer can
where you can get the best quality meet people and enjoy the good
coffee and where you can relax. coffee.

3. Should Starbucks make the $40 million investment in labor in the stores? What’s
the goal of this investment? Is it possible for a mega-brand to deliver customer
intimacy?
Since the goal of this investment is to meet completely expectation in terms of customer
satisfaction. As Starbucks believed that the customer satisfaction gap could primarily be
attributed to a service gap between Starbucks scores on key attributes and customer expectations.
The plan is to invest the $40 million annually in labor for its 4,500 stores, then:
40,000,0000
The investment comes to one store = = $8889 per store
4,500
According to below table, number of customers that need to turn from being satisfied to being
highly satisfied:
Satisfied Customer Highly Satisfied Customer
Number of Starbuck
4.3 7.2
visits/month
Average ticket size/visit
$4.06 $4.42

Total revenue/month
4.3.(4.06) = $17.458 7.2.(4.42) = $31.824

Total revenue/year
17.458.(12) = $209.496 31.824.(12) = $381.888

Difference between 381.888 – 209.496 = $172.392

 Satisfied Customer and
Highly Satisfied Customer

The difference in revenue per year between satisfied customer and highly satisfied customer is
about $172.392.
8,889
Number of customers from satisfaction turn into highly satisfaction = = 51.56 ~ 52
172.392
customers per store
It indicates that in order for Starbucks to break even for this investment, it needs to turn 52
customers from being satisfied to be highly satisfied in one store. When Starbucks had polled its
customers to determine factors of customer satisfactions - in particular, speed of service had been
mentioned most frequently is the driver for satisfaction that needs this additional investment.
If customer satisfaction does not increase, an alternative solution is to acquire new customers
come to stores.
Break‐even in Break-even in ticket
Current situation
customer count size

Average hourly rate $9.00 $9.00 $9.00

Total labor hours per

360 360 + 20 = 380 360 + 20 = 380
week, average store
Total labor hours per
360 / 7 = 51.43 380 / 7 = 54.29 380 / 7 = 54.29
day, average store
Labor cost/day,
9.(51.43) = $462.87 9.(54.29) = $488.61 9.(54.29) = $488.61
average store

Average ticket $3.85 $3.85 $3.90

Average daily
570 577 570
customer count

Revenue 570.(3.85) = $2194.5 $2220.24 $2220.24

Revenue – Labor 2194.5 - 462.87 =

$1731.63 $1731.63
Cost $1731.63
In break‐even customer count:
Revenue = (Revenue – Labor Cost) + Labor cost/day = 1731.63 + 488.61 = $2220.24
Revenue
An additional daily customers should come = - Average daily customer count
Average ticket
2220.24
= – 570 = 6.7 ~ 7 customers
3.85
In break‐even ticket:
Revenue = (Revenue – Labor Cost) + Labor cost/day = 1731.63 + $488.61 = $2220.24
Revenue
An additional should be spent by a customer = - Average ticket
Average daily customer count
2220.24
= – 3.85 = $0.05
570
Since there is a link between customer satisfaction, customer acquisition, and average ticket size,
then if the investment will increase the customer satisfaction it would be reasonable. Therefore, I
believe that Starbucks should invest the money in labor. In addition, the company should do
more analysis of its customer base and identify areas where people are less satisfied because of
the speed of service and invest in those locations only.

Case Two (Chinese Pharmaceuticals Limited, CPL) assignment questions:

1. How does the nature of Chinese Pharmaceuticals’ business impact the design of its
supply chain?
Drug suppliers create a link between the pharmaceutical factories and customers. Chinese
Pharmaceutical is a Hong Kong company supplying Chinese herbal medicine products to the
Hong Kong market. While the company had expanded its product portfolio, but its best-selling
product remained Noto37, a Chinese herbal medicine used to control cholesterol and blood
pressure levels which is primary ingredient was Notoginseng.

The company had an established supplier the raw materials in China. The pharmaceutical
manufacturer have been stocked the ingredients in Hong Kong leased warehouse facility after
taking the test of raw materials. This process takes at least 90 - 100 days. Then, the company
worked with a local transportation company to arrange delivery of new stock of Noto37 to its
local customers, usually once or twice a weeks.
Due to the long process and low accuracy of its sales forecast method and inventory
management for the company. The company had faced an out of stock phenomenon (Noto37).
Therefore, Chinese Pharmaceutical needs to implement its design of supply chain reasonably to
avoid out-of-stock product conditions.

2. What are possible supply chain management options that Chinese Pharmaceuticals’
general manager might have to more effectively manage the inventory and avoid
being out of stock?
There are various factors impacted the stock levels :
 Biweekly deliveries of different quantities to almost 1,000 retail outlets.
 Intermittent delays in supply of new stock from the manufacturer.
 Drought conditions in Yunnan that could impact supply of the key Notoginseng.
 Company’s weekly promotions plus the company’s own intermittent promotions
contributed to fluctuations in demand and on stock levels.
Upon the above factors, here are the strategies that general manager is able to take in managing
their inventory effectively:
 Implement a fixed-order quantity model in inventory and check inventory position
continuously to determine the economic order quantity (EOQ).
 Always prepare a safety stock to anticipate a surge in demand and reduce lose sale.
 Improve the sales forecast method by considering seasonality and trend.
 Doing good coordination with the sales and marketing department of the company to
actively promoting their products .
 Establishing good connection with other suppliers to find out information on conditions
that have the potential to cause delays in delivery.

3. In this case study, a simple exponential smoothing forecasting model is applied and
the value of alpha used is 0.4. It seems this is not the optimal value for alpha. How
can the near-optimal value of alpha be determined?
The table below was calculated exponential smoothing forecasts for these time series data in
Excel and the value of alpha used is 0.4.
To determine the near-optimal value of alpha by using Solver in Excel to determine the best
smoothing constant or alpha value (α). In addition, the required smoothing constant (α) is in the
range from 0 to 1.
These are following steps:
- Supposing that the best alpha is to minimizes MAD. Then, for Set Objective, we click on
the MAD value (cell number F44) and click Min for minimization.
- The variable value wanted to change is alpha, thus click the alpha value (cell number J5)
for the By Changing Variable Cells.
- To ensure alpha is less than or equal to 1, then click Add to add the constraint (cell
number J5 <= 1).
- Then click checked box that Make Unconstrained Variable Non-negative.
- For Select a Solving Method as Non-Linear and click Solve.
After solving, the optimal alpha value is 1. That means to obtain the smallest MAD possible
(MAD = 479), the optimal value for alpha is to about 1.

4. What is the benefit of allowing alpha to vary? Discuss the adaptive approach.
A key factor to accurate sales forecasting with exponential smoothing is the selection of the
proper value of alpha (α). As mentioned above, the value of alpha (α) can vary between 0 and 1.
Regardless of the initial value (α ) selected, a new value of alpha have to be adjusted periodically
to ensure that is providing accurate forecasts and reflect closely the actual demand by using
various values of alpha.
The adaptive exponential smoothing (AES) method is a derivative of simple exponential
smoothing (SES). The value of alpha is systematically changed from period to period to allow
for pattern changes in the historical data. Adaptive exponential smoothing is beneficial to use
for:
 A product that has been sold for a long period of time, only to encounter a dramatic
market change in recent periods.
 The data is non-seasonal and shows no trend.
Comparing between simple exponential smoothing (SES) and adaptive exponential smoothing
(AES) in this case, it showed that:
 A smaller MAD (MAD = 825) is in SES approach, whereas AES approach resulted a
bigger MAD (MAD = 1,104).
 The actual sales data was seasonal and showed the trend. In this case, the winter months
were the clear peak season for sales of Noto37, and the sales decreased during the
subsequent spring season.
 Therefore, the adaptive approach is not well appropriated for the improving the low accuracy
of the sales forecast method.

5. Are there any ways to further improve forecasting accuracy? What are the
advantages of the full exponential smoothing model (i.e., when trend and seasonality
are considered)?
To improve forecasting accuracy, a further method that could be considered is full exponential
smoothing model. That is a model combining three aspects of the time series:
- A typical value (average)
- A slope (trend) over time
- A cyclical repeating pattern (seasonality).
There are several parameters including α, β, and γ. For each component (level, trend, seasonal),
there is a smoothing constant that falls between zero and one.

Comparing three methods, full exponential smoothing model obtains the smallest MAD (MAD =
765). From the calculation of the error rate, it can be concluded that the level of forecast
accuracy more fits and will be better if company uses the full exponential smoothing method.
 Method MAD
Simple exponential smoothing 825
Adaptive exponential 1,014
smoothing
Full exponential smoothing 765

There are some advantages of the full exponential smoothing model:

 It is suitable for forecasting data with trend or seasonality.
 A useful method when there is trend in the data along with seasonal variations.
 Help to identify trend direction and seasonal change.