Session 6 Homework

Hi all,

Session 6 on segmentation and perceptual mapping, got done y'day.

Since we had structured, metric data this time - running R was quite smooth in-class, unlike in session 5.

Pls find uploaded on LMS R code and related datasets for session 6. Pls try to replicate classwork analyses at home with these.

Update: Some folks asked for additional reading material on the clustering methods themselves. Here goes.

The 3 main data clustering methods we covered were: (a) hierarchical clustering via the hclust() function in R

(b) the popular k-means technique ( kmeans() in R )

and (c) the powerful model based clustering technique of Fraley and Raftery (2002) under the Mclust() func in R.

In the second half of session 6, we covered perceptual mapping and joint space maps.

For those seeking greater interpretative clarity on the p-maps, pls scan through this old blog post on precisely this topic from 2012.

For those looking to explore particular methods further, I hope the hyperlinks embedded above help.

Coming to the HW:

Given that session 5 homework has been delayed, am inclined to go easy in this one. HW here shall be light. Only.

There are two small, simple group homeworks for this session. Both easy to run using the code provided. Your interpretation of the analysis is of interest to me.

Homework 1: Segmentation (group submission)

Kirin Brewery Limited, founded in Yokohama, Japan over 100 years ago was the largest beer firm in Japan, and among the 10 largest in the world.

They entered the world's largest beer market, the US, in the 80s and 90s.

You can read up about Kirin and the main characteristics of their offerings here.

Kirin performed some MKTR upon entry and some of the data they collected in available for our analysis.

Read in 'Kirin basis variables.txt' and run the associated code given in 'R code for session 6 HW.txt'.

Hint: You may want to spend a minute carefully reading the attributes they collected data on. What constructs were they after?

Now answer the following Qs:

1. How many segments are there (under mclust)? What is the ratio of the largest to smallest segment?

2. Characterize the segments. Give each an informative name.

3. Read up on the flagship offerings of Kirin Brewing USA here. There are three main Kirin beer variants: Ichiban, Light and Free.

Given each variant's particular features, which segment should each variant target? Why?

Deliverable: Write your answers in a PPT slide deck not exceeding 3 slides. Name it as your_group_name.pptx and drop into appropriate dropbox.

Homework 2: Perceptual Mapping (group submission)

We return to the PDA case we did in class for segmentation. Read in the datasets and run the code as given in 'session 6 HW code.txt'.

Answer these Qs below.

1. Who are PDA ConnecTor's closest competitors? On what attributes do they compete?

2. Present a quick, simple SWOT analysis based on the JSM for the PDA product.

3. Which products or product types seem to have the highest preference?

4. Any white spaces for potential entry that you see?

Deliverable: Another 2 slides in the same PPT as the one above.

Deadline: A week+ from now. Midnight 28-sept Sunday.

Any queries etc. contact me or use the comments section below.

Ciao.

Sudhir

P.S. Was some 20 minutes short of time in Section A. Overcompensated perhaps and was 20 minutes ahead of time in section B. Finally, Section C finished just right.

Sorry about the mismatches. My aim is to present a consistent experience in all sections.