Inthis graph, we can clearly see that the data analysis was done based ondiscrimination. This is a clear indicator which segment of people based on age,usage preference, education level, etc. would likely to accept the PDA product. In question 2, my recommendation was Segment C (cluster:2,5,9).
In this graph, we see a mixer of age groups from 28 to 51. But mostly48 to 51 in cluster 2 and 5. Their level of income is much higher than cluster9.
No one from cluster 9 has ever owned a PDA, their usage/demand on cellphoneand PC are also relatively lower than other clusters. A better choice of focusgroup may be segment B, which the income level for cluster 3 is slightly aboveaverage and cluster 7’s income level is off the chart. Cluster 7 has higherincome level, better education, deal more with technology such as cell phoneand computers. Many of them also have experiences with PDA already. A similarlygood recommendation may be cluster 6, similar attributes with cluster 7 butmuch younger professions, more innovative than cluster 7. However, I noticedthat from the discrimination spreadsheet, the Dendrogram categorize clustersdifferently. If I was to re-categorize the segments, it would have been:Segment A (1,8,5) Distance (8,727.07)Segment B (4,9) Distance (2,139.
58)Segment C (2,7) Distance (4,507.81)Segment D (3,6) Distance (4,513.54)Question 4: This analysis helped me to identifyeach segments by features/attributes, demographic groups, and discrimination.It allows me to compare the data and see which cluster or segment is bestsuited for targeting by considering both demand and supply point of view. Themethod of data collection is effective and easy to understand. It helped me toidentify which segments of consumers would have a need in conneCtor’s PDA andprioritize these segments based on profit maximization. In addition, the toolhelps to analyze which features should ConneCtor focus on future products andwhat the consumer tastes and preferences are.
Overall, it is a helpfultool/method based on the data provided above . Question 5: Since the data/sample is fairlylimited, it may not reflect the market in the bigger view. Also, setting upcluster size could be difficult. If the cluster size is set too high the resultis scattered and hard to measure, if too low, the result is too crowded and hardto measure. Managers would have to go through several trail and errors to findthe suitable measurement size. I also believe that more questions may be askedto eliminate some of the biases during the data collection process.
Anotherproblem is that the analysis always segments clusters with 2 end differentoutliers, and the attribute variables are sometimes not similar but ratherextreme. For example, in question 3, I would rather segment cluster 6 and 7together than (1,6) or (3,7). It is simple but sometimes it doesn’t make sense,like placing 2 extreme clusters together. Question 6: I would recommend the company toconduct exploratory research on consumer preference/POV on PDA in general, justto see how consumer feel about the idea or generate suitable questions to askin descriptive research. Then conduct descriptive research with larger samplesize.
After they found the focus and problems of the current PDA products, theycan then improve their product to meet the demands of their targeting segment.It would best for them to just improve their market research and found outtheir suitable segment and launch a product to meet the demand of the targetingsegment.The data collected within the analysis would providethe company with a means of roughly understanding which group of people totarget but not the needs of the consumers. For now, I wouldrecommend the company to focus on their current software and hardware updatesto create better customer and company relationship to build customer loyalty.At the same time, the can continue to focus on targeting segment that areinnovative/professions/high income level consumers.