Week 13 Question: Due April 19
Answer one of the following questions as a comment on this post.
Submit your contribution by 11:00 AM on Thursday, April 19. It should be insightful (don’t just repeat back facts from the class), but it doesn’t need to be more than a few sentences. Most of all, I want to know what you think!
- Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
- Let’s say you performed a cluster analysis and came up with a series of market segments, describing the demographics of three different sets of customers for your business (big spenders, average spenders, small spenders). Your manager is reluctant to trust your analysis because she has a lot of experience and this contradicts what she believes is true about her customers. How would you convince her that your analysis is correct?
David Schuff
2) If put in this situation I would sit down with my manager and show him that the analysis derives from statistical study of a certain sample size and is backed by evidence. If your manager doesn’t trust your analysis because his past experience says otherwise, I would explain to him that different trends tend to change and that you cannot always rely on past data analysis. In order to further prove your point, you can get a greater sample of people to show him that your data is accurate.
Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
A car dealer could use clustering to assess the affluence levels and personal preferences of their customers. For example, BMW could use clustering to assess which type of customers are more inclined to purchasing the different series (3,5, and 7) of cars. Looking at this clustering can identify trends and patterns that will allow BMW to tailor specific features in the different models of their cars. Or, they could use this information when launching the new yearly model of the cars.
Question 1: I think one of the good examples could be the students’ SAT scores. For example, some students are better off in Math, English, or maybe just in Writing. So if we used the clustering analyzing, we can perform the analytical data just to figure out which candidates would the colleges or universities pick the most of. More importantly, I believe that each university rated the SAT scores differently. This is because some wants more into Math candidates or may just the Writing candidates. That is why if I can used the clustering data, then I might use it on the students who have taken the SAT test.
#1. A cluster analysis is an effective tool for business marketing. For example, how does an airline business allocate a limited budget to target the right segmentation of customers in off-peak season? By grouping related or different data, an airline company might track back customers’ orders, with not only the right demography, but also different detailed categories about their income, age, location, order data, and how they order tickets. Do they shop on line or from local agents? Therefore, it makes sense for the company to either focus on local media or the digital market.
The analysis question I would like to answer: What kinds of students are more likely to use Tech center as a place to study and what is the number of hours spent there for each segment. In order to find out about the potential members of each segment I set the x-axis that it will represent the number of hours spent and y-axis will represent a frequency of use. The sample will consist of all students and the variable will be based on students GPA (initially to find out if there is any correlation, later there is a possibility to assign other variables to find out a common variable for objects in certain cluster). By performing a clustering method, results will show how the frequency and amount of hours spent in Tech center correlates with the variables of objects.
#1
A business question that could be answered using clustering is for class scheduling and deciding how many classes there should be for certain subjects. The information that could be taken for this would be student ID, major, year, and prerquisites. This can determine how many students would be registering for a specific class and able to determine the approximate amount of openings for each class.
One business question that could be answered using clustering is if a Wendy’s wanted to know the most popular item ordered on their menu. Management could create a cluster for each Wendy’s chain. The next step would be to chose some of the clusters and selcet a sample of people from that cluster. By doing this they would be able to keep track of each customer’s order and determine which item on the menu is most popular.
Clustering could help us to answer this question: How many and which specialty doctors should be in a hospital on a daily basis? Clustering will help us to answer this question because it collects the data and analyze it. This analysis could help us to figure out how many patients come to the hospital everyday and also what is the specialty that most patients demands. Thus, helping the owners of the hospital to save money because doctors won’t be wondering around without anything to do.
2. Cluster analysis is an extremely effective and reliable tool. Cluster analysis takes historical data and uses statistical analysis that is used by many different industries with positive outcomes. Cluster analysis uses sound data and gives the best possible probabilities for said industry and, is more often than not, correct in its conclusions. This type of analysis can potentially save a company a material amount of expenses used in otherwise, expensive and possibly inefficient methods.
1. Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
I currently work at the writing center on campus and my directors could have used clustering to assess how many tutors to have working at certain hours of operation. They would use data such as students schedules and certain times in the semester that are most pressing for assignments and papers to determine the busiest hours of operation and how many tutors to have during certain hours. For instance, perhaps mid day would mean having more tutors and early in the morning having just a few. Also they could look at information such as majors to determine which students are likely to even visit the writing center.
1. A business question that could be answered with clustering is determining what the most popular products for a retail store like Best Buy were. Each location would be a cluster and a sample of those clusters would provide people who could be selected and tracked. They could track what items people buy the most of (or additionally, when and with what other items, etc) and determine what is most popular.
1. Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
An example of clustering is used at my old job as a pharmacy technician to allocate workers and business hours. During our busy season between mid-September and mid-May (usually during a school year), we’ve noticed an increase in number of customers and, unfortunately, an increase in the amount of wait time per prescription. Days of the week are also evaluated during this “peak season” to further allocate resources and to prepare inventory. For example, Mondays, Fridays, and Saturdays are the busiest days of the week for us, and therefore we need more employees year round, especially between mid-September and May. Consequently, we need to be sure that our inventory is able to handle the larger influx of business on those days to better serve our customers. We collect this data by running daily, weekly, and monthly reports of the number of prescriptions that are filled (new RXs and refills) and by running a bi-monthly report of our most frequently used items behind-the-counter. Other reports include frequency of ordering over-the-counter items such as cough drops during winter months and poison ivy lotions during spring and summer months.
Your manager is reluctant to trust your analysis because she has a lot of experience and this contradicts what she believes is true about her customers. How would you convince her that your analysis is correct?
If my manager does not trust my analysis, I would have to include her in the process of performing the cluster analysis. If she had a better understanding of the demographics of the situation and the steps to take to perform the cluster analysis, we could be able to come to an agreement.
1) When insurance companies charge premiums depending on an individual’s expected loss, they set up homogeneous pools that gather people who have similar risks. Insurance companies can use clustering to group similar individuals in the same risk pool. The data they would collect can be age, credit score, gender, medical history, location, habits (smokers/non smokers)..
1. Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
Grocery stores such as Whole Foods could answer business questions using clustering. They could determine which types of customers are more likely to shop at their stores. They could collect data such as age, household income, sex and zip codes which would allow them to asses where new Whole Foods locations should be opened to help grow their brand. Additionally, they can assess which products sell most and in what locations. This will assist them in keeping heavily purchased items in stock and accessible to the consumer.
1) A business question that someone could use clustering to answer could be where should a band or musical artist make stops on their tour. To perform that analysis they would be able to look at past profits, fans at a location, venue capacities, and the costs associated with different locations, and age distributions of the population.
1. Store like BestBuy could use clustering to identify the different types of ereader. They could cluster the ipad ereaders and Amazon kindle ereader into different categories. The data collected would include type of ereader bought, age, gender, occupation, whether they commute or drive to work, because usually people that commute buy amazon kindles over those who drive. So using clustering, they can develop target advertisement, etc.
Question 1.
A business question using clustering would be determining demographics between a set of people or a certain category. By looking at a group of clusters, one can determine a certain set of demographics such as age and demographics. An example of would be: if I wanted to start a car repair shop business. I could look at the group of clusters and determine the people who are old enough to drive cars to see how profitable my business can be.
Name and describe business question that you could answer using clustering. What data would you collect to perform the analysis?
Where are the best locations to establish a company’s supply chain warehouse? Cluster analysis can help determine which where their customer base is located across the country. From there the company can analyze what is a sufficient number for warehouses and where are the optimal locations they can establish their warehouses. The location of the customers would be the main data collected where the actual distance from the center of the cluster would represent actual distance.
1. Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis?
A credit card company could use clustering to help target their customers. For example, through clustering analysis they can figure out the probability of different affluence levels or purchasing patterns to a certain type of credit card. In that way, they could have a better idea whether the person they are sending the information to is interested in that type of credit card.
2. I would convince the manager that the cluster analysis is reliable because it uses statistical analysis. Having past experience is not enough for the manager to predict anything because trends can change over time. Cluster analysis utilizes numbers from statistical studies to make conclusions, so I would consider this to be more reliable than a person making conclusions off of statistics. Cluster analysis can help the manager avoid any problems of predicting incorrectly based off of what they think they know.
#1
Rental companies such as Zipcar and Phillycarshare can use clustering to determine population densities and hence where to place their cars. They could also base their placement on crime ratings, so there is less of a chance of their cars getting broken into. Their ideal locations are highly dense, middle income, low crime areas that can all be determined through clustering with accurate data. They could also base their decision based on median income per a certain radius, or based on median age per certain radius because the younger population would tend to rent more cars as well
A watch company could use clustering to do targeted marketing stategies for watches in different price levels. It could collect data like income, and affluence grade. They could now customize ads for different clusters of customers depending on what price level of watch they would most likely buy.
1. Cluster analysis is helpful for the company to develop a best marketing mix. When customers purchased a same types of goods or services, different customer have different consumption patterns. For example, customers’ buying motives can determined by the needs, cognition, learning, cultural, social, family, small groups. In addition, customer satisfaction level and repurchase opportunity also can be confident in. The different of customers can divided into the interests of customers, barriers of choice, bargaining power, and so on. To learn about patterns, company can develop a different marketing mix to maximize its profit.
Question #1
A business question you could answer using clustering would be what premium to charge customers for an auto insurance policy. Insurance companies would collect data such as: age of the driver, gender of the driver, location of the driver, use of the vehicle (business or pleasure), type of vehicle, age of vehicle, and of course accident/violation history. All of this information would allow insurance companies to separate customers into groups which would determine their respective levels of risk. This in turn determines the amount of their premium.
Answer #1
I think the cluster analysis can be very useful around seasonal periods such as Christmas for example, to determine a person buying patterns and realize what trends has taken place. What are children most likely requesting from their parents this Christmas vs. last year purchases that were considered popular. Since there are so many brands to choose from and to be able to narrow down to what’s hot and what’s not that’s where clustering can come in to play. To zero in on the marketing trends and segmentations of when, how many, what do the customers buy more than other products out there. I they buy a particular toy that usually people who buy that specific toy will lead u to determine whether they may purchase another toy that usually parents buy for their kids together because your able to cluster and determine buying patterns. Learn the market trends and analyze them more.
A business question that one could answer using clustering involves stakeholder analysis. When making business decisions that could affect the future of a given company, it is important for that company to carefully examine everyone and everything that will be affected (both positive and negative side-effects). Through clustering companies can distinguish which stakeholders are dominant and which ones are dormant. If they discover that a particular business action will negatively affect a definitive stakeholder, they may want to change their plans and/or cancel them entirely. Data that could be collected to perform this analysis includes the power held by the entities, the urgency of their individual situations, and the total time during which the stakeholders were a crucial part of the organization.
#1. Clustering could be used in many business situations, especially when a company offers products in multiple market segments. For instance, a company like Sony produces electronics for many different types of consumers. They could use clustering to efficiently market products such as computers to different market segments, such as a performance or high-technology segment. By looking at variables such as affluence level, age, and gender they could better tailor marketing efforts through clustering groups of customers and targeting those who fall in a certain range, most likely to purchase that specific product.
2) Let’s say you performed a cluster analysis and came up with a series of market segments, describing the demographics of three different sets of customers for your business (big spenders, average spenders, small spenders). Your manager is reluctant to trust your analysis because she has a lot of experience and this contradicts what she believes is true about her customers. How would you convince her that your analysis is correct?
I would tell her that with all due respect, this is a concrete quantitative analysis and I would assure her the statistics are correct. If the algorithm is run correctly, and the plot shows clustering that has a strong statistical significance, then it would be prudent for us to move forward with these results. I would tell her that we are essentially leaving income on the table and wasting marketing dollars by no segmenting our customers this way. Finally, I would create a business plan and present it to my manager and propose that we run a small scale trial run to see if it is effective.
#2
First I would show her why to my best knowledge the raw data is accurate. Then I would show how, when graphed, the data is clustered together. I would then explain to her that the K points let us associate data points that are close to each other. I would explain why some of the data may look off to her because some points that are close to two centroids may be put in one cluster when they could almost fall into another.
Cluster Analysis could help a company figure out what group of people is purchasing their products. For example, A business question we can use a cluster analysis for what group of customers are buying certain brand of laptop. Students may usually buy a brand of laptop, and business people would usually buy another brand. A cluster analysis can bring all this information together and show what your different clusters are.
First of all, I will use validation tools to see if my data is accurate and mistake did not occur. I think in such situations, you have to look at data and at the same time use your judgment and reason before making a decision. If my manager has lots of experience in her field, then she probably knows a lot about her customers. In this situation the best way is to collect a new sample data and verify the correlation between variables as well as the correlation between sales and each variable individually. A new sample will decrease the chance of average error and error by chance.
The example of using, the example clustering is if you running an ebay store to selling computers, in order to find out how many you sell per month and which model is the most popular. The information that could be taking would be computer’s price, color, and model. This could also determine that which one is the best seller.
#1 Clustering analysis can be used for many market research tools to understand the correlation among a certain set of characteristics to answer a business question. For a small company selling something unique to a niche market they could benefit from using cluster analysis. There is a lacrosse store in my town that opened up a couple years. Deciding on a geographic location to open up they probably conducted cluster analysis to analyze the schools, teams, success of teams, and affluence of different areas. When they saw these characteristics displaying significant patterns in my area, they most likely decided to open up here in comparison to other area they were deciding on.
#1. A business situation that could use clustering could be when a retail company sends emails to their customers. They could group their customers based on various factors like age, location, how often they buy, etc. and then based on these factors, send different marketing promotions or product information to them. The clustering will determine which customers are more important to market to and keep up to date on products, and which customers might need that extra incentive or promotion to come in.
Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
A business question that one could answer by using clustering is where their next store location should be. When choosing between two different locations a store can use clustering to decide which location would suit their store best. They can use information about the residents located in the neighboring towns such as income, amount of household residents, age of the residents etc.
An individual Walmart store could use cluster analysis to better allocate inventory. By looking at age, gender, and ethnicity of the area around the store, they would have a better idea who their primary customers are and who their primary market should be. This way they could expand marketing towards that specific group and increase sales. Now their inventory could be tailored toward the specific needs of the area. ie. more baby clothes, less mens clothes. More diapers, less teenage toys. More Spanish food aisles, less bakery foods and deli items. This cluster would provide an explanation for past sales trends and help fuel future growth.
1. Name and describe a business question that you could answer using clustering. What data would you collect to perform the analysis? Don’t use an example we’ve covered in class.
Stores such as Targett can utilize clustering by finding, which items sell best at certain times of the season. Target can collect data such as age, race, gender and affluence grade to determine, which types of customers will purchase certain products at specific times of the year. In a broader sense, it will help stores operate efficiently to determine how much inventory is needed for a specific product.
#1
A good business situation that could be used for clustering is if a busibness were to look at what type of cars people were most likely to buy when looking for new cars. They could look at the demographics and see what the age was of people who were more likely to buy a brand new car that was equipped with everything opposed to a car that came standard, also what class the car was. Also, they could look at if more people were more likely to buy a car, truck, or SUV for their preferable automobile; and also they could see if the area that people lived in made a difference in the type of automobile they purchcased. This type of clustering could help a dealership know approximately how many of each type of car they should have shipped in.
Clustering would be very useful in medical or law school application processes in order to segment classes of students by their interests listed on an application in addition to specific scoring on different sections of an exam. While an applicant may show interest in cardiology, by clustering exam results by question theme it may steer the student in a direction that they will be more successful in and have a higher aptitude for learning in.
In turn, this will allow for a medical school to increase their chances of extremely successful students being able to fund the school in future years due to better prepared and properly assigned students.