MK 470 Exam 1 - Lecture notes 1-4 PDF

Preview

Summary

These are all the notes from Nita Umashankar's MKTG470 class. They include the information on the slides as well as information she talked about in class. I received an A on this exam. ...

Description

 Exam 1 Lecture and Slide notes Surveys

Data source most closely observed by managers ● Transactional ● Relationship ● Benchmarking  Limitations of Surveys ● Respondents may have a lack of memory or feel fatigued ● Are not good at following trends over long period of time or over real time ● Difficult to measure changes in the population because they are taken at single point in time ● Generally can not provide strong evidence for cause and effect  Limitations of Segmenting on Demographics ● Assumes demographics correlate with behavior ● Very little understanding of the consumer themselves (we are very diverse and different)

Marketing Intelligence

Process of acquiring and analyzing information in order to understand the market (both existing and potential customers ) to determine the current and future needs and preferences,attitudes, and behavior of the market, and to assess changes in the business environment that may affect the size  Why do we need it? ● Producers have little contact with consumers ● Channel has little knowledge of consumer preferences and attitudes, changing taste ● Need to understand competition ● Marketers need to be more accountable ● Customers are incredibly diverse and change over time and science needs to be used to capture this.

Marketing research Process

1. Agree on research process a. Problems / opportunities (MOST IMPORTANT) b. Decision alternatives and research users 2. Establish research objectives a. Research question and hypothesis b. Boundaries of study 3. Is benefit > cost 4. Design the research a. Design the questionnaire 5. Collect the data 6. Prepare and analyze data

7. Report research/ strategic recommendations Market Research

Types of Data

A critical part of marketing intelligence and helps by providing accurate, relevant and timely (ART) information ● Links consumer,public,customer to marketer through… ○ opportunity/problem ○ Evaluate actions ○ Monitor performance ○ Improve process 1. Strategy development a. Set performance objective b. Establish competitive advantage 2. Marketing program development a. product/channel decision b. Pricing c. Personal selling 3. Implementation  ● Heterogeneity - subjects being different than one another due to random chance ● Variance- measures how far each number is away from the mean ○ Helps you see the volatility (rapid changes)  ●

●

●

Exploratory (can generate all possible reasons for a problem while descriptive and causal narrow down the causes) ○ Interviews ○ Focus groups ○ Observations Descriptive ○ Trade journals ○ Popular press articles ○ Research on industry as a whole ○ Ex. Glad Trash bags example Causal ○ Internal sales data ○ Big scale survey data ○ Content analysis ○ Social media scraping

 * sample groups get larger from exploratory to causal  Data deals with volume,velocity ,veracity (uncertainty) , and variety  Types of Errors

●

Sampling error - difference between a measure obtained from

●

a sample of population and the true measure that can be obtained only from the entire population Non-sampling error- all other errors associated with a research project

Madewell (Recommerce) 

Opportunity - hailed as the new ebay and new fast fashion; morally right because you are recycling

Examples Coke vs. Pepsi Twitter Yahoo vs. Marissa Mayer   Stitch Fix (style quiz)     Smart thermometer

 ● ●  ● ●

By converting the 4 point scale to a 5 point scale it allows for a neutral Use different words (sometimes and occasionally are too similar in meaning)

 ● ●

Snapchat

●

● ● Competitors / complements

Tweets about coke outpaced tweets about pepsi by two to one by the past 90 days At her time at Yahoo the company had experienced decreasing employee morale and

Lets companies know areas where the flu is most occuring to generate ads to them more often Ethical vs privacy issue Snapchat is experiencing an inefficiency stage where they have the most users compared to Twitter but yet are making less ad revenue Longitudinal - everybody is likely to change over time Heterogeneity in behavior



          

Sources of Secondary Data Secondary Data

Limitations

● No control over data collection ● Collected for another purpose ● May not be accurate and may be outdated ● May not meet data requirements Benefits ● Low cost and effort ● Sometimes more accurate ● Sometimes only way to obtain data  Internal sources of secondary data may include accounting data, sales reports, inventory management, customer database  External sources may include online databases, trade directories, and published data sources  Different types of data

●

●

●

Cross-sectional - data at one single point in time across different samples ○ Little or at no expense ○ Does not tell you the cause and effect Longitudinal - tracks the same sample at different points in time ○ Not a lot of change or variance is occuring in between customers Panel - sample taken at multiple points in time across various customers

 Static vs. Dynamic ● Static - the relationship between the different types of data does not change ○ Fixed size and are predictable ● Dynamic - structure grows as continuous information is required Unilever Ice Cream

● ●

Capitalize on trend while trying to stay competitive Looked at their supply chain to see if it would be feasible

Observational Research (when do you use each?)

●

Scanner data is used to look at behavior patterns ○ Independent variable (promotion, private vs generic, customer journey, complementary items) ○ Dependent variable (total purchase amount, variety seeking, interpurchase time) Home audit (Qualitative approach) is when companies come into your home and see how you use the product over time ○ Can help them see how product is being used and use the information to further market the product Mail Diary is panel data that customer pen down to describe

●

●

how they use the product over time ○ Negative - selection bias may occur ■ You need a representative sample to avoid this ○ Customers may have a hard time recalling what they might have done with product Purchase Panel Data

Limitations ● Selection bias - proper randomization of sample group is not achieved and not representative of the entire population (the whom?) ● Mortality effect - the loss of subjects as the research continues throughout time. ○ Can fix this by creating incentives for the subjects ○ Can occur in Panel or longitudinal  ● Testing effects (demand effect) - giving you the answer you want to hear or the answer that reflects my ideal self ○ There should be no leading questions and answers should not be public but more intimate and anonymous Benefits ● Analyze heavy buyers and their purchasing habits or characteristics ● Brand switching rate (loyal rate) ● Repeat purchase rate



Qualitative Insights Focus Groups vs. In-depth

Dominos Focus Group

Benefits (Focus Groups) ● More stimulation than an interviewing ● Less pressure ● Larger sample size Negative ● Random selection ● Statistical reliability (overall consistency of smaple) ● Projectability of research results ● Groupthink  *in-depth interviews might be under testing demand effect ● ● ●

Changed perception, product, and process Merged to e-commerce by providing digital experience Used A/B testing against every process they implemented

 Negative components User generated content

How to quantify UGC

Different types of measuring scales to capture attitude

● ● ● ●

Themes (keywords used in reviews) +valence Ratio of negative to positive words Total words/ total sentences Conditional because it is dependent

●

Nominal - objects assigned to mutually exclusive labeled categories that have no necessary relationship ○ Ex. Are you a resident of california? Race? ○ If you are not in one bucket you are in the other ○ Yes or No questions ○ Losing a lot of information Ordinal - ranks objects or arranges them in order by some common variable ○ Does not provide info on how much difference there is between options ○ Arithmetic options limited to median or mode Constant sum scale - respondents allocate a fixed number of rating points among serial objects to reflect relative purchase Ratio - type of interval scale with meaningful zero point ○ Is possible to say how greater one thing is from the other ○ Only scale that permits comparisons of absolute magnitude ○ Ex. what is your income? How old are you? Interval - numbers used to rank objects also represent equal increments of the attribute being measured ○ Differences can be compared ○ Entire range of statistical operations can be employed for analysis ○ 1-7

●

● ●

●

 *an unbalanced scale is not always wrong but depends on the variance of the topic and if people lean more toward positive or negative (traffic or complaint forms) Multiple- item scales

Types of Attitudes

Measure a sample of beliefs toward the attitude objects and combine the set of answers into an average score ● Likert scale - correspondents answer to which degree they agree or disagree with a statement ● Semantic-differential scale - respondents rate each attribute object on a number of five or seven point rating scales bounded by polar adjective or phrases ○ High price vs. low price are not trade offs and so all variance will go to one side of the scale  Attitudes are mental states used by individuals to structure the way they

perceive their environment and guide the way they respond to it ● Cognitive (Knowledge) component ● Affective (liking) ● Intention (action) Standardization of measurement

● ●

 

One to one - correspondence between the symbol and the characteristic in the object that is being measured Rules for the assignment should be invariant (never changing) over time and the objects being measured...