Marketing Research

1
Marketing Research

More than just books
online focus groups, surveys,
online chat rooms, data mining, ethnographic,
demographics etc.
2
Class Activity

• Subject of Research How could UAH attract more
  business majors to UAH? (services, quality,
  instructors, etc.)
• Group 1 Focus group, one moderator to
  facilitate discussion not lead discussion.
• Group 2 Questionnaire/Survey by writing 5
  open-ended questions and soliciting answers.
• Group 3 Questionnaire/Survey by writing 5
  closed-ended questions and soliciting answers.
• Group 4 Questionnaire/Survey by writing 5
  scaled questions and soliciting answers.

3
Questionnaire Design

• Open-ended questions An interview question that
  encourages an answer phrased in the respondents
  own words.
• Closed-ended questions An interview question
  that asks the respondent to make a selection from
  a limited list of responses.
• Scaled-response question A closed-ended
  question designed to measure the intensity of a
  respondents answer.

4
Marketing Research Defined

• The systematic and objective process of
  generating data for making marketing decisions.
  Ex. Determining strategic value of customer if
  there is a market for a certain product what
  customer profile to target etc.
• Online Research offers many advantages over
  traditional methods Data can be collected in a
  much shorter time and at a lower cost and w/ less
  errors in data collection.
• Studies shows online research can take ½ time and
  can save as much as 80.
• Marketing research companies ex. MarketResearch,
  IT specialists include firms such as
  GartnerGroups BI study, Dataquest, Yankee
  Group, Forrester, Meta Group, and
  Giga. Kleindl, p. 191

5
Marketing Research Providing Insight to Entire
Enterprise
Enterprise Resource Planning (ERP)
Supply Chain Management
Marketing Automation
Customer Relationship Management
Sales Force Automation
Database
Database
External
External
Internal
Database
Internal
Management Reporting Systems
Manufacturing Systems
Business Processes
Financial and Accounting Systems
Logistics/ Inventory Control Systems
Human Resource Information Systems
Kleindl, p. 203,hpc
6
Uses of Data Mining for Market Research

• Market Segmentation Aids in identifying common
  characteristics of customers who are likely to
  react to similar marketing strategies. Allows
  for one-to-one marketing practices.
• Decision Support Provides data to guide and
  support management decisions.
• Customer Churn Aids in the prediction of which
  customers are likely to switch to a competitor.
• Fraud Detection Rates the likelihood that a
  transaction is fraudulent.
• Customer Service Provide service to customers
  based on the customers past experience and
  detects patterns of customer problems.
• Interactive Marketing Identifies the prospects
  most likely to respond to direct marketing
  efforts. Predicts what each individual visitor to
  a Web site would be most interested in seeing(and
  then delivers that content).
• Market-basket Analysis Groups products or
  services that are most likely to be purchased
  together.
• Trend Analysis Aids in short, medium, and long
  term decisions by identifying differences and
  trends among groups of customers over a given
  period of time. Kleindl, p. 199

7
Pitfall of Data-Mining/Market Research

• Using ad hoc theories When odd relationships
  are found, the researcher may develop a theory to
  fit the data. Correlation is not necessary
  causation.
• Not taking no for an answer If one has a
  preconception, it may be tempting to let the
  computer search to find it.
• Storytelling It may be tempting to develop a
  story to fit the data.
• Using too many variables The more variables,
  the more likely the computer is to find
  relationships regardless of what true
  relationships exist. Difficult to focus on
  relevant data.
• Online users may not represent a random sample of
  overall population.
• Researchers must learn to use data-mining
  techniques as a tool in model building.
  Kleindl, 191, 202

8
Steps in Marketing Research

• Problem Definition
• Research Design
• Sample Selection
• Data Collection (primary)
• Data Analysis
• Report Preparation

• Online research techniques may require problem be
  related to Internet users. May not be
  representative of users.
• The Web offers easy access to secondary research.
  Allows both qualitative and quantitative data
  collection. Can show pictures, other Web sites,
  or interactivity.
• Allows the desired sample to find the site and
  answer questions at a low cost. Limits the
  randomness of sample selection and may introduce
  a self-selection bias. Online panels can use
  databases to select qualified individuals for
  research subjects.
• Can help limit interviewer errors. Logic checks
  can be built in to limit contradictory or
  nonsensical answers. Randomizing presentation of
  items can eliminate order bias.
• Automated software can analyze the data directly
  from the Web, limiting data entry errors and
  providing continuous results.
• Low cost and ease of use may lead researchers to
  minimize the problems of using a nonrandom
  sample. Charts and tables can easily be
  developed from software. Kleindl, p. 192

9
Problem Definition

• Define exactly what you are trying to determine.
  ex. Is there a market for a gourmet deli in
  certain area of town. Or Is there a market for a
  direct online wine distributor.
• If doing this research online would the sample be
  a true representation of the potential target
  market.

10
Online Research Design

• Identifies which questions must be answered, how
  and when the data will be gathered and how the
  data will be analyzed.
• Surveys a number of companies have developed
  software that enables them to place surveys
  online and then aids in the analysis of results.
  Allows a researcher to develop questions using
  categorical (a,b,c or 1-7) measures. Via email or
  online w/ CGI returns.
• Panels A panel allows researchers to pull
  respondents from a known pool. Panel members
  supply individual profile data and can then be
  selected as part of a sample pool for clients.
  This improves sample selection by allowing
  researchers to control who responds the the
  research. Ex. www.intelliquest.com
• Online Focus Groups 1) Ongoing groups use
  forums or message boards to discuss the research
  topic. Can last for days, or weeks, allows
  moderator to post follow-up questions. 2)
  Real-time chat groups allow several focus-group
  members to interact online at the same time and
  have real-time discussions. Controlled by
  moderators who can kick out unruly respondents
  into password-protected rooms. Ex.
• Kleindl 193

11
Sample Selection

• The goal in sample selection is to obtain a true
  sample of the market targeted.
• Challenging online because on self-selection
  bias, characteristics of people online (educated,
  have money, technically inclined), leaving out
  part of target market because they are not
  online.
• Types of Errors
• Measurement error occurs when there is a
  difference between the info desired by the
  researcher and the info provided by the
  measurement process.
• Sampling Error when a sample somehow does not
  represent the target population.
• Frame Error another type of sampling error
  arising if the sample drawn from a population
  differs from the target population.
• Random Error occurs because the selected sample
  is an imperfect representation of the overall
  population. Lamb, p. 269

12
Data Collection
Internets Affect on Data Collection

• Ability to collect previously unavailable
  information.
• Customer contact for surveys is more efficient,
  rapid and lower in cost.
• Ability to conduct large scale information
  gathering.
• Contacting customers groups that were previously
  hard to access is made easier.
• Observing consumer interaction is more viable.
• Krishnamurthy, p. 270

13
Primary and Secondary Data

• Primary Data Information collected for the
  first time. Can be used for solving the
  particular problem under investigation.
  Advantage is that it will answer a particular
  research question that secondary data cannot
  answer.
• Secondary Data Data previously collected for
  any purpose other than that of the current
  research project.

14
Data Analysis

• Data collected is subject to statistical
  analysis in order to make proper inferences,
  determine relationships and draw conclusions.
• Data mining the process of inductive computer
  analysis of large datasets aimed at finding
  unsuspected relationships between variables.
• Real-time data analysis data collection and
  interpretation can happen so quickly analysis can
  portray current market trends, not just
  historical data those of the past year or six
  months.
• Individual-level data info such as
  demographics, purchase history, attitudes,
  customer behavior, response to promotions on an
  individualized level.

15
Report Preparation

• Plethora of statistical analysis tools and report
  query or generation available.
• Ex. SQL, Crystal, Excel spreadsheet software etc.
• Warning Statistics can be misleading, lie
  Correlation does not mean causation. Ex.

16
Database-Marketing Process
Data mined by software that drills down pulls
info for marketing managers
Customer interacts w/ the business
Profile of ideal customer is developed and used
to target clusters
Info from transactions, warranty cards, coupons,
sweepstakes, etc. is entered in data warehouse
Marketing strategies developed promotional
strategies personalized
New data is merged w/ past and purchase data.
Redundant and conflicting data is scrubbed and
stored.
Data shared w/ local marketing effort
Kleindl, p. 201
17
Strategic Value of Customers

• Paretto Principle or 80/20 Rule Dictates that 80
  of profits may come from 20 of business
  customers.
• Strategic Value of Customer Data Database
  marketing allowing businesses to limit non
  productive reach by limiting efforts to those
  individuals who are not likely to react to a
  marketing strategy.
• RFM Measure Customer value can be based on
  recency, frequency, monetary measure.
• Majority Fallacy when a business, focusing on
  just an RFM measure, serves only the heavy half
  and avoids the lower or nonuser segment, missing
  potential customers who could become profitable
  in the future.

18
Behavioral Model in Data Analysis
Ethnographic
Past behavior such as inquiries or purchases
Future purchases
Profile characteristics demographic,
psychographics, or geographic data
Kleindl, p. 198
19
Competitive Intelligence (CI)

• A continuous process involving the legal and
  ethical collection of information and the
  monitoring of the competitive environment, giving
  managers the ability to make strategic decisions
  and become a more efficient and effective
  competitor.
• Society of Competitive Intelligence Professionals
  www.scip.org views the collection of competitive
  intelligence as a way for managers to make
  informed decisions about marketing, RD, and
  long-term business strategies.
• Fulds Internet Intelligence Index
  www.fuld.com/13/index.html contains close to 500
  intelligence-related Internet sites that link on
  topics as varied as macroeconomic data,
  individual patents, stock quote info.
  Kleindl, p. 195

20
Ethical Snooping8 Recommendations Competitive
Intelligence Trade Industry

• Observe legal restrictions. Of research companies
  and applicable laws.
• Avoid misrepresentations. Do not disguise the
  identities of the researcher and the company.
  State the clear reason for data collection
• Do not release misinformation. Do not try to
  throw competitors off by releasing false
  information. This could backfire and erode the
  credibility of the firm.
• Never ask for or exchange price information with
  competitors. This could directly violate
  antitrust laws.
• Dont steal trade secrets. This could be cause
  for legal action.
• Do not offer bribes. This is illegal!
• Do not hack other sites. This is illegal!
• Protect information sources. Do not forward
  email from a source. This could jeopardize
  reputations and jobs. Kleindl, p.195

21
Business Oriented Search Engines

• Hoovers Online publicly traded company research
• NewsDirectory.com links to the websites of more
  than 2000 newspapers.
• ClickZ This is a search and information page for
  decision makers interested in marketing related
  e-commerce issues
• High Beam Electric Library access to online
  articles and reference books.
• Competitive Intelligence Guide
• Fortune 500 Online research on fortune 500
  companies.
• Inc. 500 Database research on privately owned
  companies
• U.S. Securities and Exchange research on
  publically owned companies Kleindl,
  p. 195

22
Additional Sources of Secondary Data

• American Marketing Association
• BLS Consumer Expenditure Surveys
• Bureau of Economic Analysis
• Bureau of Transportation Statistics
• Easy Analytic Software
• EconData
• Nielson Media Research
• Service Intelligence
• World Opinion
• Columbia University's, Center for International
  Earth Science Information Network (CIESIN)
• USA Data

23
Conclusion Importance of Marketing Research

• Marketing research can help managers improve
  quality of decision making, trace problem,
  maintain and enhance existing customer relations,
  better understand marketplace, identify trends,
  gauge perceived value of goods, services,
  branding, and level customer satisfaction.
• Online Research can speed up the marketing
  research process and can result in more accurate
  information at a significant cost savings.
• Marketing Research is becoming an ongoing
  process in businesses in order to attain and
  retain competitive advantage!