Latest Advances in Internet Market Research

1
Latest AdvancesinInternet Market Research
Trip Trepagnier HCD Research, Inc.
2
Agenda

• Evolution of Internet Research
• Sampling Issues
• Current Applications Limitations
• Case Studies
• Cutting Edge Programming
• Summary

3
Internet Research

• Background

4
Early Internet Research

• Began Circa 1995
• Experimental Studies
• Quick Turnaround Projects

5
Limitations
Early Internet Research

• Short Surveys with Simple Questions
• Largely Limited to US Respondents
• Technical Problems
• Inexperienced Users
• Confidentiality Concerns
• Convenience Sample (Opted-In Panels)
• Sample Bias Concerns

6
Early Uses of Internet Marketing Research

• 1995 Internet Services Commercially Available
  to Health Care Companies
• Applications Experimentation, Rush Projects
• Sampling Available e-Mails (5-10 of physicians)
  from Target Market Segments
• Formats Simple Yes/No Questions, Multiple
  Choice
• Reporting Post Response, Traditional Reports

7
Current Uses of Internet Marketing Research

• Applications Experimentation, Rush Projects
• Classical Marketing Research Techniques
• Sampling Available e-Mails from Target Market
• Available Universe from Target Market
• Formats Simple Yes/No Questions, Multiple
  Choice
• Skip Logic, Adaptive, Randomization, Timed
  Intervals, Images, Animation
• Reporting Post Response, Traditional Reports
• Real-time, Interactive, Stat Analysis

8
Market Segments

• Medical Professionals
• Consumers/Patients
• Business to Business

9
Current Internet Applications

• Quantitative Research
• Product Development
• Product Concept Testing
• Pricing Research
• Awareness/Trial/Usage Studies
• Message Development and Testing
• Communications Concept Testing
• Post-Launch Tracking and Effectiveness
• Sales Force Effectiveness/Message Recall
• Qualitative Research
• Online Bulletin Boards
• Tele-Web Discussion Groups
• Web-Assisted One-on-One Interviews

10
Internet Limitations

• Quantitative Research
• Very Large Samples of Physicians
• Consumer Research in China
• Local/Cultural Issues
• Qualitative Research
• Product Demonstrations
• Hands-On Exercises
• Facial Expressions/Body Language

11
Internet Usage
12
World Internet Usage And Population Statistics
Percentage
Growth Rate 20002005
Population numbers based on data from The World
Gazetteer. Internet usage from Nielsen/NetRatings,
the International Telecommunications Union and
www.internetworldstats.com
13
Internet Usage
Selected European Countries
Population (MM)
59
63
49
43
39
Growth Rate 20002005
Population numbers based on data from The World
Gazetteer. Internet usage from Nielsen/NetRatings,
the International Telecommunications Union and
www.internetworldstats.com
14
Internet Usage
Selected North/South American Countries
Population (MM)
68
14
68
16
26
Growth Rate 20002005
Population numbers based on data from The World
Gazetteer. Internet usage from Nielsen/NetRatings,
the International Telecommunications Union and
www.internetworldstats.com
15
Internet Usage
Selected Asian Countries
Population (MM)
67
67
68
9
5
69
Growth Rate 20002005
Population numbers based on data from The World
Gazetteer. Internet usage from Nielsen/NetRatings,
the International Telecommunications Union and
www.internetworldstats.com
16
Sampling Methodology
17
Two Ways of Sampling

• Old Approach
• Use e-Mail Addresses Obtained from Panels,
  Relationships with Web Sites or from Past
  Projects
• Current Approach
• Randomly Sample from Universe of Individuals
  Representing Target Group

18
Sampling
Medical Professionals

• Random Selection from Universe of Health Care
  Professionals
• Ability to Match Prescribing Segments to Contact
  information for any Physician Segment
• Rigorous Screening of Opted-In Registrants

19
Current State of Sampling

• Assumption gt90 Physicians in Western World Can
  Access Internet
• Crucial Aspect of Execution
• Random Sampling from the Universe of Target
  Segment Minimizes Bias
• With Proper Sampling, Diffusion of Internet Use
  by Professional Segments Eliminates Sampling
  Error Concerns
• Unless Project Requires Response in 24 Hours,
  Should Always Sample The Universe Of Interest

20
Regional Differences
Europe

• Limited Physician Population by Specialty and
  Country
• Greater Use of Telephone for Recruiting
• Longer Time in Field
• More Frequent Use of Panels

21
Sampling
Consumers/Patients

• Cast Wide Electronic Net
• Screen All Opted-In Registrants
• General Consumer Databases
• Special Interest Databases

22
Niche Markets

• Hard-to-Reach Patients and Caregivers
• Oncology
• Cardiovascular
• Respiratory
• Target Specific Patient Types using NexCura
• Detailed Clinical Information
• Hundreds of Data Elements and Variables
• Over 900,000 Registrants, 80 Opted-In
• 600,000 Oncology
• 250,000 Cardiovascular
• 50,000 Respiratory

23
Reporting Results
24
Reporting

• Traditional Comprehensive Analysis and Reporting
• Online Real-Time Interactive Reporting
• Basic Statistics
• Cross Tabulations
• Raw Verbatim Comments

25
On-Line Reporting Examples
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
Case Studies
32
Quantitative Case Studies

• Conjoint Analysis
• Choice Model
• Message Optimization
• Creative Concept Testing

33
Conjoint Case Study

• New Medical Instrument
• Learn Value of Product Features
• Understand Price Sensitivity
• Determine Optimal Set of Attributes
• Conjoint Example

34
Conjoint Analysis Results

• Relative Importance of Product Attributes
• Part-Worth Utilities of Each Attribute
• Ability to Perform Simulations
• Market Share under Various Scenarios
• Measure Price Sensitivity

35
Conjoint Example
Relative Importance of Product Attributes
36
Conjoint Example
Part-Worth Utilities of Each Attribute
Processing Throughput per Hour
37
Conjoint Analysis Simulation Model
38
Choice Allocation Model

• New Drug Development
• Test Copy for Product Labeling
• Learn Value of Alternative Labeling
• (Two Label Categories, Two Sets of Copy)
• Choice Model Example

39
Choice Model Results

• Model of Individual Preferences
• Aggregate Preference Share
• Aggregate Sales Volume

40
Choice Model Example
Aggregate Preference Share Analysis
41
Communications Research
42
Communications ResearchCase Studies

• Message Development/Optimization
• Concept Testing

43
Message Optimization Case StudyNew
AnalgesicOrthopedic Surgeons, Rheumatologists
and PCPs
44
Message Optimization

• Three-Phase Process
• Ideation
• One-Dimensional Screening
• Multivariate Evaluation

45
Phase One
Ideation/Message Development

• Create Database of Descriptive Elements
• Categorize Message Elements
• Text Messages
• Clinical Trial Data
• Product Images or Charts
• Competitors Statements and Images

46
Creation of Message Elements and Categories
47
Phase Two
One-Dimensional Screening

• Eliminates Ineffective Elements within Each
  Message Category
• Targeted Respondents Evaluate Each Element
  Individually
• Reduces Number of Messages to be Used in
  Full-Concept Analysis
• Minimizes Respondent Fatigue
• Rating Scales Based on Ad Objectives

48
Screening Output
--Ratings of Lowest Efficacy Elements for
Product A--
49
Screening Output
--Ratings of Top Ten Efficacy Elements for
Product A--
50
Phase Three Multivariate Evaluation

• Determines Optimal Combination of Messages
• Respondents Evaluate Complete Concepts
• Message Optimizer Derives
• Relative Importance of Message Categories
• Value of Each Individual Element

51
Phase Three Deliverables

• Derived Relative Value of Each Individual Element
• Derived Relative Value of Each Category
• Message Value Simulation

52
Phase Three Deliverables

• Message Simulator
• Provide Tools to
• Perform What If Scenarios on Simulator
• Create Combinations that Reflect the Target
  Segments Hot Buttons
• Change and Re-evaluate Messages and Changes in
• Market New Indications
• Competition Regulatory Changes
• Pricing

53
Phase Three Sample Deliverables
Demo Worksheet
54
Print Concept Testing
55
Communications Model

• Breaking Through/Getting Attention
• Communicating Message
• Affecting Emotions
• Changing Attitudes
• Influencing Behavior

56
Emotions

• All Marketing Communications have an EMOTIONAL
  Component
• Even for the Most Rational Messages
• Measuring Emotional Impact
• The Ayer Emotion Battery
• Quick
• Simple
• Validated

57
Ayer Emotion Battery
? All Emotions Felt ? Emotions Felt Most
Strongly
Ask

• Not bored, attentive

• Active, aroused, playful, entertained, excited

• Wishful, desirous, hopeful

• Benefited, enlightened

• Affectionate, loving, innocent, pure

• Competent, confident, in control, proud

?Positive Vector ?

• Virtuous, reverent, moral

• Serene, restful, pure, soothed

• Sorrowful, thankful, afraid, regretful

• Worshipful, reverent

• Sad, angry, powerless, helpless, distressed

• Amazed, astonished

?Negative Vector ?

• Spiritual

• Tense, conflictful, anxious

• Ashamed, embarrassed, mad, guilty

• Suspicious, skeptical

• Confused, uninformed

• Disgusted, irritated, enraged, contemptuous

58
Ad Test Case Study

• Marketing Research Firm Developing New Sales
  Collateral
• Testing of Communication Messages Completed
• Next Step Quantitatively Determine
  Effectiveness of Different Brochures

59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
Ad Test Case Study

• Methodology
• Sequential Monadic Design
• Split Sample into Three Groups
• Conduct Monadic Test First, Repeat Process for
  Remaining Two Ads, End with Paired-Comparison
• Measure Breakthrough, Communication,
  Believability, Impact of Key Words, Overall
  Effectiveness
• Demo Survey

63
Test Results

• Breakthrough and Saliency
• Communication, Believability, Relevance and
  Effectiveness Scores
• Message Communicated
• Focus of Attention
• Headline / Copy Stopping Power
• Emotions Evoked
• Verbatim Responses

64
BreakthroughNumber of First Mentions (Concept A
n49, Concept B n50, Concept C n51)
Number of Mentions
65
Focus of Attention
66
Message Communicated Through Graphic Image in
Concept B (n50)
67
Headline / Copy Stopping Power Concept B (n50)
Participants were allowed to select more than 1
word from the headline
68
Ayer Emotion Battery

• Emotions Evoked while Viewing Ads

69
Ad B tested strongly for resonance in several
high positive vector emotions, and generally
produced appropriate and positive emotional
responses.  Highlights included Wishful,
desirous, hopeful 63 and Benefited,
enlightened 51.
Emotions Felt Overall While Viewing Ad B --
Proportion of Respondents Reporting --n50
Q20. Please select all the emotions that you
felt while viewing the advertisement. Note
Emotion factors are ordered from positive to
negative leveragability.
70
The emotions respondents felt most strongly
parallel the emotions felt overall. Specifically
Wishful, desirous, hopeful 32 Benefited,
enlightened 30
Emotions Felt Most Strongly While Viewing Ad B --
Proportion of Respondents Reporting --n50
Q21. Please check the one emotion that you felt
most strongly while viewing the ad. Note
Emotion factors are ordered from positive to
negative leveragability.
71
Positive/Negative Emotional Vectors (n150)
Concept A
Concept B
Concept C
Percentage of Responders
72
Most Effective Brochure (n150)
Concept A
Concept B
Concept C
Percentage of Responders
73
Why Brochure is Most Effective(n150)
74
Example of Verbatim Responses
In a few words, what was the main point that the
ad was trying to communicate?

• Partial compliance looks like a small problem on
  the surface, but can be a bigger problem
  underneath.
• Partial compliance and increase with new med
• Noncompliance is a significant obstacle to
  recovery
• Partial compliance becomes a much bigger problem
  over time.
• Medication compliance is large problem that is
  mostly hidden from view
• Most patients do not receive the full benefit
  because of their partial compliance
• Non-compliance is a big, largely hidden problem
• The problem is not as evident as we think, hidden
  factors and consequences are real hiding under
  the surface
• Noncompliance poses a big problem to the
  effective treatment of schizophrenia
• Recognized poor compliance is only 'the tip of
  the iceberg'
• Compliance is a bigger problem than you might
  expect.
• The biggest problem with medication treatment of
  schizophrenia is partial compliance.
• What we know of non compliance is just a tip of
  iceberg
• How many people become non compliant with their
  medications and its effects
• Large problems generally hide beneath what appear
  to be less significant problems

75
Storyboard/Video Applications

• Ice Age 2 Trailer
• Measure Overall Effectiveness, Message
  Communication, Likeability and Kawai
• Employ Real-Time Probes to Understand Reactions
  to Key Scenes
• Demo Survey

76
Active Interest Graph Output
77
Active Interest Graphs Kawai
ltAge Group Totalgt
In the end the little burrow reached his
mommy/animal audience
Possum catapults in the air/hits the tree
Manny/Ellie/ Finding true love in the most
unlikely places
Retract the claws please
Fish breaks through ice/ Were gonna die
Then where is your big happy family?
Scrat catapults into the air
Scrat fills up with water and shoots off wall
of ice
Why does the burrow go home?
Vertical lines indicate scene breaks (Green Just
Beginning, Red Destiny)
78
Summary

• Internet Well Suited to Quantitative Research
• Execute Complex Studies
• Reach Wide Audiences
• Ideal for Print and Animated Concepts
• Meets Respondents Schedules
• Creates Interactive Environment to Understand
  Reasons Why

79
Thank You!
80
HCD Research, Inc.260 US Highway
202/31Flemington, NJ 08822908-788-9393hcd_at_hcdi
.net