How to Present Research Findings to a Nontechnical Audience

1
How to Present Research Findings to a
Nontechnical Audience

• Tom McGoldrick
• Vice President, Consulting and Business
  Development
• Questar CEM
• May 21, 2009

2
Who is Questar?

• Established 1985 24 years in Customer
  Satisfaction research/consulting
• Remain one of a select few companies focused
  exclusively on customer experience and
  satisfaction
• 120 professionals based in Eagan, Minnesota, USA
  
• Loyalty building through the customer experience
  is singular focus
• Serve multi-unit restaurant, retail, and
  hospitality organizations
• Experienced in unique characteristics of
  franchise environment
• Offer value added focus for Operations, Research,
  Marketing, Training Development, and Human
  Resources

3
Start with Survey Design

• Every response option needs a word
• Targeted research questions
• Be sure to include relevant demographics needed
  for analysis
• Enough sample for each important unit of analysis

4
Tell a Story

• People are hardwired to listen to stories.
• When listening to a story you automatically place
  yourself in it.
• The less technically experienced your audience,
  the less you can use numbers and jargon.
• Try to remember that most people didnt take
  college statistics, and those who did hated it.
• If you took it from me, I apologize.

5
Believe You Are the World Expert on Your Topic

• Your audience is there to learn from you
• Draw conclusions and make recommendations
• Guess with authority
• Why is the research being done?

6
Encourage Questions and Conversation

• Questions are your best tool for gauging interest
  and understanding.
• When people are talking you know they are
  thinking about the information.
• Most people have limited tolerance for lectures.
• Some people cant understand new concepts unless
  they can discuss them.

7
The Great Handout Debate
When should you handout copies of your
presentation?
Results from a Recent LinkedIn Poll
8
The Great Handout Debate
Gender Difference?
9
The Rule of Three

• Tell them what you are going to say
• Provide basic summary at the beginning
• However, you MUST stop yourself from going into
  the details there and force yourself to wait
  until you get to the actual detail slides to
  cover those points.
• Say what you have to say
• Tell them what you said

10
The Other Rule of Three

• Enough material they disagree with to create the
  illusion of control
• Enough material they already know to validate
  your expertise
• Enough new material to prove your value

11
Be Careful with Statistics

• Dont prove Mark Twain right
• Lies, damned lies and statistics
• The audience wants to know
• The sample is valid
• When differences are significant
• How to read the chart
• What conclusions you draw

12
Simple Charts
This slide was in a presentation by a keynote
speaker from City University in London. This is a
flowchart related to the causes and impacts of
obesity in Britain.
13
Beware of PowerPoint

• It is too easy to

Over Design
And
Over Animate
14
Choose Analytical Tools Carefully

• Percents not Means Average (unless you really
  enjoy telling people that Mean means average)
• 65 Very Satisfied vs. an Average of 4.78
• Chi-square, t-test, P .034897
• Most people just want a rule of thumb
• More than 4 point difference in percentages
• Color code and trust
• The Ultimate Question by Fred Reichheld
• Larger scale makes the math work well
• However, the story is what makes it compelling
• Promoters and Detractors

15
Linkage Research
16
Brand Consistency Curve
4th QTR 08
1st QTR 09
17
Penalty Reward Profile
18
Be Careful with Charts

• Charts up and tables down
• Consistency matters
• Be very careful when using different scales
  during a presentation
• Color matters, be consistent
• Only change formats for a reason (pie, bar,
  line)

19
Avoid Jargon

• Actually, avoid your jargon, use theirs.
• Customer/Guest/Client
• Revenue/Managers Operating Profit/Margin

Kendall Tau
CAHID
Multicollinearity
Bimodality
Kurtosis
Orthogonal
homogeneity
ANOVA
Newman-Keuls
Voronoi tessellation
Chi-Square
20
Use Relevant Examples

• Reliability vs. validity
• Bathroom scale
• News polls tend to have only 1100-1300 responses
  and have a very good track record
• Ask for examples
• Most people will try to put data into real life
  context and can come up with an example when asked

21
Sound Bites

• Use comments from the survey to flesh out and
  demonstrate the conclusions you draw.
• I saw an employee use the same mop to clean the
  floor and your tables!

• Awards Won
• Bronze Star
• Emmy
• Fulbright
• Guinness Book of World Records
• Numerous Other Military decorations
• Many literary, academic and research awards

22
Summary

• Tell a Story
• You Are the Expert - Make Recommendations
• Try to Keep Slides and Charts Simple
• Encourage Conversation