1、1,Top Ten Survey Best Practices Webinar,Sponsored by SurveyGizmo Ed Halteman and Christian Vanek September 27, 2007,2,Christian Vanek Christian is a founding partner of SurveyGizmo and the lead software engineer. He comes from an 11-year consulting background focusing on marketing and content manage
2、ment tools. Christian is based out of Cambridge, MA. See SurveyG to sign up for a Survey Gizmo account.,Webinar Survey Experts,Dr. Ed Halteman Ed has a masters degree in applied mathematics, and a Ph.D. in statistics, and he has specialized in survey design for over 10 years. Ed currently heads Surv
3、ey Design and Analysis founded in January 2003 and is available for comprehensive survey design services. Contact them for help getting more out of your next survey. For more information go to SurveyDNA.com or call 303-818-3679.,3,The Agenda!,Webinar: Top-Ten Best Practices,Welcome The Process of Bu
4、ilding SurveysTop-Ten Best PracticesQuestions and Answers,4,Overview of Survey Process,Design (1),Creation (7),Collection (1),Analysis/Reporting (1),5,Top-Ten Best Survey Practices,Best Practice #1: Start with your objectives, not questions,6,Top-Ten Best Survey Practices,Best Practice #1: Start wit
5、h your objectives, not questions,Examples Kwik Phone Telephone Company services small to medium-sized businesses. Objective #1: To monitor customer satisfaction so that identify improvement opportunities quickly to maintain our customer base. Objective #2: To decide where to focus our limited improv
6、ement resources. XObjective #3: To improve customer satisfaction. Think Actions/Decisions,7,Top-Ten Best Survey Practices,Best Practice #1: Start with your objectives, not questions,Questions Overall how satisfied are you with Kwik Phone? Very satisfied, Satisfied, Neutral, Dissatisfied, Very dissat
7、isfied, Uncertain,8,Top-Ten Best Survey Practices,Best Practice #2: Introduce Your Survey & How Long it Will Take,9,Top-Ten Best Survey Practices,Best Practice #3: Include “Not Applicable” and “Dont Know” (especially on required questions),10,Top-Ten Best Survey Practices,Best Practice #3: Include “
8、Not Applicable” and “Dont Know”,How often do you get asked a question where you answer, “I dont know” or “Im not sure”? Yet, on surveys many just dont think to include them.The Reason? They are focused on their perspective (I want an answer) and not that of the respondent!What about the argument, “
9、Dont know is a cop out I want to force them to think about it and give me an opinion!”Forcing respondents contaminates your good data.It creates lack of closure which keeps respondents distracted on subsequent questions.Method: For all questions ask yourself, “Will this be applicable to everyone? Ho
10、w would I answer it? Could dont know be a possible answer?,11,Top-Ten Best Survey Practices,Best Practice #4: Use clear and concise questions and options,12,VS.,13,Top-Ten Best Survey Practices,Best Practice #5: Nice-to-knows Dont CountHow often do you hear? “It would be nice to know what are custom
11、ers think about such and such. . . .” or “Id like to know . . .” This usually comes up during question generation but can happen w/ objectives as well. “Id like to know what customers think of our new product (feature) ideas.” The Easy Fix for “Nice-to know”s Ask “What actions/decisions will it driv
12、e?” Example Objective level: What action will it drive? None, our product is already in development! Invoicing is someone elses responsibility a NTK! Question level: “How satisfied are you with Kwik Phone invoicing?”,14,Top-Ten Best Survey Practices,Best Practice #6: Use branching and show/hide to s
13、horten your surveys appearance.,15,Top-Ten Best Survey Practices,Best Practice #7: Test your Survey,16,Top-Ten Best Survey Practices,Best Practice #8: After testing, do a “conceptual analysis”So youve followed 1-7 best practices and you have a great survey ready to go. STOP! Lets go one step further
14、. Imagine youve executed the survey and you got date back. Think of what the data might look like. Actually make up some numbers, take them for a spin. Now what will I do if these were my results? Will it allow me to meet my objective (s)? Do I need all of this information? Or are some just NTK? Jus
15、t thinking about this will stir up ideas about how ready your survey is.,17,Top-Ten Best Survey Practices,Best Practice #9: Use your survey results as an incentive.,18,Top-Ten Best Survey Practices,Best Practice #10: Sample whenever your target audience is greater than 500* *roughlyBenefits of sampl
16、ing Better accuracy Shows respect for respondents time Builds good will with customers Saves time and money Consider National Polling Remember you never get everyone to respond. What action/decision are you trying to make Exceptions Accuracy needs for breakouts Customer touchpoint,19,Summary Top Ten
17、 Best Practices,Start with your objectives, not questions Introduce Your Survey & How Long it Will Take Include “Not Applicable” and “Dont Know” Use clear and concise questions and options Nice-to-knows Dont Count Use branching and show/hide to shorten your surveys appearance. Test your survey. After testing, do a “conceptual analysis” Use your survey results as an incentive. Sample whenever your target audience is greater than 500*,
