TELL A FRIEND KEEP IT SHORT! Web surveys and focus groups are an invaluable feedback resource because they enable you to reach many customers at once. They also allow you to collect extensive qualitative feedback, often rich in clues as to the "why" of customer behavior, while avoiding the high cost of transcription. When they were new, long web surveys escaped the challenges of "survey fatigue," but the novelty has worn thin. Most people need a compelling reason to complete any survey, but especially one that is several pages long. Online surveys with miles of scrolling on a page are now just as uninviting as multi-page printed surveys stuffed into a #10 envelope. Sometimes, less is more. In this issue, we spell out five reasons to keep your surveys short, and describe six ways to pare down your surveys without losing value. GOOD REASONS FOR BRIEF SURVEYS 1. Brief surveys are more likely to be completed. Make it easy for people to say "yes" to your survey by keeping it short and to the point. For many purposes, five to ten questions is enough, including space for one or at most two open-ended comments. A one-page survey with up to 20 questions (including demographics) is often considered acceptable, particularly if all items use the same response scale and are presented in a "grid" or table. 2. Brief surveys can give you an effective snapshot of overall satisfaction and general trends - often that is all you need. 3. Brief surveys are less costly to run, analyze, and report. 4. Results can be summarized in compact visual form so everyone in the organization can grasp results easily. 5. You can send out brief surveys more often for improved tracking of changes. An example of a brief and informative survey is the Silicon Valley Business Sentiment survey, sponsored by the Leavey School of Business at Santa Clara University. The Survey Company has hosted and analyzed this monthly survey since February 2002. The SVBSI has just 6 rating questions, with a handful of demographic items for employed respondents. This monthly survey captures forward- and backward-looking sentiment in the Silicon Valley as seen by a 10-15% sample of MBA's in the area. Monthly results are plotted in a simple trend chart (posted at http://www.surveycompany.com/SCUBI/SVOutlook.html), with summary data and demographics available for additional review. SIX WAYS TO CONDENSE YOUR SURVEYS Use any of these methods to keep survey length at a minimum without sacrificing the quality of feedback you obtain. 1. Take advantage of the web's capability to create branches in your survey. Respondents see only pages relevant to them, based on their responses to branching questions. 2. Avoid asking demographic questions for which you already have data in your database. To include demographics in the analysis, link individual responses to pre-loaded data on each respondent. 3. Send all customers or employees an abbreviated version of the survey. For this survey, choose only the outcome items (overall satisfaction, commitment/loyalty questions, and the like). Invite all respondents to complete an in-depth follow-up survey. Randomly select volunteers for the follow-up survey. Examine demographics to see if the volunteers are representative of the total sample. Send the in-depth survey to all or part of the volunteers; adjust the sample if necessary to create a representative group. 4. Break the survey into parts, and administer each part to a subset of your target respondent group. 5. Send longer surveys only to your most profitable customers or clients; all others receive the short form. Notify your top clients when to expect the invitation, and if possible, follow up with them when you receive their responses. 6. If you have archived raw data from previous surveys, use statistical techniques to identify the most representative items in each category. Create a mini-survey based on those items. Validate the survey by administering the new survey and the original and compare demographic representativeness as well as results on questions in common. If the abbreviated version yields results similar to the longer version, you are justified in using the shorter version in the future. WOULD YOU LIKE HELP CONDENSING A CURRENT SURVEY? The Survey Company will provide a free 15-minute consultation for one survey on request. Send a copy of your survey to SmarterSurveys@surveycompany.com, along with your contact information (name, company name, your position, telephone number, and a good time to call). One of our principals will contact you to discuss ways to condense your survey. IN DEFENSE OF LONGER SURVEYS When you need an in-depth understanding of the experiences, needs, and desires of your target group, the methodology of choice is often focus groups (online or face-to-face). But often a comprehensive survey is desirable because of the credibility of sheer numbers and the ability to analyze results from many angles. Under the right circumstances, longer surveys can generate response rates of 45-60% for customer surveys, higher for employee surveys. Rates of this magnitude are not uncommon when the stakes are high, when respondents have reason to be confident their responses will have an impact, when significant incentives are offered, and/or a when strong relationship exists between the respondent and the survey sponsor. Tips for increasing response rates to longer surveys will appear in a future issue of Smarter Surveys™. What topics would you like us to address in future issues or web site content? Please send suggestions to jgainen@surveycompany.com, or call toll-free, 877-666-2486. If you find Smarter Surveys(tm) helpful, please pass it on. New readers can subscribe at http://www.surveycompany.com/subscribe.html. To unsubscribe from this list go to http://www.surveycompany.com/unsubscribe.html. |
