“What do my customers want?” is probably the most basic question for any marketer. “What do people really think?” is at the root of most research department initiatives. New tools and the use of online mechanisms have made getting the answers to these base questions easier than ever. But to get this feedback with accuracy, certain steps must be taken. Similarly, specific actions must be avoided. Here are some of the key items to keep in mind.
- require respondents to answer questions anonymously
- avoid “hot button” terms in poll questions or answer options
- allow participants to view the statistical breakdown of answers by others
- include expert opinion and analysis on the topic being addressed
- ensure that your polls focus on current topics and issues
If people are required to give their name when answering one or more poll questions, they are less likely to participate in the first place. Even providing people the option to give their name can cause them to answer questions inaccurately; they are then much more likely to think about themselves in the third person and what they should answer rather than presenting the most accurate answer for the particular question. Since the results will be presented in the aggregate or based on anonymous demographic breakdowns, you don’t need or want respondents’ names.
You may have heard the term, “garbage in, garbage out” (or GIGO). Don’t solicit garbage input by fanning the flames with emotionally-charged or widely-used words or terms. If, for example, you are conducting a poll on healthcare reform, don’t use the word “Obamacare.” Conservatives are more likely to react negatively, and liberals positively, just based on that word. The likelihood of the verbiage swaying the results is high and will result in politically-charged data as opposed to unbiased research on healthcare opinions.
People answer polls most often because they are interested in how they compare to others. It’s basic human instinct to compare and contrast yourself to others. Consider providing results beyond (for example) the percentage of respondents voting “yes” versus the percentage of respondents voting “no.” Aim to also provide interesting result breakdowns such as by gender, age, income, state or geography, and other relevant demographics. As the poll provider, it is a “value add” to those taking the poll. Users will appreciate the information, as it allows them practical self-reflection and a deeper understanding of their own beliefs.
Again, GIGO. A corollary to this is to provide the groundwork for an informed decision. The best methodology for this can be found in the format of providing expert opinions on each side of the debate, allowing for a careful analysis of both sides of any issue before respondents answer. Many people just don’t have enough information on some topics to provide a true and informed opinion. Do yourself a favor and take out the guesswork for your respondents; provide the opinions and analysis, as succinctly as possible, to ensure you are getting the best (and most accurate) outcomes to your inquiries.
People care about current events, not about stale or outdated issues. Bore people too much and you’ll lose them altogether. Relevancy counts both in terms of response rates and in terms of fidelity of the results. What’s more is that posting polls on topics that are outdated gets little if any media attention – and just the opposite will occur for issues that are considered “hot.” It depends on your goal with the poll results, but today’s media cycle is quick, so conducting polls on timely topics is the key to getting attention from users and the press alike.
A series of a few poll questions can be an exciting process for the respondent, yet participating in a long survey can be tedious. Ask too many questions, and your response rate will drop off noticeably. Even worse, people may start clicking on any answer just to get through the survey. Respondent fatigue, as the phenomenon is known in research circles, kicks in somewhere north of ten questions. Keep it short, because less is more (accurate).
Keeping it concise will count in sentence construction and word use as well. Treat a poll like news column text. Short sentences, succinct writing, and simpler words work best. Get the message across, but don’t repeat yourself.
It’s an old trick in political polling to ask questions in certain ways to pull out desired outcomes and spin the results in your favor. When conducting corporate polling, you’re probably not trying to get anyone elected; you want accuracy. Don’t ask: “Widget XYZ has won the Such-and-Such Award; would you agree it is the best product of its type?” Instead ask: “Do you believe Widget XYZ is the best product of its type?”
Avoid delving too deep into controversial topics or getting “too close for comfort” with personal inquiries. Don’t conduct rapid response polls on tragedies like terrorist attacks or school shootings – while the events might be topical, timely and top-of-mind, these types of tactics can backfire on you miserably, painting your company in an insensitive light. When trying to gather respondents’ psychographic profiles, ask about their feelings toward innocuous subjects, not “loaded” topics like abortion or gun control. It’s too much, too soon.
You have the results of your poll or survey. You also have the psychographic results from all the individual respondents. It is tempting to go back and offer some wonderful deal to those that answered they have, in essence, an interest in your product or service. Don’t do it. It violates the most basic rule of polling and surveying – privacy rights! Indeed, those that answered the poll(s) will feel their privacy rights have been violated, and they would be right. Your poll respondents are a sample of a wider population, not the tip of a sales pipeline.
Polling your customers and prospects has become a near-imperative business tactic. It is a valuable way of gathering data on the market and your future sales targets. Your competitors are doing it, their intelligence is increasing because of it, and you’d be remiss not to do it yourself. Various tools and vendors in the market have made polls and surveys available to many businesses, from large conglomerates to entrepreneurs. But whether you work for a Fortune 500 or for yourself, you must set up your polls properly, follow best practices, and ensure accuracy of the results; otherwise you are doing more harm than good to your mission. Follow the advice outlined in this article and you’ll be in good shape from start to finish.
More expert advice about Management and Leadership
Photo Credits: © alphaspirit - Fotolia.com; Check Man, Cross Man and Jump Man © ioannis kounadeas - Fotolia.com