A bit of a cheesy title, I’ll admit. But just be glad I didn’t try to make Roll-y Poll-y a thing. (It wasn’t from lack of trying.) Regardless, if you’re considering implementing an informal poll of a specific audience, there are a few major considerations and some resulting best practices that you should be aware of. Because acronyms make everything easier, just think D.E.O. (not Dio, the amazing metal band from the 80s.) Design, Experience, Outcome.
Let’s be honest—if people click through to your Sponsored Poll, and it’s a boring, tedious mess, they’re going to bail on you. Bailed-on polls tell no tales. So how do you get around that?
Keep it light and fun
Remember, this is something people are participating in as either a distraction, or probably instead of working. It shouldn’t feel like work (see examples below). You’re literally competing with the entire Internet here. Keep text minimal and concise. Avoid a ton of jargon and industry terms. You may be dealing with people from all expertise levels, so don’t lose them by burying them in geek speak.
It’s a sprint, not a marathon
You need to be considerate of polltakers’ time, so try to keep your questions to 10 or less in an informal poll. If you can provide a progress bar or a “# of #,” people will appreciate it. That way, they can see an end in sight.
Think ease of use
Multiple choice or multi-select questions are best. This not only makes it easier to answer, it also lets you make sure your data is usable. Fill-in-the-blank and open-ended questions just leave things too vague, and “I don’t know/No opinion” will not get you much worth reporting. Also, check your poll on multiple devices. Are the questions and answers easy to read and mobile/tap friendly? Does everything render well? Could you communicate with vibrant images rather than lots of text? Yes? Do that.
The content and relevance of your questions are equally important, if not more, than the design. Don’t make people feel like they wasted their time filling our your poll because you asked questions that provide useless data.
Don’t lead people on
Your questions need to be as neutral as possible. Ask what you want to know without leading people into the direction you want them to, whether that’s with the question or the answer. For example, don’t ask “Don’t you think that marketing automation is the best thing ever?” And don’t provide answers like “of course” or “obviously.” Those allude to the fact that these answers must be the obvious truth, and can skew your results.
Check their koalafications
If data comes from the wrong audience, does it even really exist? Consider asking a qualifying question at the beginning of your poll (see example below). While this isn’t an absolute, it can really help you make sure you’re getting data from the people you really want it from. For example, “Are you currently, or do you plan to be seeking a marketing automation solution this year?” Then continue with questions around what they’re looking for and their experiences. The qualifying question itself could even end up being an interesting data point.
No fluff stuff
As I mentioned earlier, many people are filling this out in their spare time, so respect that time. Don’t waste their time asking questions you’re not even going to use the responses to. The only exception to this could be maybe ending on a humorous tone, as a “Thanks for getting through all that. Which of these is your favorite marsupial?” Leave ‘em with a smile, and they won’t hate that (see example below).
As with any marketing strategy, your end result should be considered at the beginning of the plan. How will you use this data? Are you informing assumptions? Creating shareable content? Simply taking the temperature around a topic?
Don’t incite a data scientist riot
An informal poll comes with a caveat—this isn’t scientifically collected data. The sampling isn’t random, and it’s not representative of an entire demographic. That kind of research can take an exorbitant amount of time and be very expensive. So with that in mind, when reporting your results, be cautious of your verbiage. It’s ok to say “73% of respondents want a marketing automation solution” but you can’t say “73% of B2B marketers want a marketing automation solution.” It’s a fine line, but an important one, especially to people who conduct true, scientifically relevant research.
All forms of bribery accepted
You know what you want out of this scenario, but what’s in it for the polltaker? Consider your audience and what their motivations might be. Would winning a gift card be in their interest (see example below), or would first dibs at the final results of the poll be more in their interest? You’ll need to think about whether your audience is intrinsically or extrinsically motivated and cater to that.
Seems like a good deal of work, right? And we haven’t even discussed promotion of the poll. Luckily for you, MarketingProfs offers a Sponsored Poll product that helps you achieve all of the above, and Content Creation to help you put together your results. Chat with one of the fine folks on our team to learn if we can help you get a feel for the audience you’re trying to reach!
Your fellow pollster poll-star and