I was saddened, but not surprised by the recent decision that LinkedIn would not be allowing their site to be used as a B2B survey sample source.

Some have theorized that the decision was based in part on whether or not it was a financially viable option. Others have said the restrictions being put on what types of surveys could be used for the source were too limiting for the broader MR community.

I see it as a sign that as a community, market researchers in general have lost touch with what common everyday people are willing to do with their time, as well as not having any clear vision about how we can interact with the public who are using social media.

As a moderate user of LinkedIn, and suspending for a moment that I work in this industry, I can honestly say I didn’t join the site to be hammered with 30-40 minute surveys. Let’s face it, this is where the majority of our work has gone. It’s harder to get people to cooperate, so when we do find willing participants, we have to ask them every single question we need under the sun.

Over the last couple of years I’ve worked on some internal initiatives focused on how to fundamentally reinvent the research experience for consumers. This work has involved quite a bit of testing on ways to present questions in a nicer format graphically. However, the key principals focus on better written surveys because at the end of the day, a horribly written survey made “pretty” by using loads of Flash is still a case of the proverbial lipstick on a pig.

By “better written” we mean surveys that are clearer, more user-friendly and more appealing to respondents. By greatly improving how we write our surveys, we can help cut down wordiness and remove ambiguity. This plays a part in the wider efforts to increase respondent engagement. We believe higher respondent engagement means lower drop-out rates, greater respondent retention or future participation, and richer data for our clients.

Here are a few tips to improve your survey with an eye towards respecting the conversation with a public we desire so much to hear from.

  1. Write to be understood – The language of ANY questionnaire should be simple and quick to understand.
  2. Avoid unnecessary scene-setting – Often, long preliminary scene-setting is the result of not thinking clearly about what we want to say.
  3. Minimize necessary scene-setting – Some IS legitimate when it develops from previous questions.
  4. Question language – Ask questions in simple language and cut the chit-chat. Aim for a maximum of 10 words for a question.
  5. Answer language – Present answers in the sort of language the respondent might use in speaking to us and make the response options self-explanatory.
  6. The meaning of words – The shorter the word, the faster the communication. Respondents will not feel patronized if you use short words.
  7. Spelling and grammar – Just get it right! When you don’t it reflects negatively on the competency and professionalism of our profession.
  8. Translations - Translators should understand that we are not looking for direct translations, but rather ones that capture the essence of what is being asked.

As always, there’s a lot more supporting these eight steps than I can write here, but the key takeaway is the realization that the more words we use potentially increases the chance we have of asking a biased question. Additionally, talking down to the public is a sure-fire way to decrease future participation in market research and the further thinning of the potential sampling pool.

At the very least, using less, more direct words and easier response choices should shave time off of your survey length. Don’t clutter it back up with 10 more additional questions.