A few weeks ago I wrote a post called Research Polling Finally Enters the World of 2.0.

It’s about a cool little widget called Urtak.

I’ve installed one of these on my blog (see side bar just under upcoming appearances).

So far the results have been fascinating.

I’ve received over 1,000 responses.

Some interesting tidbits about the BrandTwist readers:

- 30% of your Moms are on Face Book
- 67% of you have fired a gun
- 90% describe yourselves as happy
- 74% spend more than 2 hours a day goofing off online
- 71% would like to work for Don Draper
- 100% say you like this Urtak widget

I’ve just added about 40 new questions.

A sneak peek:
Keep reading »

Read this post and check out my BrandTwist Urtak poll on the sidebar.

There’s a fundamental flaw with the majority of research polls.

By their very design, they make assumptions on what questions are relevant and/or predictive.

In other words, the poll results are only as good as the questions that have been asked.

But what if the assumptions on what’s important are wrong? Then the data is not so useful.

For example, I could do a survey on airline travel satisfaction and ask lots of questions about seat comfort.

However, maybe I didn’t think to ask about food quality, which might be the key driver for a lot of respondents.

In this case, the relevance of the data is limited by the assumptions and bias of the researcher (me).

My friends at Urtak have set out to change this.
Keep reading »