A short while ago, I wrote about Twitter’s new polling feature. I was excited when the feature was rolled out to me so I could try it myself.
Now, I’m not the world’s biggest Twitter user. I’m certainly no celebrity. But I’ve been active on Twitter for over five years and at the time of this writing, I have 423 followers. (Want to boost my ego and add to the total? Follow me @kimkohatsu!) My 400+ users are comprised of professional contacts, brands I’ve mentioned or contacted via Twitter, and some friends and family. According to a 2012 study (the latest I could find), the average Twitter user has 208 followers. I’ve got over twice as many — not bad, right?
Well, not if you look at the first Twitter poll I ran. I asked a generic question that I figured most people have an opinion on: what’s your favorite salty snack? I chose popcorn and potato chips as the options, because hey, they’re my top two.
Which salty snack do you prefer:
— Kimberly Kohatsu (@kimkohatsu) November 16, 2015
Like all Twitter polls, the voting is open for 24 hours. And at the end of those 24 hours, not a single person had voted on my poll. According to Twitter, 27 people saw it, as you can see in the screenshot below. But no one bothered answering. Good thing I didn’t ask about anything important! This is also pure conjecture, but I believe the two total engagements were both me, trying to see if anyone had answered my poll yet.
Perhaps I’ll have better luck next time, but if this first experiment is at all indicative, then for me, Twitter polls was a total bust. Contrast this result with the first PickFu poll I ever ran – I wanted to test two logo designs for a neighborhood salon in Virginia. I got 100 responses, and because women make up the bulk of the salon’s business, I was able to target the question so all the responses came from females. All 100 also explained what they liked about the logo, and I could segment them by age, income, ethnicity, and education level. The total time it took to run the poll? Under two hours.
Was this experiment at all scientific? No. Am I a bit biased in favor of PickFu, seeing as how I’m writing all this on the PickFu blog? Okay, sure. But it’s no mistake that when I looked at my organic engagement, I was also encouraged to promote my tweet and pay Twitter for an additional audience. And even if I had, Twitter can still only offer me impressions on my Twitter poll – not guaranteed votes, and certainly not written explanations.
As a marketing professional, I work with brands that spend thousands on Twitter advertising, and it can be a powerful platform for customer engagement. Even with all its demographic targeting, however, I still don’t think polling on Twitter is the best approach – especially if business decisions are being made based on the poll’s results. There’s no privacy or non-disclosure agreements, and you don’t know at the outset how many responses you’ll get. People outside your target demographic (and outside the United States) are welcome to vote, not to mention your competitors. You also can’t poll on images.
Below is my side-by-side comparison. Has yours been any different? Feel free to comment, or hey, you can always tweet at me. Unlike my 423 followers, I’ll actually respond.