Online Polls and Voting
December 31, 2009
Since launching ShopSavvy back in 2008 we have participated in one or more online polls or votes each quarter. I groan a little each time, but at the end of the day we need the exposure winning might create. For example, last year we were nominated for a Crunchie by internet voters, but ultimately lost out to imeem (ahem). More recently Tim O’Reilly’s team pitted ShopSavvy against our primary competitor in something called the O’Reilly iPhone Smackdown. About the same time we found out we were also nominated for the CES Mobile App Showdown. Both contests relied solely on internet votes.
Of course we have a small advantage in these sort of internet based polls/votes - we have millions of users. Any time we get into one of these online contests we ask our users to vote. For example, we created a bit.ly link for the O’Reilly vote and sent it to a segment of our users via a push message. We kept the message live for three days sending more than 61,000 voters to the O’Reilly website to vote. Click here to check out the stats for yourself. Ironically about 10% of these folks voted for our competitor because they viewed the request as ‘SPAM’ (which it sort of was, but hey the app is free so we should be able to ask a favor every few months). Earlier today we got a note from O’Reilly explaining that they had taken down the poll (we were winning by about 80% or so). Here is the note:
Thanks for your note and apologies for the confusion. We pulled the smackdown because there was lots of bot traffic skewing the results. It seems that our web team implemented the smackdowns with fairly naive, casual poll software. Votes are handled by simple <a> links with only modest cookie-based attempts to prevent multiple votes from the same machine, all of which means that they’re quite susceptible to skewed results via bots. In this case, it looks like a bot or spider stumbled into the page and kept pounding both vote links. I’ve removed this smackdown from the site and will hold off on additional smackdowns until we find out more and/or get a more robust poll solution in place. We’ll give the smackdown another try then. Again, sorry for the trouble.
I suspect they weren’t used to so much traffic. We reached out to them and shared the bit.ly stats so they could determine that the source of the traffic was most likely from actual voters. Of course it is a little frustrating that we used our ‘goodwill’ to generate votes only to have those votes discarded. I doubt we can ask these users to vote for us a second time for the same contest.
The CES Mobile App Showdown is still underway (8 days left). We are currently in first place with 16,457 votes. We have limited the number of users who have received a push request to vote for us so that we don’t swamp the CES servers. I couldn’t help but feel that it was silly that we had to limit the number of requests so that we didn’t skew the votes too much in our favor (as we did in the O’Reilly vote). As long as these sort of votes/polls persist we will happily participate, but I wouldn’t mind some sort of multilateral agreement whereby we would all agree not to have online votes/polls. How about you?
