5 Kinds of Bad Statistics, and How They Can Kill Your Internet Marketing Campaign
Ian Lurie May 6 2008
Beware bad statistics. They can kill your internet marketing campaign with bad assumptions, incorrect assessments, and by sending you far off track.
By ‘bad’, I mean:
- Statistics that are collected from a small population. Alexa, for example, collects a lot of their data based on Alexa Toolbar users. It’s very possible those users behave differently than the rest of the internet population. So being top 10,000 on Alexa may not mean a heck of a lot.
- Statistics that are just a rough model of the real data. Google’s toolbar pagerank (the pagerank number you see in the Google Toolbar) is based on the true pagerank. And, by Google’s own admission, it’s totally inaccurate. Don’t obsess over it.
- Insufficient samples. “9 out of 10 visitors to our site made a purchase. We had 10 visitors”. ‘Nuff said.
- Subjective data. “90% of our visitors really like our site.” Really? How do you know? Did you watch them through their monitors?
- Statistics collected in focus groups. Focus groups are worthless, in my opinion, for gaging visitor response to a site design or an interface.
Avoid bad statistics.
Even no data is better than bad data – then you know you’re flying by the seat of your pants. With bad statistics, you’re flying with no pants.
Ian Lurie is CEO and founder of Portent Inc. He's recorded training for Lynda.com, writes regularly for the Portent Blog and has been published on AllThingsD, Forbes.com and TechCrunch. Ian speaks at conferences around the world, including SearchLove, MozCon, SIC and ad:Tech. Follow him on Twitter at portentint. He also just published a book about strategy for services businesses: One Trick Ponies Get Shot, available on Kindle. Read More