Hacker News .hnnew | past | comments | ask | show | jobs | submitlogin
How changing a single word increased click through rate by 161% (visualwebsiteoptimizer.com)
13 points by ankneo on Oct 4, 2012 | hide | past | favorite | 11 comments


And yet they're missing the real probable truth here: when I ignore a 'request quote' and click a 'request price' it's because I don't want to waste thirty minutes with an agent AT ANY POINT in the transaction. I want a set price. I want to buy now from you, or from someone else. But I want hard information right now.


Nah, not missing it. Well aware of it actually but that was not reported to us by Veeam, so instead of trying to fudge, we just stick to reporting for now.


Exactly. I bet "Pricing" would have an even higher CTR, since most people don't want to have to request anything, but want a simple list of products with prices.


Note: pricing info is not published. So they changed an accurate link title 'request quote' - which tells people they'll need to fill in details to find out pricing - to an inaccurate one 'request pricing' which wont actually give the user what they want (prices). So the click through rate increased, but did any more people actually fill in the get-a-quote form? I'm guessing not.


I was thinking the same thing. This is just tricking customers into thinking the link contains less red tape and annoyance than it really does.

It's the equivalent of having a ride at a state fair where you sit there and someone throws dog poo at you. You find that "Request poo flung at you" as the title above the door attracted 0.54% of average fair goers. You then find that if you change the title above the door to "Buy cotton candy" that more people peek their head in the door :P

The real measurement here is how many looked in and still submitted to getting dog poo flung at them? Or to translate that back into original, how many actually go through with requesting a quote?


"with 100% statistical confidence"

I'd love to know the sample size that gives you 100% statistical confidence.


VWO rounds off 99.9x% to 100%. Sample size was 3500 visitors approximately to control and variation each.


The point your parent is alluding to is that "statistical confidence" (confidence interval) can never be 100%.


'Marketing team finally gets marketing'

'Atomic weight of cobalt 58.9'


thanks for advertising your stupid A/B test tool and masquerading it as an article


In fairness that describes a lot of blog posts.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: