4 ms·
On the surface the statements seems to make sense and the arguments are well put across, but unless someone actually A/B tests it and proves that it makes a dif
by SimonPStevens 15y ago
On the surface the statements seems to make sense and the arguments are well put across, but unless someone actually A/B tests it and proves that it makes a difference to readership, or conversions (or whatever your goal is) then it's currently just unsubstantiated theory.
- patio11 15y ago5 points of karma says null result, but I'll have at least one anecdotal data point for you in a few days.
- SimonPStevens 15y agoAwesome. I'd be interested to see results. Email me if you get any data your happy to share (details in my profile).
- ekanes 15y agoThanks. Please post results to HN, we'd love to see 'em!
- rlander 15y agoOne little known tip that I've been using for years to reduce bounce rates is to increase the text size by a few pixels. I've successfully reduced the bounce rates on a few sites with that hack alone. On others it made little difference. I guess the older the audience, the more pronounced the effect.
- pbhjpbhj 15y ago>I guess the older the audience, the more pronounced the effect. // Was this a tested hypothesis or is it literally a guess?
- rlander 15y agoLiterally a guess. I've worked with very few sites that target older consumers, so my data points are limited, but in one of these the bounce rate dropped by more than 20% (home decor industry). Unlike more recent experimentations, I did not conduct a formal AB/multivariate test, but the font size had been the only internal change at the time. Also, I have no idea whether these lower bounce rates translated into higher revenue. Also, all of the sites mentioned above were female-oriented.