3 ms·

It is essential to an extensive A/B testing first. Otherwise how can you possibly know that 11px is better than 10px?

• ryantownsend 16y ago
  I'd go much bigger than 11px... I think it's pretty much a given that the download trial link is way too small. It needs to be clear for people that they can give it a try without having to fork out $500 immediately. I don't think you need A/B tests to prove that - it's pretty much a given that 10px is too small. I'd also agree with @patio11 - the website is outdated, unprofessional and I wouldn't trust it. I'd pay a decent web designer a couple of thousand to get a bespoke design that will be much more suitable than what you have now.
  • patio11 16y ago
    You don't even have to pay a couple of thousand -- $50 for any of the SaaS web templates from themeforest.com or $70 or whatever for one of the product highlight ones from WooThemes would be a quantum leap in the perceived quality of the site. You can always reboot the design later with more professionally placed drop shadows and Helvetica after the software has sold a bit.
• michael_dorfman 16y ago
  Clearly, he's not getting enough conversions now to be able to get a statistically significant result in a reasonable time frame. I'm all for A/B testing, but the problems here are at a much more fundamental level--it is about understanding your market, and matching your customer's expectations. There will be plenty of time for A/B testing, once the sales volume is there to support it.