4 ms·

When I was starting out, I was so excited to do A/B testing. I was going to A/B every word on every page. The truth of the matter is, A/B is worthless when star

by dmazin 16y ago

When I was starting out, I was so excited to do A/B testing. I was going to A/B every word on every page. The truth of the matter is, A/B is worthless when starting out because you need a hundred conversions or so for both A and B to have statistical significance. I'm getting less than 20 visits per day, but with a very good conversion rate, actually. As for the URL - well, I'm not really into making a micro ISV/SEO-blatant site. Facebook sign up - I really should have that. I have personally never clicked a landing page video. I think the only example I know of it actually being a good decision is Dropbox. I want to see a description with images and text (something my page lacks, really).

michaelpinto 16y ago
1. I don't mean A/B testing on the web — I mean in person. It's part of figuring out your pitch — which is really figuring out what the site is. 2. Having a good url is a form of communication, or branding. It took my over a minute to understand what your site was — people start with the name of the site and work their way down. If you hear of a url and don't get what the site is off the bat you've failed before you've started. 3. You need to learn what your personal preferences are don't matter — so while geeks may like pictures and text normal people might not. Geeks loved MS-DOS which is why Windows conquered the world...
- dmazin 16y ago
  1. Good point. 2. I get your point, but are you talking about choosing something like achievergradetracker.com for the primary URL? 3. You have another good point, however I think that personal preference is more important than people admit. If we didn't infuse some personal preference, especially for aesthetics, into web design the whole web would look like fucking infomercials, which probably convert like Jesus. I guess that's kind of irrelevant as a video walkthrough is really something I have no good reason for not trying.

by dmazin 16y ago

michaelpinto 16y ago

1. I don't mean A/B testing on the web — I mean in person. It's part of figuring out your pitch — which is really figuring out what the site is. 2. Having a good url is a form of communication, or branding. It took my over a minute to understand what your site was — people start with the name of the site and work their way down. If you hear of a url and don't get what the site is off the bat you've failed before you've started. 3. You need to learn what your personal preferences are don't matter — so while geeks may like pictures and text normal people might not. Geeks loved MS-DOS which is why Windows conquered the world...

dmazin 16y ago
1. Good point. 2. I get your point, but are you talking about choosing something like achievergradetracker.com for the primary URL? 3. You have another good point, however I think that personal preference is more important than people admit. If we didn't infuse some personal preference, especially for aesthetics, into web design the whole web would look like fucking infomercials, which probably convert like Jesus. I guess that's kind of irrelevant as a video walkthrough is really something I have no good reason for not trying.