My Photo


« SlideShare Kicks Ass, Picks Up $3M From Venrock & Friends (+ me!) | Main | Memo to Google, Yahoo, Microsoft, & AOL: How to Turn 500M email logins into Facebook Platform & a Crapload of Revenue. »

Wednesday, May 14, 2008



Plus, don't forget a killer autoresponder email series to bring people through the experience

dave mcclure

absolutely sean, i think people should be doing lots of landing page testing & customization based on several factors:

* search keywords (SEM or SEO traffic)
* referral source (search engine, social platform, affiliate partner, blogs, widgets, etc)
* user profile data (if available, either from your site or others... see FB/Google Connect / etc)
* stage of customer life / # of visits / previous landing pages seen
* geography / language
* demographics (age, gender, etc)
* behavioral mode (browsing, searching, buying, gaming, social, etc)
* time of day / day of week / day of month / season

it's almost brutally obvious that only one single version of home page / landing pages won't be optimal for each of these, much less in multiple combinations. but invariably, most startups just have one home page, or a very small # of [untested] variations.

at the very least, choose 2 versions of your home page / landing page to always be testing.

at the most, setup multivariate testing options based on variations in copy, graphics, buttons, etc & test for optimal conversion.

Sean Ellis

Hey Dave, that slide also jumped out at me in Josh Kopelman's post. The other concept I really liked was serving different homepages for people that abandoned the acquisition funnel at different stages (and later return to the website). Don't know why I never thought of this when I was at LogMeIn. We were pretty sophisticated on A/B testing many lifecycle emails. I quickly forwarded the idea to former colleagues. I'll be sure to implement it at Xobni.

The comments to this entry are closed.