Typically A/B tests are run until they reach a statistically significant result. The article doesn't include the sample size, but with a sample size as small as 500 users for each test an increase from 1.6% to 3.2% would be a change larger than 1 standard deviation, and thus can be assumed to be more than just random noise.
I'm a little late to this party, but most of the conversation here has centered our not cleaning your plate, but that is hardly the crux of the problem. Most of the waste happens through the supply chain, from farmers not bring "ugly" produce to the distributors, to food spoilage during transportation, to prepared foods that don't get purchased at the store.
My friend wrote a great book about this last year:
Since reading this book and Barbara Kingsolvers "Animal Vegetable Miracle" my family gets our food almost exclusively from the farmers market or directly from a farm via a farmshare.
A feature I'd love to see is should I refinance or not.
I've currently got a convoluted system of spreadsheets which I have to update every time I want to recheck against current rates. I'd basically like it to tell me in which scenarios it would make sense to refinance. I input things like amount of principle still outstanding, my current rate and term, how far I am into that loan, when I plan on selling, what the refi rate/term options are. The outputs I get are total cost (in fees and interest) by the prospective sale date, and total equity by the prospective sale date.
It is not necessarily in a lenders interest to give borrowers this information, since borrowers are trying to optimize based on paying the least amount possible.
we're building out our refi functionality now. would be great to have you as a test user? we're going to have analysis and data not available anywhere else. info at smartasset dot com.
Interesting idea, i'd love to hear from anyone who's tried this and see if they've gathered any data on it's effectiveness (a/b tests, heat maps etc...)
We implemented a "mega box" drop down at the University of Nebraska (http://www.unl.edu/) in 2009. We didn't do a production a/b test but we did do in-person a/b testing using a custom web app that tested user's ability to find things in the navigation in the mega box compared to a traditional flag model. Mega box won.
Also note that this nav and template isn't unique to the homepage like most higher Ed institutions but is used site wide by all units. That was an epic battle story for another time.
thanks for the info, i'm currently working on a redesign from a site that has all of the top 2 levels of nav showing in a side bar and the proposed design has a single level with "mega menu" style dropdowns, I'll bring this solution up as an option we can focus group. Also, I like the ability to pin the menu open, nice touch
+1 to this idea, particularly cleaning up the URLs. You should also setup a "browse" type page for users to browse through the topics. So if I went to http://iorad.com/tutorial/topics/microsoft-word/ I'd get a listing of all tutorials based on microsoft word. That'll help your SEO for searches like "Microsoft Word Tutorials"
Also, while clicking around your site I got quite a few ColdFusion errors with the full debug output. At the very least I'd set restrict debug output to your local IP's within ColdFusion Administrator.
Durham, NC and the Triangle in general are fast becoming a great startup hub. It's definitely a great place to live and the proximity to the universities provides a large talent pool for startup founders to draw from.
