I want to point out 3 in particular that will really help you gather qualitative research. A/B and multivariate testing is great to show the quantitative difference in conversion between a test control and treatment(s) but it doesn't answer the question WHY one outperforms the other. In some cases you may be able to make an educated guess (big button converts better than smaller button - it must stand out more), but in other cases you can't.
Enter usability testing - with real people, who can tell you in their own words what they like or dislike about your site, how attractive are your offers, how does your site measure up to user expectations and more. Often retailers avoid testing because they think it's time consuming or expensive - but it doesn't have to be. Check out the following resources and you'll have no more excuses not to user test!
Five Second Test
Five Second Test is "a simple online usability test that helps you identify the most prominent elements of your user interfaces." I suggest you upload a screenshot of your home page, category page, product page or email creative and choose the "click test." The click test asks testers to click on what stands out in the image and then describe what they clicked. (The other option is to ask testers what they remember about an image, but that might be more effective for offline/print campaigns when it's important a viewer can recall your brand messaging or offer).
There's also entertainment value in taking part in the testing process. Just click on random test and play! For me it was a dose of reality that my short term memory often can only remember one pathetic thing after a 5 second gaze of an image.
When a new visitor arrives at your site, your site must answer 3 key questions in the visitor's head:
1. Where am I?
2. What can I do here?
3. Why should I stay?
Bryan mentioned a few user testing survey sites like Feedback Army, Amazon Mechanical Turk and 3rdPartyFeedback.com. My favorite is Easy Usability as you can create a survey and also qualify the testers who provide you feedback. For example, if you sell mobility products and disability aids, you want to test with people who would typically use your site rather than some punk kids off the street (no offense to punk kids). Similarly, if you sell streetwear and band memorabilia you want testers who are interested in that.
The feedback you receive may also be helpful in developing personas. Check out this sample test result. Your first test is free and you should receive a report back within 24 hours.
For the deepest qualitative testing, you want capture a site user's verbal comments and on-screen actions (and see how your site displays on various machines, not just your web designer's 45 inch iMac). For $29, UserTesting.com will recruit a pre-screened tester and deliver a video and written summary of the test which is ready in about as long as it takes to bake a potato (in the oven, not microwave). You can ask for testers within an age range, income level, gender or level of Internet savvy. Considering you only need 4 or 5 user tests to get good insight, this is a super-inexpensive way to get involved.