Standard academic surveys
Just to get this out of the way – if you’re doing complex surveys for academic purposes Qualtrics is a great choice. Lots of questions types,
Purpose: detailed academic questions
Price: starts free for academic use
Quality: Very good!
Quick “in-flow” Surveys
Sometimes you need feedback from the current visitors to an existing site. This can be intrusive, but if done rarely is a great way to get information. When we started our site redesign two years ago we needed data on the major destinations of our visitors – what were they trying to do on the website. We also were aware that internal users (faculty, staff and students) probably had different goals than our external visitors (prospective students, parents, local community members) and we wanted more information about those differences.
Purpose: quick surveys of your current users
Price: starts free for 100 responses; then monthly charges based on number of responses desired
Design Review Surveys
What about when you have a couple of designs drawn up in photoshop or the tool of your choice, you’re not ready to make a website with them yet, but you’d like to see what your users think? The hot-spot and graphic survey tools can help you with this. We’ve used two different ones.
Our first design survey we used IntuitionHQ which offers hot-spot testing only. You create a series of questions that can be answered by clicking somewhere on a graphic you upload. The place the user clicks is recorded, as is the time it took for the user to respond. Average times for all users are available as well as the results for each individual. If you have two designs you can upload both and have Intuition automatically divide the participants in half so that you can compare two designs more easily. It’s a little cumbersome to ask the user a question about one design, then the same question about another design. Plus it may skew the results: if the designs are similar the time on the second design might benefit from the experience of answering the question for the first design. Best to test each question on only one design. Unfortunately only two paths are possible at this time so it gets tricky if you more designs.
Links to the surveys can be posted as a URL, or sent in an email. I believe you don’t need a huge number of responses for these types of tests, which is good because getting thousands as we did for the first type of survey would be prohibitively expensive.
The second time we asked our users to review some designs we selected a different tool. This time we wanted more than just hot-spot questions, we wanted some qualitative feedback as well. There were quite a few companies offering composite question types, and we selected a fairly new product from the Zurb company. They create quite a few different tools to help you design and evaluate your sites. We selected VerifyApp.com . The interface for building surveys is attractive, but a little difficult to figure out. Lots of flexibility – they have quite a few test types including memory tests, preference tests, mood tests, and the hot-spot test which they call a Click test. Plus you have multi-variate testing which allows you to compare multiple variations against a control. Each user that takes the test is presented with the control and with one of the variations and asked for a preference.
The users LOVED it – several emailed to say they really enjoyed taking it. That’s not a response I’d had before! The results can be viewed while the test is on-going and can be seen for each user (Raw Data) or in composite view.
Purpose: composite surveys including timed hot-spot testing
Price: free demo, then starts at $19/month without multivariate or demographic testing.
Quality: **** Excellent
So lots of tools out there, these are just a few that we have used. The main thing in user testing is to test early and often. And remember you can get valuable data from 3-8 users, so there’s no excuse for not getting out there now and asking a few people about your site.