|Registration Pages and Visitor Loss |
A Jakob Nielsen Usability Study
Next to having one-hit wonders leave after seeing half of the home page, I've often thought that the most important page to design is a sign-up or registration page. I know that, when I'm the visitor, a lot of sites lose me right there -- and if I get that far on the site, they certainly had me interested to a high degree up to that point.
The new issue of PC Magazine has a very practical usability article co-authored by Jakob Nielsen [zdnet.com].
The print version (but not this online article) has screen shots to illustrate the points he makes, which help a lot. In addition, he zeroes in on my long-time suspicion:
|"...common website drop-off zones, such as the home page and the registration area, where the danger of losing visitors is the greatest." |
Designing a form page often gives me willies -- different browsers vary so much in their implementation of forms that I almost always end up creating two versions, if not more. On top of that, clients often want to inventory the visitor's entire house before they will allow a registration.
I hate the idea of losing a prospect with a confusing or overwhelming page, especially after they have come so far as wanting to sign-up.
I keep feeling like I do things the hard way by creating different pages by trial and error. Does anyone know of a resource that documents browser differences with forms elements, such as size of input areas, extra pixels added around various boxes, etc.?
I would also appreciate any observations or discoveries about how to make this all important page work out well.
First, I stand guilty as charged. I never really put that much thought into signup pages (this site - perfect example). I didn't put a great deal of time or effort into the page - rather bland actually.
Registration or order pages where pii is asked for is another story. Order pages, I go to great lengths on. I feel if I'm asking them to id themselves in a bit of detail, they deserve a good explanation of how the data is going to be used.
Site design itself is also a very difficult question I am still wrestling with after all these years. When is too much time? When is not enough? What pages require the extra time? With users demanding fast loading pages, yet wanting nice looking ones, the choices are endless. I've pretty much copped out with the "kiss" method for all these years and focused on promotion instead. However, there is a quote in that article that really made me think twice:
"When people leave bad Web sites, they rarely come back to see if you got it right in Version 2."
That is so true. Just earlier today, I was at the local news stations' [ktiv.com] website and was quite impressed at how it looked today. I remember three years ago it was a mess; bad links, bad design, you name it - they'd broke every rule in the design book - it looked really poor. There you go - 3 years since I'd been back! In that time, they've made major improvements and even offer RA's of their newscasts - groovy. All from a site about 10 miles away. Big lesson there. First, impressions still count. That starts with a signup page.
I had the pleasure in the past year of re-designing a site whose main purpose was to get seminar sign-ups for anyone of a 20 seminars. The original design had pulled in a grand total of 3 signups over one month, so the site owners were going to be happy with almost anything.
The client drove in traffic with half page ads in the Boston Globe, so all I had to do was make the thing easy to use -- a bit of a challenge since the target demographic was retirement age and not very web-savvy, in most cases.
Well, we filled the seminars, and over half the sign-ups came from the web. But I spent about 2 full weeks on 12 pages or so, and most of that was on the sign-up page.
My observation is that forms are the "techiest" looking thing a web page can throw at a newbie. Also, anxiety runs high about maing "mistakes", loss of privacy, and a host of other issues that require constant "comfort information" to be close at hand.
Getting the input areas well arranged on the page makes a big difference. Setting defaults and text box sizes sanely helps a lot as well. And it's this last area where browser differences drive me crazy.
It seems to me that a registration form might be one of the few good places to use an exit pop-up to cover some of the essentials while reducing the complexity of the main page. The same might go for mouseovers.