Christian Manzella Rotating Header Image

August, 2010:

IA, Metamorphosis, & Design

I was mowing the lawn the other day and, although I spent $50 extra at the time I bought the mower for an electric start mower, i pulled the rip cord to get the mower started.  Why?  Because the electric start requires the battery to be charged.  This requires plugging it into an electrical outlet, preferably overnight.  Too often, my son or I forget to do it, or we simply don’t do it (since it would require at least three extra steps…)

So it got me thinking…what a great example of graceful degradation, or better yet, metamorphosis.  I mean, it’s similar to the concept of hybrid cars leading up to the eventual culmination in electric only cars with battery stations instead of gas stations.  We’re so very stuck in the way we do things; creatures of habit.  Very few of us really want revolutionary idea and even fewer want their world turned upside down by new inventions.  This is why redesign requires so much more than first design.  Inclusion of existing facets, nuggets of the previous user experience are a must to keep visitors from being overwhelmed with a feeling of alienation.

When building a better wheel, all consideration must be made for the existing solution.  Before a new frame and information approach can be defined, a full assessment must be made of the existing site and solution.  This needs to be done for a number of reasons, including salvation of existing content, brand discovery, and learning about your client.  More importantly, your assessment becomes your launching point.  Every item you glean from the existing site and solution becomes the outline from which you need to spin tales of what will and will not work in the new iteration of theNextBigThing.com.  The information you retrieve, before you even start architecting, is of the utmost importance to your redesign process.

The first question to be explored is how sophisticated is the current solution…. If the existing wheel is nothing more than a coconut with two holes bored in it and a bamboo shoot shoved through, then not a lot of work needs to be done outside of taking inventory of the current content.  Essentially, when you’re starting by redesigning a website that was someone’s FrontPage high school project (we all started somewhere), then you’re designing from a clean slate.

Of course, there is the possibility that your client’s nephew’s girlfriend had a keen, natural talent for design, and if so you should take note of what you review, visually.  Information Architecture is not typically about the visual design (masterpiece) which follows, but it is about visual cues.  There’s no question that strong existing visuals establish both a product brand and a use brand for a particular site.  Pay attention and record what’s obvious.  Record what’s distracting.  These could end up being the very items that keep return visitors comfortable upon the release of your newly redesigned usability showcase.

With a more sophisticated previous site, the work really begins.  Taxonomy of existing (and future) content becomes a small, first step; a single-digit percentage of the entire project.  The navigation structure and associated visual cues now need to be vetted.  What proceeds, really, is a series of questions for you to ask and bang your head against a wall as you take notes, ask again and again, torturing yourself with the good and the bad in which the site users are able to swim through these cheesy seas that your client previously called a website.

What makes sense? Has the content been categorized correctly? How many ways can the content be accessed? Who are the outliers… those good ideas that came well after the launch, but have no real home? Does the user know where they are at all times?

Do a complete investigation of the structure of the navigation of the site. Be critical, in order to build a better solution, but again (sense the theme, people), take special note of what is there and is working.  Despite that inexperienced ad agency’s best attempts at creating something a cat would vomit onto the internet, there’s bound to be kernels of good user experience, ripe for reuse. Keep what works.  It’s critical to maintain that user comfort.  Once you’ve exhausted the possibilities within navigation and location cues, the whims, the good, and the horrific, move on to layout.

Most of your time will undoubtedly be spent in layout.  It’s easy to discount the existing layout in favor of all those tricks that you know will work, but this path quickly becomes a precipice.  The layout is what users will become the most confused about when they jump from page to page, looking for a quick bail, or a link to a cart.  When they go to click the details button and it has suddenly become a home page link, your abandon rate will sky rocket!  Look closely at the visual paths.  How well does the site layout take the user across the screen to relevant areas of the site?  Is the eye drawn in the right direction?  A lot of sites have the right idea about layout, but simply don’t pay attention to the details.  Does it just need to be cleaned up?  Is there just a lack of consistency in gutters between content blocks creating a sensation of discontent for the user?  Make sure to identify all of the appropriate spatial use patterns.  Keep anything that you can, provided it won’t make you feel like you can no longer put your name to the project. Someone else’s layout, but with better usability principals applied, can go a long way.

If you’ve done an appropriate assessment of an existing website solution, you should end up with one to two pages of notes per content category of the site, one to two notes per layout of the site and four to five pages of notes regarding the navigation.  And, of course, your catalog of content.  You now have the tools to begin the process of merging those existing thoughts and ideas into what the client has asked for and what you know to be the right answer, based on your experience and training.

Too often, people expect to employ drastic change overnight, which is anathema to what humans prefer (hint: most sheep prefer comfort over innovation). Don’t be afraid to plan for multiple versions of a release. When working with an existing (quasi-)sophisticated design, there may be drastic departures from the original UI involved.  The larger the brand, the greater the traffic and use of the site, the more disconcerting this can be to the public audience.  It may make sense to release multiple versions, paced over a timeline, wherein you intentionally evolve the site from point-A to point-B.  This will involve creating multiple IA sets, or, at the least, multiple visual designs.  However, in the long run, this can prevent unwanted attrition of site users, something particularly important for transactional commerce sites… but let’s face it; no one wants to lose traffic.  Get the marketing group involved, if you can.  It’s a great opportunity to tell a story with the website and to drive repeat traffic.  Tease users with what’s coming and inform them about what’s been released.  Highlight changes and provide clear, well-written help areas.  Whether they are pages or tooltips make your user assistance easy to understand.

Metamorphosis is a process, by any definition.  Redesign is, in fact, a metamorphosis.  The easier you can make that process on your users, the greater chance you have of retaining your existing traffic and driving new users to your site.