Blog

the writings of

When Thinking of Redesign

July 10, 2014

Pondering redesign cycles as compared to an iterative design approach and how those approaches scale across an increasing number of projects.

In theory, I'm against the concept of redesign cycles.

I tend to believe that they do more harm than good. It's not that a fresh take on a subject can't be good - it can be. It's more a problem with the overall cycle that redesigns promote: do a bunch of work designing something, don't improve it at all for some extended period of time, then start all over again.

The problem with redesigns

It feels like wasted potential to me. There's that old adage, "perfection is the enemy of progress." I think that plays a role here. If you assume a design won't be perfect the first time around (it won't), then it stands to reason that you would want to push out a good design first, then push it toward perfection.

In other words: iterate.

So that's been my philosophy for some time. Design once, then iterate toward perfection; as opposed to designing once, letting the design stand (and gather moss), then redesigning again later.

Overall I find that approach satisfying (and it tends to work). As an interesting side effect, it enables celebrating small victories - something I also believe in quite strongly.

Over the long haul

However, as of late I've been pondering the long term viability of the iterative approach. All designs have flaws - some of which only the designer ever noticed and some of which become glaring to all.

Some of my longer standing projects are starting to show some of these warts. Of those warts, some seem like they would result in more aggravation to iterate through than it would take to just redesign everything.

I think it's a combination of this phenomena and lack of in-house design skill that has led to the typical "design, then redesign" cycle in the first place.

When quantity beats quality

While I personally fall further on the "iterate" end of the spectrum, it's really about the right approach for any given project. Sometimes, a redesign is what's needed. I just don't think it should be assumed as the right approach.

In other words, I think it's toxic to assume that any design you make will eventually be redesigned. How can you ever expect to make a great design under that sort of assumption? Instead, assume your design can be improved over time as long as the core of it is your best work.

With that being said, there is an issue of quantity. If you're anything like me, you have several projects at various stages of "completeness" all being designed/developed/maintained in parallel. So, if you're flying solo or low on man-power, even your best intentions to iterate ofttimes will fall by the way side for long enough to make it seems as though the design itself has stagnated.

All in all, I think iteration is the Eden of design (or maybe some better metaphor). I think it's where designers should want to be. It provides an avenue to create the very best work imaginable. The challenge, aside from culture change, is scaling. How can iterative design scale across an ever-growing number of projects?

Culture change

I don't have any real answers here - particularly not for a solo designer. It's something I've been struggling with as of late.

However, consider the current way that web projects are often handed off from one set of designers to another. What usually happens during those hand-offs?

I've found it goes something like this:

Client: "Hey, we want a redesign"

New Designers: "Sure! Oh my god, I can't believe that the old designers did ____!"

Client: "I know, totally. They were terrible. Fix all our problems with a redesign please.

What if all designers thought more iteratively? It's conceivable that the above hand-off could change to be more about iterating on the existing design from the previous designers.

Client: "Hey, we need work on our website. ____ and ____ are problems for us."

New Designers: "Oh, we like what you have here and here but we can do it better if we do ____"

Magical Christmas land of unicorns and rainbows? Maybe. But it seems to me like it'd be a world of great, battle-tested, user-focused products that have had time to mature, instead of just being "re-thought" every few years.

blog comments powered by Disqus