Obtaining Best Results from a 99designs Contest
99designs is a company that applies a prize model to sourcing new designs for logos, web sites, and the like: you put up a prize, designers submit their designs, you choose a winner. The company keeps your prize money in escrow and pays the creator of the winning design, keeping 20% or so as a market maker fee. They also establish boundaries for acceptable prize amounts, cultivate the community, manage the web platform through which contests run, and generally keep things running smoothly.
I recently had occasion to run a design contest for a new site layout at 99designs. I am not a designer, this was a side-project, and everyone who might have volunteered time was busy. I had heard good things about 99designs, liked what I saw when I looked over an array of recent publicly visible contests, and decided to take the plunge - with the expected worst case scenario being that I'd spend some money to learn that the buzz was wrong. Besides that, there's always the value of other perspectives: how are you going to inject fresh vision into your own thinking on web layout without meeting new designers and interacting with them in the course of their work? Working with completely random new people is an interesting prospect.
Because 99designs is a global marketplace, you'll have a better chance of meeting interesting and capable folk than if you were hiring local talent. The pay-off in building connections alone would be worth the price of entry for many people. How much would you pay to meet and accurately evaluate the quality of a dozen or more professional designers from outside your present network? That's exactly what happens in a 99design contest.
My contest went well, and I'm pleased with the result: a couple of very viable connections for future freelance design needs, three designs that met the high bar of approval from the peanut gallery I keep around for such tasks, and no grief. The whole thing took about ten days and cost me a little under US$1000 - the price of a silver package for designing a couple of different web page templates.
While 99designs evidently tries to make things as smooth as possible for first time contest owners, there are some points in which best practice differs from documentation: the designers who use the site form a diverse community of people, with its own evolving cultural norms. The most important point I learned along the way is that you should always run blind contests, so that contestants and onlookers cannot see one another's designs - the majority of designers will not enter a non-blind contest, and you cannot switch between blind and non-blind modes once your contest is launched. Unfortunately, this isn't terribly evident from either the site documentation or the community forums on a first pass. Unless you already know, the most likely way for you to find out is to be messaged by a designer asking you to switch your contest to blind mode.
For many of the same reasons, you should probably also make your contest private - though this seems to be less important to the designer community.
A second point should be self-evident to anyone who has worked with designers in the past: the effectiveness of the contest is directly proportional to the quality of your design brief and the quality of your feedback. You have to be as helpful as possible when it comes to outlining your thinking: what you like, what you don't like, and where you'd like to see designers explore the range of the possible for you. For example, my brief looked much like this:
[My site] is a blog with a focus on [specific subject matter], and read by [specific demographics, with some examples]. I like the idea of relevant wall to wall images incorporated into the page backdrop and mid-ground, or making up the header. Here are a few examples:
[Link to example #1]
[Link to example #2]
But they are examples of the (very general) type of approach I'm talking about - really stepping up with a relevant, large image to make the page stand out, without overpowering the actual content. The imagery on both of those sites conveys an import part of the intended message.
A successful entry will be a grand departure from the present site; something very different. The greatest lacks in the present site design (not that it really rises to the level of "design") are: (a) bad use of whitespace, leading to it feeling cramped, (b) no imagery. Good use of sweeping imagery in the layout makes a site look high class, professional, and other good things.
The blog will always be a wall of text: there will never be images in individual posts. The most important thing on [my site] is text, and text is what should be front and center and delivered to the audience. They come to read. The trick is to have that be the case and not have a site that looks boring or cramped or low-rent or "so 2003" or any of the unfortunate attributes it presently has. The readers are well-educated and sophisticated. They are impressed by the writing, and it would be good if they were also impressed by the site design.
The two page designs I'm looking for are:
1) Home page: showcase the full text of several recent posts, as well as the most important resources that I'd presently like to draw to a visitor's attention.
2) Resource page: e.g. a single post, an about page, an essay. A wall of text of some sort or another that must nonetheless appear welcoming.
Since the imagery was so important in this contest, and at the same time I was looking for ideas that I wouldn't have myself, I added a later section to the brief that went into more detail on that topic. I continued to elaborate on my views as the designers tested the waters with various stock imagery and good ideas.
Providing timely and decisive feedback (or at least notice that you are indecisive on a particular point, and an explanation as to why you are indecisive) is very important, as it helps to quickly identify the designers most willing to iterate their designs multiple times. My practice was to look at new designs as soon as they arrived, then wait for a couple of hours to mull over my reactions and make sure I could talk sensibly, rather than just giving an immediate reaction - I've found that designs tend to grow on you, and your first reaction to a radically new and different look is often not the right one. Then I'd post a comment, rank the design, and offer guidance for the next iteration.
The level of interaction you obtain from contestants is key. You will see drive-by submissions: a design is posted and that will be the extent of the interaction with the creator - no response is given to feedback. It isn't worth trying to push for more there; you are looking for people who are willing, proactive, and will engage in the give and take of a design conversation, not just provide a quick working proposal and leave. Lack of interaction is always a bad signal: perhaps a less experienced individual who is quickly customizing one of a few general templates and entering as many contests as possible, perhaps a significant language barrier, perhaps just a lack of interest in the specific nature of your project. Either way, it's probably not someone you want to be dealing with going forward.
What you are looking for is people who do engage, who revise their designs rapidly and proactively, and who can therefore better show you the different visions necessary to reach a good final design. It is always worth asking the creator of a design you like to show you another quite different version - as you like their work, and want a point of comparison. The winning designer in my contest produced eighteen variations on several designs in the course of a week, a good half of them unasked for on my part, with a lot of back and forth in the contest comments. Needless to say he's in my rolodex now as a resource of first resort for future design needs.
Would I use 99designs again? Absolutely. It is a highly cost effective method of expanding your stable of freelancers, and a reliable way to source good designs.