A common question on the Macintosh Software Business email list is about finding a designer to work with on your Mac application or product website. Inevitably, after all the “named” designers, who probably don’t have time to work on your project or who you can’t afford, are mentioned, someone will suggest having the work done on spec at a design contest site. Two of the most popular of these sites are 99designs and crowdSPRING.

The practice of having design work done on spec is controversial. Spec work is done on the speculation that the designer will get paid, if and only if you decide to use the work that they produce. I myself come down on the less than popular pro-spec work side of this argument. Why? Because I think the very best software developers do spec work all the time while most middling and bad developers never do any spec work.

Never heard of a developer working on spec? Every time a developer pours his heart and soul and months or years of effort into building a new product, that is spec work! No customer is required to buy your product at the end of that process. You build products with the speculation that people will want to buy them. The willingness to work on spec can be seen as a vote of self-confidence in your own ideas and abilities. Buying design work on spec is no less ethical than buying a copy of ImageWell, TaskPaper or Voodoo Pad. All of these products were built speculatively.

Another curious point is that these same ethical concerns aren’t raised about using open source software or designs. In choosing to contributing to open source, a designer has no chance of getting paid directly for their efforts. No one objects to this because the designer freely decided to participate, either for the publicity and exposure, for the investment in their career that they hope will pay off later, or for the sheer joy of doing their craft. These are the same reasons designers freely decide to participate in speculative design contests.

If you accept my argument, then you can agree that spec work is not ethically wrong. But is it effective? Can you get good designs for your product UI, logos, icons and website? It can work. I’ve run many successful design contests on 99designs and crowdSPRING, and like most things in life, I’ve learned that you get out of it what you put into it. Here are examples of some of the work that I’ve had done on spec.

The one key lesson is to be very active in your contest. Don’t put together your design brief and then come back when the contest is over to pick the winner. You’ll be disappointed with that approach and you’ll waste your money. Instead, provide detailed and timely feedback to EVERY submission. Yes this is a lot of work. Budget to spend a total of 6-10 hours of your time providing feedback during the span of a typical 5-7 day contest. Don’t run a contest during a time when you are too busy to provide this level of engagement.

Prepare to be disappointed and underwhelmed in the designs at first, but don’t get discouraged and abandon your contest. If you stick with it, 3 good things will happen.

At first you’ll get a lot of designs that are laughably bad and you’ll be tempted to write the designers off these off as bozos. Don’t. If you provide continual and constructive feedback many designers will revise the design dozens of times until eventually you end up with something that’s pretty darn good.

By providing constructive and thoughtful feedback to every submission and revision, a second phenomenon will occur. Most of these sites track what percentage of the submissions have gotten feedback from you. A big complaint from designers is that they submit a design but never get any response from the contest owner and when they don’t win they never know why and never had a chance to make the design a better fit. When designers see a contest with 20, 30, or 50 submissions and 100% feedback, they jump on it. By being responsive you end up with a lot more quality submissions than the other contests.

The third thing that happens is that the best designers on these sites often submit on the last day, or even in the last few hours. They also read all the feedback you’ve given to every other design submission so they have a very good idea of what you want. They are able to get an even a better idea of what you want than you had when you first wrote the design brief. By being active with all those not so good designs, you’ve helped other designers (and yourself) discover what you really want and what’s really important to you. I’ve found that I’m often disappointed in the designs for the entire contest and then blown away by what comes in on the last day.

The point is regularly made that these contest sites are not a substitute for a relationship with a designer and this is very true. A spec contest is a better fit for one-off static designs like logos, icons and an initial website than they are for things that need to be very dynamic and are often changing, like a product’s UIs. With that caveat in mind, running a contest is a great way to find designers that you can build a long-term relationship with.

Most designers you’ve heard of have more work than they can handle (that’s why you’ve heard of them). That means they probably don’t have time for your project and if they do it’ll be expensive. By running a contest on one of these sites you can find designers whose work you like, who are responsive to your feedback, who “get you”, and who clearly have time for your work (or they wouldn’t be doing spec work). That sounds like a great recipe to build a relationship around. Many of the designers I work with today on an ongoing basis came from interactions on 99designs and crowdSPRING.

So the next time you need some graphical design, consider a contest. It’s ethically sound, you’ll meet some great designers, and it can be a lot of fun.

Sean Johnson is a seasoned product designer and developer and the President of Snooty Monkey, LLC, a Mac and iPhone development consultancy and the creators of BubbleTimer. If you’d like to consider Snooty Monkey for development work, please contact Sean at sean at snootymonkey dot com.