[tomjen3]

The thing he glosses over is that most businesses don't need or even want a perfect logo - good enough is perfectly fine especially as it can be had for a tenth of the price.

And that is properly a bigger change than these spec work sites.

[stylejam]

Good enough in web design is a website that is kind of clear and kind of converts. If it's ok for the customer it's ok for me, but considering that I hardly believe it's ok for the customer here we have a big problem. And there's no such thing as a "perfect logo" :)

[ippisl]

But conversion rates are easily measurable. in theory it's possible to offer a results based service: improved conversion rate for a fee or profit sharing.usually businesses go for this sort of arrangement because it's low risk investment.

I bet there are some services like this , but i'm not sure it's a service fit for a small company: it just seems like a lot of work and takes a lot of time and traffic to do it right.

And it's mostly not about graphic design , but interaction design and content.

[stylejam]

I don't really understand your answer - conversion should be point 1 for every design, call it graphic, call it web, call it interaction, but you suggest it's some kind of "pro" service. In my opinion it's not, a design that doesn't convert doesn't have any business value. For what reasons someone should pay for a website if that website doesn't have any mean or any return? You understand that this is exactly what I'm talking about? A customer buys something that has no use and no impact, of course he wants to spend less, and he is damn right. And that's also why I say that a designer must be a designer, if you have no clue about human interaction you're simply not a designer - can you imagine an industrial designer crafting some crap without any idea about how the user will interact with it? No, it's totally laughable, that's why it doesn't happen.

[tomjen3]

Unless the logo is extremely bad it won't effect conversation, and it's more important to be able to change parts of the rest of the graphics cheaply for ab testing purposes than devine the result initially (a great design can improve conversation say 15% better than the result from one of those sites, but that can be done with a few an tests).

[stylejam]

I guess you mean conversion and not conversation. A/B testing is part of every sane design process, there's no "divination", only iteration. Knowing the process of design would definitely help here.