You should do as much as you can before even considering coding.
Imagine you were about to pass this project off to be brought to life - either by a co-founder or to a hired developer. What details or documentation would you provide them with? Create that now.
Decide on the core functions, sketch it out, write the copy/wording, decide where you will pull data from, what data from the user needs to be stored and passed from page to page.
Look at what people are currently using to solve this issue and why it's like that. Research pricing points and customer concerns.
Strip the idea down further and further until you are confident all that is left is pure coding. You could even step through the application in plain English - then it's just really the syntax that's left for the developer to execute.
There are 3 reasons why doing all of the above is a good thing.
1) The project remains undoubtedly your idea. You've done the thinking, the planning and it has your vision running through it.
2)Your effort in your own idea gives others more confidence in it, e.g by showing how far you've come on your own makes it more likely you'll find a co-founder willing to jump in. In the case of hiring a developer, you cut the cost back ten fold by doing this, you show you're easy to work with, decisive and you explain exactly what you want and don't expect them to think for you.
3) If you really wanted to build it, you wouldn't let lack of coding skills stop you. Doing all of the 'paper' development makes you consider if you really are the right person to execute on this idea. We put a very high value on our own time and if we can't convince ourselves to put effort in, then how will we ever convince a co-founder to join or a customer to buy. Remember, we're not always the right person to execute our own ideas. If you are motivated to do all of the above, it gives you a better understanding of how to run the business once the tech is built.
Some of the best hacks out there are merging existing technologies, no point reinventing the wheel - use what's already proven. It's an efficient process.
"I have a great idea. What do I need to do to get my invention on the market?" http://www.bpmlegal.com/pletter.html
But, I feel this would be harsh, even though it's true. So, instead, I'll ask you this: Who is going to pay for the product or service your idea will lead to?