At Guestlist we believe in short iteration and customer feedback. The best advice I can give you is to ignore the waterfall method of design that would have you believe you need to "gather requirements, sketch, create wireframe, etc." in order to be successful. There is value to such a method (if only to create a paper trail that you can use to sell your design) when you are working for clients, but when you are a co-founder the absolute best way to go about is to simple do it.
Start with a specific section of your app and create a mockup in Photoshop. Work on it until you are happy and then implement it. Don't do 10 different variations. Don't worry if you don't know how everything is going to fit together. Just get it working so you can move onto a different page.
The first designs we implemented at Guestlist were pretty terrible in retrospect. It had a lot of fatal logic flaws. We had no idea what our initial feature set would be. We knew we needed a public event pages so we started there. However, as we added more and more pages we realized our first couple of pages where shit so went back and iterated over them. Given enough time and iterations things will start taking shape. Just take a look at the very first mockup I did: http://dl.dropbox.com/u/410315/event.jpg It looks nothing like what we currently have. For reference here is the current public page: http://guestlistapp.com/events/7311
Now, it may seem like you could follow the 37Signals route and simply skip the Photoshop step altogether and get to using it faster. _Don't do this._ You cannot be creative with CSS. The Photoshop stage gives you a chance to do crazy things and not think with boxes.
This can also be applied to websites. The current Guestlist website is very weak in my opinion. It was created over a weekend. There is a much improved version in the pipeline. Not a radically different site, but an evolution that iterates over the weak areas.
The key is to never stop iterating. Never lock down the design. Never let somebody tell you you can't do something because it's inconsistent with the rest of the site. Just change the rest of the site too! Design must evolve constantly.
Comments
At Guestlist we believe in short iteration and customer feedback. The best advice I can give you is to ignore the waterfall method of design that would have you believe you need to "gather requirements, sketch, create wireframe, etc." in order to be successful. There is value to such a method (if only to create a paper trail that you can use to sell your design) when you are working for clients, but when you are a co-founder the absolute best way to go about is to simple do it.
Start with a specific section of your app and create a mockup in Photoshop. Work on it until you are happy and then implement it. Don't do 10 different variations. Don't worry if you don't know how everything is going to fit together. Just get it working so you can move onto a different page.
The first designs we implemented at Guestlist were pretty terrible in retrospect. It had a lot of fatal logic flaws. We had no idea what our initial feature set would be. We knew we needed a public event pages so we started there. However, as we added more and more pages we realized our first couple of pages where shit so went back and iterated over them. Given enough time and iterations things will start taking shape. Just take a look at the very first mockup I did: http://dl.dropbox.com/u/410315/event.jpg It looks nothing like what we currently have. For reference here is the current public page: http://guestlistapp.com/events/7311
Now, it may seem like you could follow the 37Signals route and simply skip the Photoshop step altogether and get to using it faster. _Don't do this._ You cannot be creative with CSS. The Photoshop stage gives you a chance to do crazy things and not think with boxes.
This can also be applied to websites. The current Guestlist website is very weak in my opinion. It was created over a weekend. There is a much improved version in the pipeline. Not a radically different site, but an evolution that iterates over the weak areas.
The key is to never stop iterating. Never lock down the design. Never let somebody tell you you can't do something because it's inconsistent with the rest of the site. Just change the rest of the site too! Design must evolve constantly.