How do you test software sales without building the software first?

[ozmentdws]

Hi everyone,

I have a great idea for a database/personnel management software for my particularly industry. This is a strong muse idea because software of this type does not currently exist for my industry (except for excel and access).

My dilemma is that I'd like to test whether there is interest but how do I do that without paying $1000 or more to actually build the software first?

I'd like to follow Tim's model of using websites and getting people to actually place an order. Is there a way to create screen shots of software that doesn't exist? Photoshop?

Thanks,
Danny

[dvdwlsh]

That's precisely how you do it. Mock it up in Photoshop and market it as if it exists. If you get enough interest to justify the costs, move forward.

[Peter Bowen]

Build the interface first and use it to make screen shots.

Possibly look at doing it in a series of stages and developing each stage/additional features depending on the sales of the previous one?

[ozmentdws]

Photoshop sound slike the way to go if I don't want to lay out the money. Do you need to be a programmer Peter to do what you are talking about?

[ergest79]

You have to build an interactive prototype. I'd say Flash is the way to go, though you'd need some programming. I've seen software out there that allows you to build interactive prototypes and export them to HTML. And you can even do some work in PowerpPoint http://www.boxesandarrows.com/view/interactive

And if you want to be real fancy try this http://www.axure.com/default.aspx They have a 30 day trial which could be enough for you to design the whole thing and even put it up online.

[geekbehindthecurtain]

Don't underestimate the power of pencil and paper. Especially early on in the prototype process. Photoshop and Flash are wonderful tools for nibbling out the details, however it's been my experience more than not, that people will get caught up in the details (is this the right shade of blue) and miss the forest for the trees (is it easy and a joy to use).

I've researched and developed various web applications and have found the following approach to be useful.

1) Lay out your ideas of what you want your software to do in a bullet point list.
2) Follow up each bullet point with the problem it solves for the user.
3) Answer the question, "How does this make the user's life easier?" and have that listed as well.
4) Find someone who you envision would be a user of this software, and spend a few hours interviewing them. Go over your list and listen to what they expect. Novice users are great for this, because they'll ask for things advanced users would simply work around.
5) The UI is critical. To the user 95% of the program is the user interface, the other 5% is simply "magic".

If you've move beyond these steps, then great, it's time to bust out Photoshop and start nibbling on the finer details. Mock up some interface shots, put up a three page website marketing it and be sure to include some way of collecting email addresses to announce the launch.

If you find the response to be lackluster, then you haven't wasted much more than the time you've spent a maybe $100 for the hosting, the domain and other odds and ends.

[webgal]

Think of your website as your salesperson.