Keith's answer pretty much says it all. SEO will take a long time to kick in for your site, so the sooner the better. If you are worried about competitors, then it is even more important that you get your site up ASAP -- it will be much harder to take away a top ranking from a competitor that has already established a strong google presence.

The best thing you can do is to start blogging about the general business area related to your product. For instance if you are building a new SAAS zookeeping management system, then you want to start blogging about anything related to zookeepers. The information does not have to directly relate to your product -- the idea is more to build an audience within your target market segment that you will seek monetize later. If you feel like you dont know much about your target market then it is even more imporant to start bulding that experise now -- you customers will respect your company a lot more if they sense that you understand their business and are not just trying to shove some product down their throat. Dharmeshs' OnStartups blog is of course a really good example of how this is done.

However, there is no need to to say very much about your product at this point. I dont believe in "stealth mode" but I also think that keeping a little mystery is unlikely to hurt you as long as you are building an audience within your target market. This is a lot harder to do for consumer-focused apps than business-focused apps. If you are buildng a general purpose or consumer-focused app that does not have an obvious target market then you want to start thinking about what segment will constitute your early adopters. For consumer web apps, social networks or things like that, the early adopters are frequently from within tech crowd but this is not the only way to go so feel free to brainstorm other potential use cases.

I've asked myself this very same question in the past and I found one piece of information that has really changed my point of view. In the past I wanted to make sure the site was 110% ready. I wanted all the links, content, pictures, logos, etc.

This is just paraphrased because I can't remember exactly how it went.

Google didn't wait until they had the whole Internet indexed before launching and they're doing just fine now.

I pretty much got nowhere real fast! I read this ONE blurb and now I'll put stuff out there that may not be quite done but enough for visitors. I do make sure though the information is accurate and factual because if not you'll lose your credibility.

Hope this helps!

Although I agree with the caveats expressed by others about the potential risks (and extra work), I tend to lean heavily toward the "start now!" camp.

It is very tough to develop a following for a software product (especially a desktop application), so the sooner you can begin building a community of interested prospects, beta users, and influential observers, the sooner you can begin building that momentum.

Obviously you need to avoid shooting yourself in the foot by over-promising in your messaging on the site, and I can see the risk of creating and managing the site being a distraction to your core product development efforts.

I also took a quick look at that "Coming Soon" service offered by Denis, and I think that merits serious consideration as a no-friction way to get started.

The downside to launching the site prematurely is that people might develop wrong impressions about your product. Ralph's strategy is a generally a good one; have a website presence but don't market your product on the site until 1.0 readiness is at arm's length.

That said, your specific product might be resistant to those problems, so adjust your strategy accordingly.

First, get the web site launched as soon as possible (it doesn't need much content yet - either design your own page or use a service like http://comingsoonapp.com Most importantly, don't forget to let people leave their email address and include your google analytics code on the page.

Second, start a google adwords campaign. Set a relatively low daily budget (even as low as $1/day) and every few weeks change the text of the ad. Track how many people 'react' to the ad and you'll get really valuable data about what is the best way to attract.

Third, a few months before release, change the text of the coming soon page to something like 'sign up for beta'. You'll get early adopters who really want to see your software. They'll be great both for testing it, and spreading the word.

Best of luck!

Start it as soon as possible, offer a beta sign up form or a discount to people who subscribe to your mailing list. Create a blog and regularly update it (once a week is fine) with screenshots and milestones, as well as other interesting topics that may fall into your overall category.

As for the pages, these should be sufficient to start with:

Homepage, Screenshots, Features, Blog, Contact

Don't forget to get "Placeholder" accounts on all the popular social media sites, you don't have to use them, but its good to get a hold of them and put links to your site in the bio.

Just my thoughts, and no I have not launched any software so take it with a grain of salt.

Start it right now. It can take a long time until your site starts ranking for relevant searches, so get it up and out there right now.

If you think you'll be immediately swamped with traffic, think again.

Don't wait on this,

I agree with Chris that we cannot answer this for your business. I did not have a choice but had to start a website, as in Switzerland I have to proof to government agencies, especially pension, that I am self employed. In IT this can be done with an internet presence and some bills that show my investment.

I did not advertise my product in the beginning, but establish a presence. I started to blog about my field of expertise. Later on I started to narrow down on the problem my product would solve eventually, and only shortly before launch I introduced my idea and then the product.

Absolutely, positively, unquestionably launch your website right now!!!!

On it tell people what you are producing and why your product will be the greatest thing since sliced bread.

Then have a small survey asking them about what they use now to do whatever it is you'll be doing better and at the end of that, ask them if they'd like to be informed when it is ready, and get their name and email address.

This will start to build up a following.

It will also motivate you to work harder to get the product done.

I did this for 3 years before I launched my product and got the names of 800 people who were interested. When I released my alpha, I emailed them all and most downloaded the trial.

None of us can answer the question for you without knowing a lot more. But the process..

Look at it from the perspective of: 1) what are you trying to accomplish by putting up the site now? What are your objectives?

2) what's the downside? Competitor benefit from knowing? Possibility you extend way beyond ship date and lose credibility? Etc...

3) what's the upside? Start to build a buzz. Start to generate prospects. Etc...

If you write up bullets for each of those, you'll have your answer.