Apple has been quite successfully at using secrecy to their advantage prior to a product launch. They prefer to build suspense around a new product then let loose with a "Big Bang" announcement and (hopefully) a product that surpasses all expectations.
But their recent launch of MobileMe did not go so well using this strategy. Steve Jobs acknowledged that Apple should have launched MobileMe in beta to a select group, then evolve the product prior to broader distribution.
This "iterative, incremental" approach to launching web services, websites, and SaaS applications is pretty much a universally accepted practice today.
After all, software is "soft". It can be easily crafted, modified, and evolved over time. Software-as-a-Service is arguably even "softer" than traditional software with concepts such as declarative configuration, multi-tenancy, and single code bases becoming the norm.
My personal experience working on several eMarketing and portal projects over the years is that projects that start out overly ambitious and strive to launch with a "Big "Bang" with lots of new functionality are prone to either a) outright fail and be abandoned or b) re-correct themselves, but proceed demoralized.
It's no surprise that our most successful clients and projects are those that started small and evolved their solution over time. The pattern is quite clear. That is why we've designed our subscription levels and services around iterative, incremental development.
Many people talk in terms of "SMB" or "Enterprise" software and make immediate assumptions based on price. We don't see it that way. It's perfectly acceptable for a Fortune 500 company to start with a $195 per month microsite and evolve that solution over time to several redundant/load-balanced servers.
Maybe it's the ego of legacy IT and marketing organizations that persist this notion of "Big Bang" projects. "My new eMarketing platform is $30K per month with 3 full-time consultants" sounds much more powerful than "I'm upgrading my $595 per month microsite to manage all our campaigns once I've established a consistent cost-per-opportunity metric."
Sounds odd coming from me because I don't make more money giving this advice, but I would rather have a satisfied customer start with our $195 month microsite edition and see immediate ROI rather than risk a dissatisfied customer pursuing an overly ambitious goal.
You can have massive scale aspirations and scope. Just make sure the software you're using is "soft" enough to support iterative, incremental development and plan for many small releases along the way to achieve larger goals.