This is a fantastic article. We've had a decent amount of success so far with our startup following this same mantra. Release early with a core set of features,

This is a fantastic article. We've had a decent amount of success so far with our startup following this same mantra. Release early with a core set of features, and then follow up with rapid releases (preferably one per feature). This has the following advantages: 1. Unnecessary features are the #1 source of waste in software. If you launch with a very limited feature set, it's much easier to avoid feature bloat. 2. Having customers immediately helps to fund future development, gives you a large group of real-world test cases, and a huge source of information about what the market wants in your product. 3. Rapid releases let you create constant buzz about your product. If you have six months between releases, it's much more difficult to create constant conversation about what you're up to. Your users will be happier, also, because they continue to feel they're getting more for their money, and they get excited about what comes next. (Also, if you implement their suggestions, they can feel more like they're participating in the product development, which builds a good customer relationship.) 4. Continual releases are a very powerful motivator. We've tried it both ways, and having a release every week (or every other week) for smaller feature sets is much more motivating than spacing them out over a couple months. (Note: this naturally assumes you've got a SaaS model like us. However, I think you could pull off something similar with desktop software, within reason.)