Because of this, users need to be convinced that you will in fact be supplying future versions at the quality that they want. They will feel, essentially, robbed if they discover that you sell one version and then fail to continue with vigour, or that you aren't able to keep supplying versions at the expected quality because, for example, you acquired the product from another company or you changed your own. Promises will have almost no weight there, and users will look for evidence. Essentially, they will be convinced if they see that your livelihood depends on turning out good versions of the software. You have to be seen with your fingers where your mouth is.
The converse of that, and therefore the good news, is that users will extend you a certain amount of trust, and thus accept big limitations in your immediate offering, if they do believe in your overall intent and ability to deliver software. Both Microsoft and Apple understand this very very well, as it is evident in their product history. Examples of companies that do a dismal job in this respect are Corel and Sun. In fact I can't think of any successful companies that aren't able to exploit this effect. It's almost impossible to put a new product in the market if you don't make use of it it.
What users buy will be in a certain number and size of chunks - very few from the vendor's point of view. Nobody likes buying lots of software to fill a single need, such as working with documents. They prefer instead to buy a complete package to fulfil a need, such as office, or contacts/diary, or communications. That is both for simplicity and because users perceive, correctly, that buying larger chunks is cheaper. If you have an incomplete offering compared to a need you will suffer because of this, even if the actual usage that you are failing to cover is very small. A good example is Apple's office suite that lacks a spreadsheet currently. On the other hand, users are happy, and in fact prefer, to buy separately what they see as isolated needs, such as road routing, or arts packages, or several distinct games.
From a marketing point of view, that means that it's extremely hard to make any small decisions about what products you build. When considering any particular idea for a product you have to assume that you'll have to evolve and compete with that product for as long as you wish to maintain the company's reputation, which from the point of view of the entrepreneur driving the decisions means indefinitely. When looking for a niche to situate your product you have to look not at the gap that may exist in the product landscape but at the space that the user wants to fill. Typically the user will want something an order of magnitude larger than any one gap, and will pick some complete offering, gaps and all.
The amount of money that users spend on your software will generally be a small fraction of the investment they make in it. You'd better learn to live with that - both with the low willingness to spend money and with the high scepticism and level of demand that they will confront you and your product with. You have to imagine at all times when building your software that you have in your hands the trust of your users, and you are expected to care for them. In that respect, if your software doesn't do something that's not so bad; but if it breaks, and destroys their invested time and information, it's as if you've physically hurt the people you were meant to be caring for. I think this is a maturity issue with the software industry that might one day be fixed, but for now you have to acknowledge it.