It's something I've heard many times, and something that I agree with.
You want to build a business, not a feature. Unless you can come up with more features, it's difficult to grow the business.
Then of course there is the risk that your feature will come standard in the next version of "popular software product." If your audience is niche, then you are likely safe, but if your feature appeals to all users of the product, it's inevitable that it will eventually be included in a future version of the product. You end up doing the R&D for free.