A lot of the statements within this article should be prefaced with "If you want to build a large company,...", "If you want to support a team,...", or "If you want to become more than an SME,..."<p>Targeting niches with tailored software is a perfectly rational -- and, dare a say, I good -- strategy in those cases where you want to build a product that will support (only) yourself economically. And you don't have 'Pinky and the Brain'-esque fantasies. I know it is hackneyed, but bingocardcreator.com is a fantastic example of this strategy put into practice.<p>Often, as these niches aren't already served by a professional product, there is obvious space for monetization. There are also good opportunities to develop quick expertise in that niche -- and as a result corner the market. To boot, as the consumer base is smaller, you can often develop long-lasting, personal relationships with your customers more quickly and more easily.<p>Of course, if you want to take the Internet by storm, don't build a repository of crochet patterns, but if you want to see some quick money roll in (or build up some basic business experience), I but don't see why you shouldn't. Advice has to be viewed within context.<p>Blanket statements are antithetical to intellectual nuance.<p>Not everything has to be New York; sometimes the Galapagos Islands are quite enough.
I'd guess you're more likely to fail building a product for 'everyone' than for a niche. Therefore I'm not sure it's productive to worry about the size of niches unless you really are only interested in building a huge company (which seems not to be how most people who actually do build huge companies start out).<p>If you're aiming for a niche you likely know something about it - or at least can learn & validate quickly. If you're aiming for 'everyone' you're likely either a) solving a non-problem that everyone has (e.g walking: Segway) b) solving a problem nobody has (it's too small, or non-existent).
There are products that are very popular in niche (or not so) places, and practically no one heard of them outside.<p>As a game developer I've seen plenty of those. You'll hear about them only from game developers, and other non-game developers almost never heard of them. And these are not graphics/animation/movie related - they things for managing/debugging memory/leaks/profiling/file I/O/etc.<p>I'm not sure how well these middleware providers do, but if they are small teams (sometimes just one person), and they keep good history, and are able to make $100,000 a year - then why not, especially if you can do other things in the mean time.<p>Here is some (probably outdated, and not full page) - <a href="http://www.gamemiddleware.org/" rel="nofollow">http://www.gamemiddleware.org/</a><p>I for one love Deja Insight - but who has heard of it? <a href="http://www.dejatools.com/dejainsight" rel="nofollow">http://www.dejatools.com/dejainsight</a><p>It's awesome product (not free, not open source, and not VTune/prof contentder) - but the nice GUI that it has, with tracking back makes it awesome little product that I've been using for the last 3-4 years at the studio (and only 2-3 more people, the rest of the studio doesn't know much about it - they use other things).
I'd think the general way forward in a situation like that is fairly easy: you've proven you have a team that can develop a great product for a niche market. The next step is to identify a similar market that you can use your technology and experience to break into, and then raise funding to get the resources you need to break into that market. Raising funding at this point is much easier: you have a proven track record and are profitable.<p>I think the article overstates the problem a lot - a Galapagos product is not a problem, it's an opportunity in a good starting position to finding a solution.
Or, you could end up with something unique that is not found anywhere else in the world. And when people visit that "island" it helps people codify their revolutionary ideas.
I need to think about this, and how it applies to <a href="http://teamrelaychat.nl/" rel="nofollow">http://teamrelaychat.nl/</a> Basically the service is only appealing to heavy IRC users. Others probably prefer Hipchat or something similar.
With a ceiling for growth, you have to look in the other direction: can you reduce costs?<p>Do you really need a full time team to work on this?<p>After v1.0 is done, can the rest of the team 'exit'?
And this post citing the obvious gets upvoted to the main page and the app ecosystem for touchless user interfaces i built over 2 years and posted to HN at the same time gets unoticed. Lame.