Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I've tried and failed many times to come up with some kind of small SaaS app that I can just keep running in the background. All of my ideas seem to be just too niche, like solving some tiny problem that no-one would care to pay for. Mainly a lot of different integrations and synchronization between various services. And then I tried my hand at blogs with advertising, or Amazon affiliate sites that went nowhere. And monetized Youtube channels.

Do you have any advice about coming up with ideas for real products, that solve real problems for paying customers? I guess another issue is that the market is getting more and more saturated with mobile and web apps now. It seems like you have to be very lucky to stumble on the gold ideas: where it's very easy to build, and people really want to pay for it.

Or maybe I'm aiming too low, trying to find some minimal weekend app or website. What would be an average amount of effort that you need to put into your projects before you start seeing a return? Or do you crank out a whole bunch of app ideas until you find one that sticks?

EDIT: Really big coincidence, I just read the "How I Built a Side Project" post: https://stayintech.com/info/sideproject

That's almost exactly what's happened to me, 3 or 4 times now. I come up with an interesting idea, build it, post it on Reddit and Hacker News. It gets maybe 20,000 views, then traffic slows down, then I just end up paying a lot of money to keep it running. And it's usually something basic that's very tough to monetize.



I don't have any direct answers, but the more ideas you execute on, the better chance something 'sticks', and the best thing you can do is make it as simple and as painless as possible to launch prototypes. Each time you do a project, take time to abstract the code into reusable components, templates, etc. and integrate it into a generator infrastructure with nodes that represent different 'types' of apps. Then when another idea comes to you, go to your generator node that is most similar, and spin something up with most of your codebase there, front end and back, incl. a stylish landing page. This is a serious time investment to get going, but if you're serious about launching an online business that 'sticks', you're most likely going to need to try out a lot of things.

As far as getting good ideas, I think they almost always have to come to you organically because you have a fascination or interest in the area, or you experience an obstacle somewhere, or you have a need for it personally. PG (or maybe it was Sam) spoke about getting into the right mindset for idea generation in the Stanford startup course, and the gist of it was 'go travel'... i.e. by default be in a 15-20 hr/ wk (max) work setup so you have time to explore things as they pique your interest.


I've been meaning to spin this into a full blog post, but here is the list of criteria that I use to evaluate new ideas:

https://news.ycombinator.com/item?id=9883697

Naturally, then you need to do all the traditional market validation steps, but this at least filters your ideas down to ones that maybe stand a chance.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: