Ask HN: How do you come with a startup idea?

[cristiioan]

What strategies can one employ to generate an innovative startup idea that outperforms existing options, similar to how Plausible created their privacy-focused and user-friendly analytics platform, which stands out from competitors like Google Analytics?

Rather than creating a copycat app, I'm interested in discovering an idea that sets itself apart in terms of quality and utility. What techniques or approaches have been effective for you in identifying unique and valuable startup concepts that can compete against established market players?

Additionally, what resources or tools would you recommend for conducting research and assessing the viability of a potential startup idea?

[coreyp_1]

Coming up with ideas is a habit. Once you start doing it, it's impossible to shut off, trust me!

1. Learn to be discontent. You aren't seeing ideas because you are content with the status quo. You need to be in the habit of thinking "this could be better".

2. There is a difference between a hobby idea and a business idea. Hobby ideas solve a problem, but you probably can't get someone to pay for it. A business idea may not be "fun", but people will pay for it. If you're lucky, you will find something that overlaps the two areas.

3. Businesses succeed because they solve a problem that people are willing to pay for the solution. Figure out how to monitize your idea before wasting time building something.

4. You must either already have, or have a plan to develop, expert or insider knowledge and connections in order to establish your business. For example, don't expect to build a tool to sell to banks to use if you don't know anything about the banking industry and don't know anyone in the banking industry.

5. A successful business requires (a) someone to build the thing, (b) someone to sell the thing, (c) someone to manage the thing. You might be all three of those. If you are only one of those things, then you need to partner with others who fulfill the other roles.

Good luck! We need more people building things!

[PaulHoule]

Ideas are a dime a dozen. It’s execution that is valuable. So don’t feel too in a hurry to generate ideas.

Take the Google Analytics competitor. Personally I was bothered that GA got a monopoly in the space so fast and that it took so long for real competition to come along, if it has. The interesting questions revolve around “why did GA have a moat?” and “how does a new entrant overcome that?”

Wherever there is a market leading product there is a narrative that

   * Google docs sucks
   * Microsoft excel sucks
   * Adobe creative cloud sucks
   * Facebook sucks

It is very possible to take a bite out of this kind of firm but it is not easy. Look at at Figma as a detailed case study —- it is the ability to execute that puts an idea like that on wheels.

(My current side project is something that I was thinking about 18 years ago thinking… why doesn’t this exist? But it still doesn’t, except on my server.)

[Krollifi]

Whenever you have an idea put it in a file or notebook vault. Prioritize them by putting the most interesting ideas at the top. Build and update the list of ideas for life and you will never have to start from zero when coming up with ideas.

[surprisetalk]

You can generate ideas with a combinatorial approach:

[1] https://taylor.town/an-algorithm-for-generating-ideas

To validate ideas, I try building them for 1 month to find my "real-world" blindspots.

[scastiel]

I often go to https://www.indiehackers.com/products. Sometimes I feel inspired by new projects. Note that it’s just a way to find ideas, not one to know if this idea will work or not :)

[cristiioan]

Ok, so you find an idea this way. But guess what, it is already taken by most probably another startup/small company.

[mindcrime]

There's no such thing as "taken" like that. That is to say, just because somebody else is working on an idea, that doesn't mean you can't (or shouldn't) work on the same idea, or a closely related idea. If anything, if you're working on an idea and nobody else is, you should be asking some really hard questions of yourself, like "why is nobody else working on this if it's such a good idea. Am I really that much smarter than EVERYBODY ELSE IN THE WHOLE WORLD??" and so on.[1] If somebody else is working on the idea, then that's at least small amount of validation that it might be a good idea. If they have paying customers, so much the better.

There's a quote I read once, and I wish I could find the original source so I could provide a proper citation, that I think was from Bob Parsons that goes "Never be afraid to enter a crowded market. Just be better than everybody else." That might be a little bit hyperbolic, but I think the general attitude is sound. Don't be afraid to get in there and mix it up and compete.

As the old saying goes "Startups don't die by murder, they die by suicide." Most likely, the other startup you're think about is going to die by suicide anyway. They're going to fail to execute, choose a bad market segment, screw up their marketing, screw up sales, choose SVB as their bank, or any of 1000 other things that are going to send them to the Great Accelerator In The Sky. Just don't make those mistakes, outlast them, and the market is yours!

Steve Blank[2][3] also has a lot of salient advice on these topics.

[1]: That said, somebody has to go first. So if you have an idea and you have real conviction that it's a Good Idea, then maybe you just have to go for it, even if nobody else has gone down that path. But that does, IMO, imply the need to aggressively try to invalidate your idea so you can pivot as early as possible (before wasting a lot of resources) if it does prove to be a Bad Idea. This whole topic is a lot of what @sgblank talks about in TFSTTE.

[2]: https://steveblank.com/

[3]: https://books.google.com/books?id=CUZsmgEACAAJ

[andrewfromx]

Read https://www.reddit.com/r/Startup_Ideas/ and post your own to get feedback there.

[cristiioan]

I'm trying to find startup ideas first, not to verify if my startup idea is worth. And I wouldn't verify it in such a way. I have read books already on the subject and this seems like a bad idea.

[mindcrime]

Those are basically the same thing in the end. Since generating "ideas" in the abstract is trivial. The interesting part is exactly figuring out if a given idea is worth anything or not.

All of that said, this topic has been beat. to. death. here on HN. Use the search box and you'll find more discussion of generating startup ideas than you can shake a stick at.

[mtmail]

Examples

"Ask HN: Looking for side project ideas" https://news.ycombinator.com/item?id=23290536

"Ask HN: Idea Sunday" https://news.ycombinator.com/item?id=7693262

and a couple "what would you pay $X for" threads https://hn.algolia.com/?q=ask+what+would+you+pay

[andrewfromx]

there is also https://www.reddit.com/r/SomebodyMakeThis/

[dougmwne]

Unvalidated ideas are worthless, validated ideas are gold, execution is everything.