[stevenj]

If your goal is to build a company, I think you should focus on the following:

1. Find a problem that you (or someone you know very well) have.

2. Look at the existing solutions to the problem and evaluate not only if you can build a better solution, but also how quickly people will switch to something better if they came across it today (i.e. How fed up are people with current offerings? How easy will it be for them to switch?).

3. Build and market something for a very small group of customers or users. Focus on making that group extremely happy with your product or service. (It could be as small as one customer or user at first.)

[yalogin]

What if the idea you have is for a real need/issue you face but its trivial for an established player to build in as a feature?

[snarf21]

How many "you"s are there out there? Is this a space BigCo would bother with? Would the other "you"s trust your more/less than BigCo? Is your issue really that important or does it just annoy you?

In general, the only way to find out is to talk to everyone you can that you think is "you". Ask them if they care. Ask what they'd pay for a solution.

[scribu]

Better than asking them what they would pay, ask them to actually pay a token amount, with the promise that they’ll get it back if the product never comes to market or doesn’t meet their needs.

[DigitalJack]

Email them and hope they build it. Solves your problem and saves you the effort.

[gitgud]

But then what do I do with all the time and money I saved by not building it? ;)

[dna_polymerase]

If the issue exsits within the software of such a big player, don't bother investing time (e.g something built for Google Docs, or Stripe Analytics...). If it's in the wild and a BigCo could do something, try it.

[srikz]

How aligned is it with the BigCo’s biz model? There are several examples of a product from a small startup being far more popular than BigCo’s simply because it’s too small a pie for them and therefore the execution is lacking.

I used to be worried earlier about ‘what if google does it’ for every idea of mine. I don’t think anyone needs to worry unless their idea is a feature enhancement for ads or search (taking google as an example).