Find Problems To Solve
So that suggested a solution where I could make a note after the first shampoo and after the second.
And then looking what I had to work with I realized that I could algorithmically determine the issue if I left the cap open on the first go round and if I grabbed the shampoo and the cap was open it was round 2 so just close it.
If the market is small ask what folks with high levels of disposable income might want.
For example, if you had unlimited energy you could easily create liquid fuels out of the air.
For validation: you definitely need to put in the time talking with the people who you imagine most need what you're building.
People have habits, and solutions of some kind already in place for any problem -- you need to give them a solution that better enough that they'll change their habits for it.
Knowing that any problem is in fact a series of problems the other problems get solution speculations on their own, and out of those hundreds or thousands of ideas pop out things that occasionally are both implementable and monetizable.
To give an example of how I think, one day when I was taking a shower I recognized that I'm so busy thinking about things that I sometimes lose track of whether or not I've done one round of shampoo or two.