By building a better product I mean taking what is an existing product area that sucks and making what you believe is up to modern technology standards in terms of UX and deployment model (say web and mobile). I'm not really talking about disrupting this existing area, just providing a vastly improved product that does more or less the same thing in terms of functionality. Sort of what like xero.com is doing for accounting software.
You wouldn't expect to become super rich doing this (xero probably will) but is there at least a good business in this? The markets already been tested and technology is always moving on. Do all startups need to swing for the fence and try and be the next Facebook? I see alot of swinging for the fence which is fine, but maybe this could be contributing to the high startup failure rate?
And then how much better is enough for it to be worthwhile?
The largest hurdle I see to just being better is probably the amount of effort involved in educating the customer that you are in fact better. Also people are naturally adverse to change....
Obviously there is no definitive answer...
If you solve a customer pain, then yes. There are plenty of companies that do this successfully. Dyson comes to mind. They reinvented the vacuum cleaner -- hardly a glamourous endeavor but it's clearly a better mouse trap.
Look at all the deal sites -- Group On, Living Social, etc. All are better mouse traps of all things coupons. Image that, reinventing the boring coupon but it works.
The only want to figure out how much better is worthwhile is to do the research and the math. Figure out what niche you want to go after and do it better than the competition.
I don't know if you need to be 36% better (although I'm intrigued @B Mitch on where you got the number), but it does have to solve a customer pain.
What you describe might be better summed up as, as good a mousetrap, delivered more conveniently to its target audience. And - if only in a limited sense - that is a disruptive strategy.
To be successful, you want to know three things.
1. Is the apparent market a real and active market? (Sometimes sucky products persist because there really isn't much going on, for the incumbent or for you)
2. Is there a significant challenge for the incumbent to match your improved experience? (Work back from the current UX to infer the underlying processes. Three cheers when you can see evidence of complexity!)
3. Do you have ready access to customers or channels? (If the existing players have the market sewn up, your only realistic option may be to partner with an existing player.)
That's a good set of gates for your own prequalification of apparent opportunities. Now you're in classic startup territory, and you'll be weighing standard questions such as whether you will choose to attack the broad market or specific segments and niches, what impact a fast follower would have on you, and so on.
I wouldn't pick an overly saturated market, but yes. If you can differentiate yourself by making a higher quality product then you can certainly be successful.
When Gray Poupon came out they positioned themselves as a higher quality mustard, being sold in a small glass jar with fancy labeling. And they sold it for twice the price of the other mustards. People bought it.
In other cases there can be a niche or variation of a product that has great potential without anyone realizing it. A good example is extra chunky spaghetti sauce. In the 70's you would not find extra chunky spaghetti sauce at the grocery store. It wasn't that spaghetti sauce companies weren't listening to their customers. It was that, when asked, their customers didn't actually know what they wanted. After a better methodology was employed by Howard Moskowitz, he found that 1/3 of Americans prefer extra chunky spaghetti sauce.
By the way, all of these stories and more insights can be found in this TED Talk from Malcolm Gladwell (a great video).