I understand what you're saying. I'm aware of pretty much everything you said and I try to build startups in that way. I only work on projects that I can start small and grow. For example I had an idea for a more 'human' analytics app, which instead of showing graphs and statics would explain what was happening and ways to improve things with text. So I thought I'd keep it lean and started with just Twitter analytics and no body cared for it. I still love the idea but I don't have the time or resources to build it into the app I want it to be.
Thanks though, I do appreciate the advise.
"So I've been trapped with building less imaginative ideas because they're easier and I know how to do that."
I would argue that this is a very fatal flaw in your thinking that you should correct asap. You're making it impossible for yourself to succeed.
It's exceptionally difficult to start big / expensive / complex businesses without the ability to raise equally large amounts of capital. And raising large amounts of capital for that type of business is nearly impossible without the experience and or connections you'd properly expect it to require.
So let me put this out there: do not try to start such types of businesses, unless e.g. you've got the kind of wind at your back that Elon Musk did after selling PayPal (to then start something like Tesla or SpaceX, which required immense capital and amazing execution, drawn on from huge amounts of experience and plenty of brilliant influences and employees).
Imgur, Reddit, Craigslist, Facebook, Twitter, Google, YouTube, eBay, Microsoft, LinkedIn, and on and on. These were not massive, highly complex concept businesses to get started (some of them become very complex of course). In fact, they were small'ish concepts, narrowly focused, with the capacity to snowball massively.
Zuckerberg didn't need a billion dollars to build the first version of Facebook, and it wasn't complex. Ditto Twitter or YouTube or Snapchat and so on. Facebook started at one college, Zuck knocked over one bowling pin market, and then the next and then the next.
What's big and elaborate about the core of Candy Crush? Draw Something? Absolutely nothing. You'll find most successful businesses start out very simple; in hindsight, they make you think: holy shit, that's so stupidly simple why didn't I think of it.
Even difficult to start businesses typically start from a simple premise or a simple attack approach. Take Tesla for example, Elon didn't jump right into trying to make the model S (or a mass consumer electric car). He built up experience, an amazing team, capital, manufacturing experience, technology, a sales channel, a brand, global hype - by first building the low-production roadster for a small group of rich people (which could afford it, which then helped fund the next level of development).
Simple idea + easy to get started + low cost to build first version/s + the ability to snowball massively = what you should be building (strictly my opinion). It helps if you can derive sales from day one to use customers as your funding, instead of giving up equity, but that's not a requirement depending on what direction you want to go.
Also, very importantly: the idea itself matters a lot less than you may think. Ideas are nearly worthless, a dime a dozen. For every Facebook, there are dozens of people that had the same idea and didn't execute / didn't ever build it.
This is one of my favorite articles on this topic:
http://cdixon.org/2010/08/21/the-bowling-pin-strategy/