Hard to say without more detail, but licensing is often a quicker, cheaper, less risky route. Assuming you can get over the patent hurdles and the technology delivers a real benefit to some segment of the customer base, the key issue is whether your technology can be grafted onto or incorporated into an existing product to deliver that benefit. If it can, then plugging it into the established infrastructure (marketing, sales, production, distribution etc) of an existing company makes a lot of sense.
Licensing is often the first choice in the technology transfer world for all these reasons - generally speaking you'd only want to create a new company if a market doesn't already exist, or if you will derive a significant advantage from developing, adapting or melding together production technologies that don't already exist in the supply base.
If you're a lone inventor you might want to consider filing with the UK's Intellectual Property Office first - they're a bit more user-friendly than the US equivalent. Make an initial filing (which you can do without professional help if you're willing to put in the work) and then negotiate with potential licensees under an NDA (to preserve your options... for example, there are circumstances where you might want to drop the application and re-file).
You'll need to revise and finalise your application within 12 months of filing, which will cost you $4k-$6k if you engage a (UK) patent attorney. At that point you would almost certainly want to transfer the application into the PCT system (and drop the initial UK application - you can pick it up later if you want protection in the UK and dropping it at this stage avoids complications later).
The PCT application will probably cost you another $4k-$6k if you engage professional help. The advantage is that you're then positioned to file in any/all of the 170-odd PCT countries and you've delayed the cost of those national applications by 30 or 31 months from the initial filing.
The game, if you go this route, is to nail a licensing deal - preferably before the PCT application (i.e. +12m from initial filing), but definitely before the national phase (i.e. +30 months from initial filing) which includes a duty for the licensee to meet the patent filing and maintenance costs.
It's not an easy route (but then again, none of the other options are either!)but pull it off and you might enjoy a lucrative royalty stream without having to re-mortgage your house or be beholden to investors
Because I feel this product not only replaces the old but also opens an entirely new line of improvement for years to come, I am leaning towards a startup. I'll have to look into investors and/or "crowdfunding", but my problem is I do not have much experience when it comes to starting a business. What is the best way to go from virtual design to prototype?