As noted here and in our documentation, much of the DEVONthink experience is what you bring to it, how you think about your data. For example, some people are taggers; some like deeply nested group hierarchies; yet others a combination of the two.
Here’s a something to think about: If you can use the Finder, you can use DEVONthink. DEVONthink just has a brain underneath it all.
If you were going to manage your files only in the Finder, how would you do it?
Also, what are your needs, in terms of databases?
Single monolithic databases are possible, but we usually suggest smaller, more focused databases. Even if the separation is as simple as work and personal. You can make as many as you need to, as long as tey make sense and you can find what you need. As a personal example, I have a financial database where I archive my bills. I can tell you what my electricity bill was in November of 2013. It’s a database that I get into monthly, neatly segregated from say, my work database. And considering I am in work databases far more often, there’s no need to have my bills available in an open database all the time.
So I’d start with that core idea of what databases you need / want.
Then you can develop a workflow of how you’re processing and storing incoming data.
PS: Do you know of Øystein Sevåg? Amazing musician, if you’re not familiar.