Do you think 12-15 main domains and maybe 30-40 subdomains spread out would be a good start?
That definitely sounds like a good start. You could even start with
less; I mean, it all depends on how much you can do in a day, and how big of a network you want to have.
But
regardless of what you want to
now or this week or even in the next month, you could buy all the domains now and let them
sit.
It's like when I make Apfelwein (apple wine). Being a lover of good German Apfelwein, I sometimes drink a lot of it (that wine isn't gonna drink itself!) - and then I regret it the next day sometimes. But man it's so good. The thing is, I can't
buy it - you can't just buy a bottle of Apfelwein in the USA (not anywhere that I know of). So I have to brew it myself. To do that, I have to buy the raw materials (apple juice, corn sugar, brewing yeast), have my brewing equipment ready, and (if I want to be drinking more wine in 6 months) I need to plan ahead. It
takes a while to brew. The day I drink my last bottle, I want to have the next batch in
reserve. When it's gone, it's gone.
So what does that have to do with the price of tea in China? Let's say you want to start a small network...and add new sites to it
periodically. (People "drip-feed" every other damn thing these days, why not do it with
domains?) You could do that with brand-spankin' new domains...or with dropped domains you snap up on an ongoing basis. Or you could think a few moves ahead like Bobby Fisher...and always have some domains being
"aged".
For instance: domains you buy and then put a
landing page on. (I am of course talking about domains for which you
cannot acquire the
old content - if they were dropped.) A
good landing page. One that has...oh...maybe a nice graphic (named for your keyword, with alt text, you know the drill), some relevant text to whatever the domain will be used for eventually, an H1 with your KW, etc. and maybe some related outbound links to some authorities in your niche. That's all I need to tell you without showing you actual working examples (and revealing one of my niches - because I'm so shy), but I just checked and I have three such pages ranking on page 1. So these pages
can rank (depending on the competition) just based on how good they are and how old the domains are. I mean, my LPs all look appealing, so they
must be totally legit and helpful sites riiiight? No one has to know they maybe took only 15 minutes work to set up (each).
(If I haven't already mentioned that all your sites should look damn good, I'll mention it now. On-page is
important - in your code and also how it looks to the human eye. Again, a "real-life" example: You don't want to date an ugly-looking girl, do you?
Hell no, you want a foxy hot-lookin' chick in a miniskirt. Same thing here, make your site all look sexy to Google. Google will then make sweet sweet love to your sites.)
But anyway, the point I want to make is to buy any good domains you come across, but there's no need to rush to plug them into your network as "working" sites. You can make sure they at least get found, indexed, and "aged" with your KWs on them, however, by spending a few minutes to put up a good LP. Then - just like a batch of Apfelwein - when you want to it, it's there.
As for google on the other hand, yes I'll work on covering my tracks to the best of my ability.