I think you started at one extreme of not planning at all, then moved to the next extreme of over planning. I think businesses have to start with a solid general plan, but things always change, of course. Find a balance.
I usually focus on one site or project at a time. I take each project one step a time. I do plan what I want to do, but I know that things can change as they go along, so I don't put too much time into thinking. However, if I went into starting a business or making a website without a solid game plan I wouldn't know where to begin. I also like to think tentatively about what my goals are for a project, both short term and long time.
There may have been times where I had small experimental side projects in addition to the large ones, but I didn't necessarily expect those to be successful.
For VectroProxy and Primus Proxy, the initial plan has always been to have a functional site that works before worrying about profit. It's good to start with a solid foundation. Next, I think about getting people to the site. Then, after I verify that people actually like it I start to think about monetizing.
The rest comes natural. For instance, as soon as I start making the site I automatically do the on-site SEO as I'm building it. You'll find that even with very light planning, your previous experiences will be a factor since you've worked on sites before and know what works and what doesn't.
Lastly, cut out what doesn't work and build up on the things that work. If one section of your site isn't very popular and doesn't earn much money but still requires work, scrap it and focus more on the part of the site that is popular. Time is money.
Also, you're pretty young if I remember, so you still have leeway to go through this phase