If I had to write a complete patient care report from scratch with, quite literally, a blank piece of paper, I would probably do something very similar to an H&P note or a SOAP note. All charting methods will have some strengths and their weaknesses. They're all a compromise and that's OK, as long as you recognize the weaknesses the charting method you use.
Currently at work, we do Chart by Exception. That's all fine, well, and good because in theory if there's not a problem with a given system, it's OK. Here's the big issue with this: when your patient literally has nothing wrong with them (all findings are normal), your chart would be nearly completely blank. So... did you actually assess your patient? That's the problem I have with CBE. Therefore I don't really use it.
I'd actually "write" longer notes if I 1) had the time and 2) I could use voice dictation. As an EDRN, I rarely have the time to sit down and chart as in-depth as I'd like to and since we don't have voice dictation software on all workstations for all users, I can't speed up that process from any of the workstations. I'm a reasonably decent typist but with good software and a fast computer, I can dictate "stuff" nearly 3x faster than I can type.
Now that my rant is over for the minute... What I suggest you do is look over the various charting methods, look at the charting method your employer requires you to use, and then take a good look at which method is going to likely be the most efficient for you. Another thing you'll have to learn is how to "store" the info for your charting in your head so you can pretty much dump it into your chart later. At one point, I could literally tell you nearly everything about a patient, except specific labs and exact VS (I'd write those down) and then dump it onto my chart and once all that's done, I'd pretty much forget those details. I got to the point of taking simple notes for timeline purposes and later reassemble it into a chart...
Once you get good with a given charting method and you're using it consistently (and it has to be able to paint a clinical picture of the patient), you'll find that the "pieces" of charting almost fall into place easily. The one thing I'm not a big fan of is "double charting" stuff. In very simple terms, I won't chart on something in my narrative that isn't already charted elsewhere where "it" has a time-stamp associated with it unless doing so is necessary to call specific attention to that item because it's a pertinent positive or negative finding and I need that to highlight why I did or did not do something. I also don't care much about billing per se, but I also don't participate in fraud. So, while I may write an activity using specific phrasing, it's to avoid having my chart kicked back to me for "adjustment" because the billers want something else said to justify a given report. As an example, a former employer wanted us not to use the word "gurney" in reference to our own gurney because we were getting stuff kicked back to us denied because "gurney" meant "gurney van" trip and that's clearly non-medical... so we used "cot" or "stretcher" or something like that instead. Those "gurney" kick-backs got paid anyway, but it took another round of billing and that just wasted time. I will not, however, write something that isn't true just to "justify" something being billed out just because that gets us paid... that's fraud. It can be tempting to do, but don't do it. I'd rather keep my license... If a run can't be billed out because of something I wrote, so be it. I'm not going to amend that chart to make it billable if I can't do so without committing fraud in the process.
Sorry... I digressed into another rant.
Get familiar with a charting method, don't "double chart" if you don't have to, and always tell the truth, even if it gets you into trouble. The "lie" can get you into even more trouble, more than you want.