Since I sell methodologies on YouTube, I often see people follow them only on the surface. I used to quietly expect that even if Zettelkasten was hard to follow, everyone would adopt PARA successfully. Here is an episode from the past.
I met one of my kind subscribers, who watches my videos faithfully, for a coffee chat. We showed each other how we used Obsidian and had a meaningful conversation about productivity. (Mostly about marketability.) The subscriber said they had watched my PARA video, but it didn't seem to suit them, so they gave it up fairly quickly. So I wondered whether they had adopted another framework, like Nick Milo's ACE, but that wasn't the case either.
They were building their own method, tailored to themselves, without using any existing framework. That struck me as odd. Productivity may be a field with no right answers, but not using a framework is the same as saying you will build your own framework. It may sound like a leap, but to me it feels similar to saying you want to become Tiago Forte or Sönke Ahrens. I cover the importance of having a framework in another piece. Frameworks: Falsifiability and Natural Selection
It seems we now live in a world overflowing with methodologies and knowledge, yet one where it is hard to understand them deeply and turn them into steady action. I suspect it's because information is so plentiful that its value has dropped and we no longer feel how precious it is. Information is easy to come by, so we use it easily, and when the effects don't show up quickly, we throw it away right away. Then we look for another method, only to discard that one just as quickly. After digging through so many methodologies this way, people start wanting to create their own. Not by revising an existing methodology, but with a new method that simply overturns its foundations. In truth, it's often less a matter of creating something than of using whatever is comfortable for them. I've also seen many people reject existing methods without even bothering to understand them.
A methodology is a kind of religion. Just as it is said that believers can enter heaven, you have to trust a methodology to succeed with it. And that trust rests on understanding the methodology's principles, effects, and purpose.
The reason you need to trust a methodology is that every methodology requires a period of floundering before you become skilled. You have to endure a stretch where you feel no effect and everything is awkward, and when you fall into a section that nobody has told you how to solve, you have to rediscover the effect, principle, and purpose and work out your own know-how to climb out. It's like driving. Just as a driving school can't teach you every possible case, you ultimately have to log more hours behind the wheel, run into various situations yourself, and solve them.
In productivity methodologies, a typical example of such a stretch is PARA's Area vs Resource classification, and in Zettelkasten, it seems to be the process of gathering notes and eventually writing an article. But without trust based on understanding, people reach this stretch, conclude "PARA is hard to sort" or "Zettelkasten doesn't help me write," and give up using it.
If you skip methodologies that someone has researched and, under the claim that "there's no right answer! The optimal method is different for each person!", just use whatever feels comfortable, there probably won't be much growth. I always view methods built from scratch, rather than found by locating the flaws in an existing method and fixing or improving it, with a skeptical eye. A system built for comfort is unlikely to have a purpose or principles, and it probably hasn't been criticized enough to be improved. It's simply a method that's comfortable for that person. In swimming terms, it's like paddling like a dog instead of formally learning the strokes. You'll float and adapt quickly, but it probably isn't the best you can do. Other people are teaching good strokes, so there's no need to insist on the dog paddle. You need to stand on the shoulders of giants.