Deciding what to capture is a skill. It is not innate, and it is not learnt all at once. It develops with practice, adjusting as you come to know your own system and your own needs better.

What we can do from the start is have some clear filters that make that decision easier without paralysing us at each capture.

The cost of saving too much

Saving has a cost that is normally not accounted for: the cost of processing, organising and maintaining what you save. A note that enters your system will eventually need to be reviewed, processed, connected with other notes or discarded. That work has a real cost in time and attention.

If you save everything, the maintenance cost of your system becomes so high that the system itself becomes a burden. The solution is not a bigger system; it is a more selective one.

The resonance filter

The first filter is the most intuitive: does this make me think something new?

It is not about whether the information is objectively important or whether the author is recognised. It is about whether for you, in your context, with your current projects and interests, this information produces any kind of mental activation: surprise, curiosity, contradiction with something you believed, unexpected connection with another idea.

If the information leaves you completely indifferent, it is probably not worth saving even if it seems “important”.

The application filter

The second filter is more practical: can I relate this to something I am doing or planning to do?

This filter avoids the opposite mistake to the previous one: saving only what emotionally excites us but has no practical application. Knowledge that is never applied ends up being only decoration.

If you can say “this helps me for project X” or “this answers a question I have been asking myself about Y”, it is worth capturing. If you cannot make that connection, the resonance filter has to be very high to compensate.

The scarcity filter

The third filter is the strictest: can I easily find this if I need it in the future?

If the information is published in a reliable place and a search engine can find it in thirty seconds, there is no reason to save it in your personal system. Your system is not a duplicate of the internet; it is a repository of things that are either not publicly available (your own ideas, learning from experience) or are hard to find, or require a synthesis that only you can make.

Use this filter especially for data, statistics and verifiable facts. A number you can look up on Wikipedia does not need to live in your system.

What is not worth keeping

With the three filters in mind, there are categories of information that are almost never worth capturing:

  • Information that confirms what you already know. If it only reinforces an existing belief without adding nuance, it has little learning value.
  • Information without your own context. An article you save without knowing exactly why you are saving it. The absence of a reason is a clear signal.
  • Consumption content without lasting value. Daily news, memes, ephemeral entertainment. It has its function, but it does not need to live in your knowledge system.
  • Generic motivational quotes. Unless you have a specific reason to keep a quote, it adds nothing you could not find in two seconds.

The opportunity cost rule

One way to think about selection is in terms of opportunity cost: every element that enters your system takes up cognitive space that something more valuable could occupy. Saving something mediocre prevents that space from being available for something excellent.

The goal is not to have a big system. It is to have a system where every element has earned its place, where you can trust that what is inside deserves to be there.

A small, selective system you approach with confidence always beats an enormous one you avoid because it is too noisy.

In the next block we move on to organisation: what to do with what you have already captured, how to give it structure without creating bureaucracy.