He spent two weeks researching a place before travelling: sixty links, three interviews he had read, an old map, figures from an official report. He made the film, and it worked.
A year later he needed the same material for another video about the same region. He opened his folder and found a file of links with no titles, half of them dead, and notes written in an app he no longer used. So he sat down and researched for another whole week, and all of it had been in his hands a year earlier.
This is the most valuable thing creators lose, and the thing they talk about least: their research.
Research is an asset, not an expense
In every video you make, you gather more than you use. The surplus is not waste; it is the first material for a future video. Whoever keeps it writes the second video in three hours, and whoever throws it away starts from zero every time.
The difference is not in the software you use but in the shape of what you keep.
The rule: keep the substance, not the link
A link on its own means nothing after a month. You do not remember why you saved it, or where the important part is, and it may not even open.
So write a five-line card for every reference worth it:
| The line | What goes in it |
|---|---|
| What it says | One sentence in your own words |
| Who says it | The body or person, and what makes them a source |
| When | The year of the material, since old facts may have changed |
| Why it matters to me | The video or question it serves |
| Exactly where | The page, the minute, the chapter name, with the link |
The first and fifth lines matter most. The first lets you understand the card a year later; the fifth saves you reading the whole source again.
How to gather and verify material before this step is in Research Before Writing: How to Gather Your Material and Verify It.
File by question, not by subject
Filing by subject looks organised and does not work: a folder called "travel" becomes another graveyard within a year, exactly as an idea list does in Your Idea Stock: Why It Becomes a Graveyard.
What works is filing by the question the reference answers: "how much does a trip there cost?", "who owns the land in this area?", "why did this craft stop?" In future you search with a question rather than a title, and the question is what will come to mind.
Put two or three words with every card in your own everyday language, because those words are what you will type into the search box a year from now.
Three cupboards, not one
Separate three kinds, because each has a different use:
- References: figures, dates, documents, articles. These are text cards.
- Visual material: old photographs, maps, archive clips, each with its source and rights status written beside it from the first moment.
- People: who you spoke to, their number or email, what they know, and whether they will appear. After two years this is your most valuable cupboard.
The second cupboard is the one that lands people in trouble a year later, because an image gets saved without its source and then used. Whoever recorded the rights status with the image spared themselves everything in Photos and Archive Footage: Where to Source Them Safely.
Pages disappear
Any page on the internet may vanish, a government source changes its links, an article goes behind a wall. So keep a local copy of every reference you build something on: a PDF, or a screenshot showing the date and the headline.
Put those copies in the project folder itself, with clear names, following the same logic as How to Organise Your Footage Files and Never Lose a Shot. A local copy is not only for verification; it is what lets you reuse the reference in a future video with no new research.
And an AI assistant does not replace a library
Many people now ask an assistant instead of going back to their references. That shortens the time and raises the error rate, because the answer arrives with no source and with equal confidence in what is right and what is wrong, as in How do you use AI for research without it handing you a mistake?
The useful use is having it help arrange your cards and summarise what you gathered yourself, rather than being the source.
Fifteen minutes after every video
This is the only habit the article asks for. After publishing any video, sit for fifteen minutes and do three things:
- Turn the references you used into cards, if you have not already.
- Write cards for the material you gathered and did not use. That part matters most.
- Write one line: "what question in this subject is still unanswered?" That line is usually your next video.
Most people at that moment are tired and want to move on. Those fifteen minutes are the difference between a channel whose knowledge accumulates and one that researches from scratch every time.
How the library shows up in the video
A library does not mean putting everything in it on screen. Figures and facts have a limit in a script, with the method in Numbers and facts in a script: how do you include them without weighing the video down?
But the library gives you two things: confidence in every sentence you say, and speed in writing a second and third video on the same subject from different angles.
And after a year
Whoever wrote cards for a full year opens their library and finds five videos with the research already done, the names of ten people they could call tomorrow, and figures they know the origin of.
Whoever did not opens a folder of links that no longer work. The difference between them is fifteen minutes a week.