There is a wiki you can browse right now for a TV show that does not exist.
It has entries. Faction pages, a timeline, characters with histories, a vocabulary the pages use like everyone reading already knows it. If you found it cold, you would take it for the companion site to something that aired, the fan-maintained record people go to for arguments about what happened in episode nine.
There was no episode nine. There was no episode one. Tales from Nowhere is the wiki and the ARG for a show that was never produced, and it lives on the Orbital Ranch site as a finished thing, not a leftover from a dead deal. It did something in a specific order, and the order is the whole lesson.
Here is what happened. The evidence came first. The wiki, the ARG, the puzzle-box of in-world pages that assume a canon behind them, all of it got built before there was a story to explain it. Nobody broke a pilot, mapped the arc, and then commissioned a fake wiki to prop it up. The wiki was the act. It stood on its own before any story arrived to account for it, and it kept standing when the story never came. The show falling through left the evidence intact, because the evidence was the parent of the story.
Now pull back, because this is a method, and it runs backwards from how most people work.
The usual order is story first. You dream up the tale, the characters, the arc, and then, if you are diligent, you build the proof to hold it up. The map goes in the back. The field report gets written when a scene needs it. Story is the main event, the world is the stagehand.
I want to flip that completely. You do not need the story first. You need the evidence. The wiki entry that describes a battle in flat, tactical prose. The magazine spread. The field dispatch written in the voice of a correspondent who assumes you already know the war they are covering. The photograph with a caption that leans on context you have not invented yet. Build one of those, and you have made something real, because the reader is now standing inside evidence of a place, and evidence is how we decide that things exist.
Think about how you actually come to believe a place is out there. Through artifacts. A ticket stub, a grainy photo, an argument between two locals about something you have never heard of. You do not need the story of the town to believe the town is there. You need enough proof that your brain fills in the town on its own. That fill-in is the whole game. Build the outline of a plot instead, and you have made a promise you now owe somebody.
Sequence is doing all the work. Evidence first, story second, and the gap between them is where the reader lives.
Which brings up the reflex to explain. When you make a piece of evidence, there is an urge to attach a label, to tell the reader this is fiction, this is for a project. Resist it. A reader who finds a convincing wiki for a war can see the war is invented without your help. The pages do that job better than any note you could staple to them. You walk in, the world behaves like a place, and you convince yourself, which is the only kind of convincing that ever holds. Your caption would just get in the way.
The reason I keep coming back to this order is not aesthetic. It makes the work lighter and the yield higher, and those rarely move together.
Committing to a full arc up front is a heavy bet. You are wagering a lot of hours on a shape you cannot see yet, and if the shape is wrong, the whole stack comes down. Building one artifact is a cheap bet with a strange payoff. A single wiki entry costs you an afternoon and can still be true a decade later. And the low stakes buy you something. You learn what the world is by making pieces of it. You write the field report, and the report tells you what the war was, because the second you have to file it, you have to know who is losing and why the correspondent sounds tired. The artifact asks questions the outline never would.
That is the trade in a sentence. The evidence convinces. The story just explains. The story is the part you add later, and by then the world has already argued half of it into existence.
This is the thread under everything I keep going on about, the root of the fictional-documentary approach. When I build out an Unstable Signals piece or a CONFLICT Magazine spread inside Terra Conflictus, that reportage is a complete artifact from a real place. The magazine is already complete as a magazine, not a trailer for a book I still owe somebody. Once you accept this, the format stops being the boss of you. A feature, a game, a serial, a comic, a text adventure, all become ways to let people into a world that already exists. The world is real. The delivery mechanism is optional.
So here is what I want you to try, and it is small on purpose. Pick a story you have not written, one that lives as an itch. Do not outline it. Build one artifact instead. One wiki entry. One magazine cover with a headline from a battle you have not designed. One field dispatch that assumes context you do not have yet. Make it good enough to fool a stranger for ten seconds.
Then set it down, look at it, and notice how much of the story is suddenly obvious. You did not write the story. You built the proof of it, and the proof started writing the story back.