Straight to the point: a worldbuilding sheet does not get better as it gets longer. Keep only the fields you actually go back and look up while drafting, and write down which chapter each entry first appeared in. A forty-field sheet never gets filled past the first few entries. A five-field sheet is still alive at chapter 30.
This is about where to put a world you have already invented, so that it is still usable in week fourteen of a serial. Inventing it is a different problem.
Why most sheets go quiet around chapter 3
You build the sheet to write things down. You open it to find things. Those two jobs pull in opposite directions, and a sheet organised for filing is miserable to search.
Abandoned sheets tend to share three traits.
The field names are categories. You made a heading called “Magic — Costs and Limits.” Three months later the sentence in your head is “how much does a second-tier binding cost again?” Category names are what you reach for while filing. Search terms are what you reach for while hunting. They rarely match.
The same fact lives in two places. Worse than not finding it. You find both copies and have no way to tell which one you updated last.
Draft and sheet live in different apps. The manuscript is in Scrivener, the sheet is a Notion page. If checking means switching windows, you stop checking. You think “pretty sure that was chapter 7” and keep typing. Continuity errors are born in that half-second.
The seven fields worth keeping
| Field | What goes in it | Why it earns the space |
|---|---|---|
| Search term | The words you will type into a search box three months from now | The title is the search interface |
| One-line definition | One sentence, written to confirm at a glance | When you are hunting, you only need to check one thing |
| First appearance | The chapter it first showed up in | This is how you know what the reader currently knows |
| Hard rule | The line this element can never cross | The first thing to collapse in the back half |
| Reader knows | What is on the page already, and what is still yours alone | Only serials need this field |
| Still open? | Whether it is a thread you planted and have not paid off | Nothing else in the sheet tracks that debt |
| Linked entries | Pointers to other entries | Keeps you from copy-pasting content |
Seven looks like a lot. Start with the first three if it does. Just keep First appearance in. It costs almost nothing to maintain and saves more time than anything else here.
What to leave out
- A timeline of events that never reach the page. Great fun at the planning stage. Never opened during drafting.
- Height, star sign, the playlist you made for them. If it will not reach the page, it is ballast.
- The map image. You need the map. Put it inside the sheet and the sheet loads slowly, and a slow sheet is a closed sheet. Keep it somewhere else and link to it.
One test decides: have you looked this up twice while drafting? Once is a coincidence. Twice means you will look it up again.
The thing serials need and finished novels don’t
Royal Road, a Substack serial, any weekly release schedule — all of it runs on the assumption that nobody rereads. You know the whole book. Your reader saw one chapter, on Tuesday, on a phone.
So “does the reader know this yet” is often the more urgent question. When you write the reveal in chapter 22, what you are deciding depends on which chapter the reader saw the evidence in. First appearance plus Reader knows: two cells, and you do not have to remember.
Copy this
Search term:
One-line definition:
First appearance:
Hard rule:
Reader knows:
Still open?:
Linked entries:
Seven lines per entry. Copy the block every time something new shows up in the draft.
A consistency pass, once a quarter
You do not have time to reread the whole sheet mid-serial. Three questions:
- Any thread still marked open that you planted more than ten chapters ago?
- Has anything written into a Hard rule field already been broken?
- Is the same fact written into two entries?
All three should be answerable from the sheet alone. If you have to go back into the draft to answer one, the sheet is not doing its job.
Where this stops helping
The sheet makes a world findable. It will not make it interesting. A dense setting and a good story are separate achievements, and plenty of writers have one without the other.
Skip it for short fiction. Under a certain length you just remember everything, and the upkeep costs more than it returns. It starts paying off somewhere around chapter 20.
Slima keeps notes, draft and every saved version inside one file tree, so a lookup does not cost you a window switch. Try it