Note · 2026-09-21

Why one-word random tables waste GM prep time (and what to roll instead)

Most GM random tables give a single noun ("Rumour", "Weather", "Treasure") and leave the GM to invent the rest on the spot. A table that puts a full sentence with a built-in hook on every line saves the exact minute of thinking a random roll was supposed to save.

A one-word random table result does not save prep time — it defers the work. Roll a d20 "Rumour" table and get back "3. A missing person," and you still have to invent who is missing, who is asking, and why anyone should care, live at the table, under time pressure. That is the same authoring work the table was supposed to do for you, just moved to a worse moment. In our experience designing GM tables, they save the most time when every entry already contains the specific noun, the specific stakes, and one loose thread the GM can pull on immediately.

What a hook-per-line table actually looks like

Compare a generic result to a full-sentence one for the same category:

Generic: "A rumour about a missing person."
Hook-built-in: "Widow Estren's youngest son came home last spring, but he answers only to a name nobody taught him."

The second version names a person, states a concrete strange fact, and implies at least two next questions (what happened to him, who or what came back instead) without the GM inventing anything. The same principle applies outside social-hook tables. A weather table that says "12. Rain" gives nothing to play; one that says "Steady following wind pushes the party along and carries their voices ahead to anyone listening on the road" hands the GM both a mechanical fact (faster travel) and a complication (an eavesdropper) in one roll.

Three rules that make the difference

1. Name something specific. A person, a place, or an object with a name reads as real; "a merchant" or "an item" reads as a placeholder waiting to be filled in. 2. Attach one consequence or question, not zero. "The well speaks" is a curiosity. "The well speaks, though it only ever repeats what you say to it first" is a scene, because it implies an experiment the players will want to try. 3. Resolve to a full sentence, not a fragment. A fragment ("Strange lights") requires grammar-building before it requires idea-building. A sentence is already usable as read-aloud text if the GM wants it to be.

Why this matters more than table count

We'd argue a 200-entry table of one-word results produces less usable play than a 20-entry table where every line is a finished hook, because the bottleneck was never randomness — it was having something specific to say when a table gets rolled. In this view, table size alone is a vanity metric; hook density per entry is the number that actually saves prep time.

Our <a href="/gm-codex/">GM Codex</a> is built on this rule: 40 system-neutral d20/d12 tables (rumours, weather, NPC secrets, faction goals, site dressing, and more) where every single entry is a complete sentence with a name and a hook already inside it, plus 10 printable prep worksheets. Try the design for free in the <a href="/tools/gm-codex-roller/">GM Codex Roller</a> — 10 of the 40 tables, no install, no signup.