Session recaps your players actually read
Two weeks is long enough to forget a kingdom. The recap is the cheapest tool a GM has for keeping a campaign alive between sessions — and the most commonly wasted one. Here is what makes recaps go unread, and the formula that fixes it.
Why nobody reads your recap
It isn’t that your players don’t care. It’s that most recaps are written as minutes — a chronological account of everything that happened, in the order it happened, at the length it happened. Minutes are for the GM’s archive. A recap is a different document with a different job: get five people emotionally back into the story in under a minute, on a phone, in line at a grocery store.
The three classic failure modes: too long (eight hundred faithful words nobody finishes), too flat (a list of events with no stakes — “the party fought cultists and found a map”), and too late (sent an hour before the next session, when its only effect is guilt). All three have the same root: writing for the record instead of for the reader.
The 150-word formula
A recap that gets read fits in roughly 150 words and carries exactly five things:
- Where we stood. One sentence of orientation. “Coldhollow, dawn after the equinox rite.”
- The three beats that mattered. Not everything that happened — the events the next session will lean on. If a beat won’t matter again, it doesn’t make the recap.
- One consequence. Something the table caused that is now true about the world. Consequences are what make players feel the campaign remembers them.
- One open question. The hook. End on the thing nobody knows yet — “the collector now knows you have the ledger. He has not decided what to do about it.”
- A spotlight moment with a name on it. One sentence crediting a specific character’s choice. The player it names will read every recap you ever send, and so will everyone hoping to be named next time.
Write it in second person
“The party traveled to Coldhollow” is a report. “You forced Verrin to speak the old name aloud” is a memory. Second person is the single highest-leverage edit you can make to a recap — it turns the reader from an audience into a participant, which is what they were. Reserve third person for what the world did in response: you acted, it answered.
Spoiler discipline
The recap is a player-facing document. The moment it leaks GM knowledge — the ambush they didn’t spot, the lie they believed — players learn to read your recaps as intel rather than story, and the document changes character. Keep two layers: the recap players see, and private session notes where the ambush lives. If your tooling makes that separation easy, you’ll keep both honestly; if it doesn’t, the layers will bleed.
Send it twice
The recap written the morning after the session is the best one you’ll write — it’s still warm. But the send that changes play is the second one: the same recap, re-sent the day before the next session as the “previously on…”. Day-after captures the memory; day-before loads it back. Tables that get the day-before briefing start sessions faster, ask fewer “wait, who is that?” questions, and walk in with plans.
Let the recap accumulate into a chronicle
A single recap is a courtesy. Forty of them, linked to the NPCs and places they mention, are a campaign chronicle — the searchable answer to “when did we meet the collector?” and the artifact a table actually keeps when the campaign ends. Whatever tool you use, link each recap to the entities it touches while you still remember; retrofitting forty sessions of links is a job nobody ever does.
Where Storywright fits
Storywright was built around exactly this loop: the Session Tracker turns your session notes into a recap seed, recaps publish to players as “previously on…” briefings before the next game, GM-only notes stay GM-only by per-item default, and every recap links into the web of NPCs, places, and story arcs as you write it. The formula above works in any tool — Storywright just removes the upkeep. It’s free to start, and your players are free forever.