Could a tabletop campaign feel like a metroidvania without copying the video game parts?

Build the map around meaningful connections, not a huge count of rooms. Give each region a few memorable landmarks and routes, then track what changes when the group solves a problem. A powered lift, a new agreement with locals, or a cleared passage can make old spaces feel different without requiring a brand-new dungeon every time.

For gates, mix physical obstacles with social and story conditions, but try not to make one exact solution the only way forward. If the group finds another plausible route or persuades the people guarding one, let that count. The fun is seeing the map open up, not guessing the one answer you had in mind.

With several players, I’d treat the party as the main exploring unit most of the time. Let them choose which lead to follow, and use splitting up for short, clear scenes rather than letting everyone roam separately for long stretches. If they do divide, set a boundary, like staying within one region or agreeing to meet at a known landmark.

Make important changes visible to everyone. If someone activates a mechanism far away, show the result on the map or give the group a clear signal, so the other players have a reason to care and a useful next choice. Avoid a setup where one character gets all the exploration while everyone else waits.

The in-between areas don’t all need a fresh fight. Give a few of them a small job: a clue to a nearby route, an unusual local problem, a risky shortcut, or a detail that makes the region feel like it has its own character. Keep these light enough that backtracking doesn’t turn into a string of mandatory encounters.

You can also change the state of a route after a major event. Maybe a passage is safer now, a local group has moved in, or a newly opened path makes the old way optional. If nothing has changed and there’s no choice to make, it’s fine to summarize the familiar travel instead of replaying it room by room.