Character and Location Breakdown: The Step Between Script and Shot List
Most breakdown tutorials skip straight from "read the script" to "plan the shots." That skip is the problem. Before a shot list means anything, you need a clean, verifiable list of who is in the story and where it happens — and that list needs to survive the whole script, not just the scene you're staring at.
Introduction: The Step Everyone Skips
Search for "how to break down a script" and nearly every result does the same thing: it opens on a scene, marks the action beats, and gets you to a shot list inside a page or two. That's useful — we've written that tutorial too. But it quietly assumes something that isn't true: that you already know, with confidence, every character and every location in the script.
You don't. Not really. Not on page one, and often not on page ninety. A screenplay introduces people in ALL CAPS the first time they appear and never again. It reuses a location under three different sluglines because the writer typed INT. THE DINER in Act One and INT. MARGE'S PLACE in Act Three, meaning the same room. A "DETECTIVE" in one scene becomes "REYES" by name two pages later. None of this is a mistake in the writing — it's just how prose describing a shoot naturally drifts, because the writer is thinking about story, not about a call sheet.
Character and location breakdown is the discipline of resolving that drift before it reaches your shot list. It's a separate pass, with its own failure modes, and it deserves to be treated as its own deliverable — not a side effect of scanning for camera angles.
- How to pull a complete character list out of a script, including the people prose tends to bury
- How to read a slugline for what it actually tells you about a location
- Why the same physical place can — and should — collapse into one location entity even when the script never says so directly
- Why this step has to finish before shot list and storyboard work starts, not after
ShotList.Studio's own pipeline reflects this: Import → Style → Review → Shot List → Characters → Locations → Storyboard → Schedule. Characters and Locations get dedicated stages of their own, not a checkbox inside the shot list screen — because a storyboard can't stage a scene correctly if it doesn't know, with certainty, who's in the room and what room it is.
scene_04.fountain
1 INT. COFFEE SHOP - DAY 2 3 SARAH (30s, tired eyes) pushes through the 4 door. The BARISTA calls out a name; no one 5 answers. 6 7 She spots MARCUS (40s, suited) at a corner 8 table, papers spread in front of him. Behind 9 him, through the window, a black SEDAN idles 10 at the curb. 11 12 SARAH 13 (to the barista) 14 Just a coffee. Black. 15 16 CUT TO: 17 18 EXT. COFFEE SHOP - CONTINUOUS 19 20 The SEDAN pulls away. DETECTIVE REYES watches 21 from across the street, phone to her ear.
One page. Two sluglines. Five candidate people, if you count generously. Let's break it down properly.
Step 1: Extract Every Character
A pass for people, and nothing else
Run this pass before you think about coverage at all. You are not deciding who gets a close-up yet — you are deciding who exists. Screenplay convention gives you exactly one strong signal: a character's name in ALL CAPS the first time they're introduced in an action line. Everything after that is lowercase, which is precisely why characters vanish from a casual read. By page sixty, "Sarah" reads like part of the scenery.
Three Categories, Not One List:
1. Named, Speaking Characters
SARAH, MARCUS, DETECTIVE REYES. These need full character entries: description, arc across scenes, every location they appear in.
Why: They drive casting, wardrobe continuity, and staging in every scene they touch.
2. Named or Functional, Non-Speaking
The BARISTA. No line, no last name, but a specific job in the scene that implies blocking and possibly a background actor booking.
Why: They still occupy space in the frame and need to be accounted for when staging the shot.
3. Background and Ambiguous Mentions
A SEDAN isn't a character — it's a prop or a vehicle. But "no one answers" implies other customers in the coffee shop who were never named. These are judgment calls, not extraction failures.
Why: Deciding what's a character versus set dressing is exactly the kind of call that should be visible and checkable, not silently guessed.
Notice that category 3 is genuinely ambiguous — a reasonable breakdown could go either way on "no one answers." That ambiguity doesn't go away because software is doing the reading instead of a human. It just becomes something you can flag instead of something you silently miss. We go deeper on how to handle exactly this kind of uncertainty in why AI script analysis needs a confidence score — worth reading once you've got the extraction habit down.
Don't resolve names as you go. Extract first, resolve second. If you try to decide "is REYES the same person as DETECTIVE?" while you're still hunting for new characters, you'll do both jobs badly. Step 3 exists for a reason — do it separately.
Step 2: Extract Every Location
Reading a slugline for what it actually says
A slugline — INT. COFFEE SHOP - DAY — packs three separate facts into one line, and breakdown gets sloppy when they're treated as one fact instead of three:
INT. / EXT.
Whether the camera is inside or outside. This affects lighting and crew logistics, not the identity of the place.
COFFEE SHOP
The physical location itself. This is the fact that should become your canonical location entity.
DAY / CONTINUOUS
A shooting condition — time of day, or a temporal relationship to the previous scene. Not a new place.
Look again at our example: INT. COFFEE SHOP - DAY and, ten lines later, EXT. COFFEE SHOP - CONTINUOUS. Read literally as strings, that's two different sluglines. Read as a location breakdown, it's one physical location — the coffee shop — with two shooting setups: an interior and an exterior, covered back to back. A location list that treats these as two unrelated entries will double-book scouting time and confuse a location manager who only needs to secure one address.
What Belongs on Your Location List:
- Master location: Coffee Shop — the address, the set, the thing you scout once.
- Setups within it: interior (DAY), exterior (CONTINUOUS) — each one is a slugline, not a new master location.
- Sub-locations, if they appear: a scene later reading
INT. COFFEE SHOP - BACK OFFICE - DAYis still the coffee shop, just a room within it worth tracking separately if it needs its own dressing or access.
This distinction — master location versus setup — is the whole reason location breakdown is its own step and not a by-product of counting sluglines. A script with forty sluglines might describe twelve real places. Knowing which twelve is what makes a shooting schedule buildable later; ShotList.Studio's Schedule stage is still a placeholder today, but the location list it will eventually consume has to be this clean to be useful at all.
Step 3: Resolve Aliases and Canonicalize
Collapsing the same thing under different names
Now do the pass that ties Steps 1 and 2 together: go back through your two raw lists and merge anything that's really the same entity wearing a different name.
Character Aliases:
"DETECTIVE REYES" and, three scenes later, just "REYES" — same person. A script might also introduce someone by role ("THE DETECTIVE") before ever giving a name, or use a nickname in dialogue that never appears in an action line at all. Each of these needs to resolve to one canonical character, not three loosely related entries that all show up separately in your shot list.
Location Aliases:
The same problem, worse, because locations don't even get the courtesy of ALL CAPS on introduction. "THE DINER," "MARGE'S PLACE," and "THE DINER - LATER" can all describe one set across a hundred pages, with nothing in the text explicitly linking them. This is where a location breakdown either becomes trustworthy or falls apart.
Every canonical character and location — no matter how it was merged or renamed — needs to carry its source lines: the exact script citations that justify why it exists and why those mentions were merged. If you can't point back at the line, you can't defend the decision, and neither can a tool acting on your behalf.
This is also exactly where mistakes happen, whether the resolving is done by a first AD with a highlighter or a language model reading the whole script at once. Two different characters can share a surname. Two different rooms can share a description. The fix isn't pretending the merge is always obvious — it's surfacing how confident the resolution is, so a human reviewer knows exactly where to double-check. That's the entire argument in our piece on confidence scores in script analysis, and it applies to this step more than any other.
Step 4: Hand Off Clean Data to Shot List and Storyboard
Why this step has to come before, not after
Here's what a sloppy character and location pass costs you downstream, concretely:
Inconsistent Staging
If "Reyes" and "Detective Reyes" are treated as different people, a storyboard's staging — the per-shot blocking that says who stands where — can place the wrong number of people in frame, or lose a character's continuity of wardrobe and position between shots entirely.
Broken Restages
When you edit a scene and ask for a restage — regenerating the blocking after a change — the system needs a stable identity for every character and location to know what actually changed versus what just got reworded.
Redundant Location Setups
Two "different" coffee shops in your location list means two scouting entries, two sets of reference images, and a shot list that can't group scenes by physical location to save a company move.
None of this is really about software — it's about sequencing. A shot list built on top of an unresolved character and location list inherits every ambiguity in that list, silently. Fix the identities first, and the shot list, the storyboard, and eventually the schedule all get to build on solid ground instead of guessing along with you.
In This Series
Character and location breakdown is the foundation of a broader cluster we're building out on script analysis. Planned next:
- How to build a character continuity sheet across a full feature script
- Reading sluglines for production logistics: company moves, INT/EXT crew calls, and time-of-day grouping
- What a location scouting brief should contain once your breakdown is canonical
- How staging and restage decisions depend on a clean character and location model
Why This Step Pays Off Later
Character and location breakdown isn't glamorous. It doesn't produce a shot, a frame, or anything you can show a director in a pitch meeting. What it produces is trust in everything that comes after it — a shot list that knows exactly who's in the scene, a storyboard that stages the right number of people in the right room, and a schedule, eventually, that can group work by physical location instead of by script order.
Skip it, and you're not saving time. You're just moving the same work — figuring out who Reyes is and where the diner actually is — into the middle of production, where it's far more expensive to get wrong.
See It Run on Your Script
ShotList.Studio extracts characters and locations as dedicated stages in the pipeline — each entry carries the source lines it was built from, so you can check the work instead of trusting it blind.
See the tool