Chapter 20: B-Roll and Coverage

Overview

You did everything right in Chapter 19. You framed the interview off-axis, lit it with a soft window key, ran two mics for safety, asked open questions, and got twelve minutes of a real person saying true, moving things in complete sentences. You have gold. And when you drop it into an editor and try to cut it — to lose the rambling middle of an answer, to trim the "ums," to put the best sentence first — the picture jumps. The subject's head snaps from one position to another on every cut, their hand teleports, the background lurches. Each edit announces itself like a slammed door. You are now staring at the single most common wall a beginner hits in nonfiction: a perfectly good interview that cannot be edited, because there is nothing to cut to.

An interview with no other footage is a hostage video. It can only be one long, unbroken take, or a string of visible jump cuts. You cannot shorten it, reorder it, or show what the person is describing. Everything they say — "these hands have made ten thousand loaves," "the shop nearly closed in 2019," "watch how the clay rises" — stays trapped as audio over a static talking head, because you have no picture of the hands, the shop, or the clay. The interview is the skeleton. What you are missing is everything that hangs on it. What you are missing is B-roll — and this chapter is about going out and getting it.

Here is the throughline this whole chapter turns on, the third of the book's five and the one you have been building toward since Chapter 1: you shoot for the edit. B-roll is that principle made physical. It is not decoration you sprinkle on at the end; it is the connective tissue that lets an interview become a film. Every cutaway you grab, every detail you insert, every little sequence you cover is a gift handed forward to the editor — who is almost always you, at a keyboard, at night, either blessing or cursing the version of you who held the camera today. This chapter makes sure that future editor blesses you.

In this chapter you will learn to:

• Define B-roll and its counterpart A-roll, and name the four jobs B-roll does that the edit cannot live without.
• Tell a cutaway from an insert from a sequence, and know what each does in the timeline.
• Shoot a sequence — a single action covered wide, medium, and detail — so it cuts into one clean, compressed action.
• Match your B-roll to the story beats (Chapter 17) so the picture proves what the interview says.
• Write a coverage plan — a B-roll shot list built from the beats — so you never leave a location one shot short.
• Think with the coverage mindset: shoot generously, motivate every frame, and capture each shot's own sound.

Learning Paths

B-roll is the great equalizer — it makes a phone documentary look intentional and a lazy cinema-camera one look empty. Everyone needs this chapter, but here is where your attention pays off most.

• 📱 Phone-first: §20.3 (shooting a sequence) and §20.5 (the coverage plan) are built for you. A phone shoots gorgeous B-roll one shot at a time; the discipline, not the device, is the whole game. Your feet and a steady hand are your gear.
• 🎥 Creator: §20.1 and §20.4. B-roll is retention — it is how you cover your own jump cuts and keep a talking-head video moving. Learn to sequence a process and you can make any topic visual.
• 💼 Pro-track: §20.4 (matching B-roll to beats) and §20.5 (the coverage plan) are the professional core — this is how you guarantee a client's story cuts together and never go back for a reshoot you should have planned.
• 🎓 Student: §20.2 (the taxonomy) is your vocabulary; §20.6 (the mindset) is the habit. The Production Checkpoint gathers all of Project 2's B-roll, so read with your documentary subject in mind.

20.1 What B-roll is and why the edit needs it

Let us name the two halves of your footage plainly. The A-roll is your main footage — the primary material the piece is about: the interview, the piece-to-camera, the subject speaking, the core sync-sound action. It carries the words and the spine of the story. The B-roll is everything else you shoot to lay over and around the A-roll: the supplementary shots — cutaways, inserts, little sequences, atmosphere — that show what is being talked about, cover the cuts in the main footage, and give the piece somewhere else to look. The A-roll is what your subject says. The B-roll is everything the viewer sees while they say it.

The names are old and slightly arbitrary, and it is worth knowing where they come from, because the origin explains the job. The term is commonly explained as coming from the film era, when an editor would build a program from two physical reels: the "A" roll held the main footage, and a parallel "B" roll held the secondary footage used for overlays, inserts, and transitions. Whatever the exact history, the working meaning has been stable for decades: A-roll is the primary track your story rides on; B-roll is the secondary footage that covers, illustrates, and enriches it. (Treat the reel-origin story as a useful Tier-2 account — widely repeated, and true in spirit — rather than a fact to build an argument on.)

Do not let the "B" fool you into thinking it is optional or second-rate. B-roll does at least four specific jobs, and an interview-driven edit needs every one of them:

• It covers your cuts. This is the job that saves your edit. When you trim an interview — cut out a stumble, join two good sentences that were minutes apart — the picture jumps (a jump cut, from Chapter 9). Lay a piece of B-roll over that join and the jump disappears: the viewer is looking at the subject's hands, not their snapping head, at the exact instant the audio splices. B-roll is the editor's escape hatch. Without it, every trim is visible; with it, you can cut an interview to ribbons and no one will ever see the seams.
• It shows what is described — show, don't tell (Chapter 17, §17.4) made literal. The subject says "I still hand-set every line of type"; you cut to their hands setting type. The words carry the feeling; the picture carries the proof. This is B-roll's storytelling job, and it is the difference between a viewer being told something and a viewer believing it.
• It compresses and controls time. A cutaway lets you skip the boring middle of a process — cut to a detail, and when you cut back the bread is baked, the pot is glazed, three hours have passed and the viewer never felt the gap. B-roll is how the edit bends time.
• It adds visual interest and controls pace. A face talking for three straight minutes is exhausting no matter how good the light. Cutting away — to the work, the place, the tools, the result — gives the eye somewhere to rest and the piece a rhythm. Variety of image is variety of attention.

Here is the failure and the fix, side by side, as an edit-timeline diagram — the first of several in this chapter, because B-roll only makes sense when you can see what it does to a cut. (We draw the timeline the way an editor reads it: video tracks stacked on top, audio below, time running left to right, cuts marked |. You met this notation in Chapter 1's pipeline and will live in it from Chapter 26 on.)

FIGURE 20.1 — Why the edit needs B-roll: the same trimmed interview, without and with cover

  WITHOUT B-ROLL (the hostage video)
  V1  [ interview: talking head ###|###|############|### ]   <- every | is a visible JUMP CUT
  A1  [ interview sync audio ####### (three answers spliced together) ####### ]
              ^ head jumps    ^ head jumps        ^ head jumps
      The audio is great. The picture betrays every edit. You can only keep ONE unbroken take.

  WITH B-ROLL (the same audio, now cuttable)
  V1  [ head ][ B-roll: hands ####### ][ head ][ B-roll: the shop ##### ][ head ]
  A1  [ interview sync audio ######################################################## ]
  A2  [ room tone / wild sound bed ....................................................... ]
             ^ cut hidden          ^ cut hidden           ^ the audio never stops; the picture roams

   Same twelve minutes of interview, same three answers joined. In the top version the joins are
   naked. In the bottom, B-roll lays over each splice: we hear the subject keep talking (the sync
   audio runs unbroken on A1) while we SEE what they describe. The cut vanishes. The interview,
   untouchable a moment ago, is now free to be shortened, reordered, and shaped into a story.

Read the two timelines as the same raw material handled two ways. On top, the interview is edited but has no cover, so every audio splice forces a picture jump — the edit is a cage. On the bottom, the identical audio runs unbroken on the A1 track while B-roll rides over the video track at every join: the viewer hears the story continue and sees the hands, the shop, the work. Nothing about the audio changed. The B-roll simply gave the picture somewhere else to be at the instant of each cut. That is the whole mechanism, and it is why professionals gather B-roll as reflexively as they breathe.

✂️ In the Edit: every trim in an interview needs something to cut away to. This is the callout that governs the entire chapter, so let it sink in. In the edit, the moment you want to shorten or reorder an interview, you reach for B-roll to hide the join — and if you did not shoot it, it is not there, and you are stuck. There is no button labeled "generate the hands." The overlap between "what I say the interview needs" and "what I actually captured" is decided entirely on the shoot day. So the working question on set is not "did I get a nice shot?" It is "have I got enough cover to cut the interview however the story turns out to need?" Shoot as if the editor will want to hide a cut every eight seconds — because they will.

💡 Why It Works: the ear forgives what the eye cannot. A viewer will accept the picture changing under a continuous voice without a flicker of confusion — we do it in conversation all the time, glancing around a room while someone keeps talking. But the same viewer instantly catches a jump in the picture when the audio also cuts, because now both channels lurch at once and the illusion of continuous time breaks. B-roll exploits this asymmetry: keep the audio flowing and the eye will happily wander wherever you send it. That is why the single most common professional cut — sound continuing over a change of picture — is invisible, and why an interview without B-roll has no invisible cuts available to it at all.

🔄 Check Your Eye. Answer from memory before reading on: 1. In one sentence each, what is A-roll and what is B-roll? 2. Name the four jobs B-roll does in an edit. 3. Why does laying B-roll over a trimmed interview make the jump cut disappear?

Check yourself

1. A-roll is the main footage the piece is about (the interview, the piece-to-camera, the core action, carrying the words and the spine); B-roll is the supplementary footage laid over and around it (cutaways, inserts, sequences, atmosphere) that shows what is described and covers the cuts.
2. It covers your cuts (hides jumps when you trim/reorder), shows what is described (show-don't-tell), compresses and controls time, and adds visual interest / controls pace.
3. Because the audio runs continuously while only the picture changes — and the eye accepts a change of picture under a continuous voice, so the splice in the interview is hidden behind the cutaway.

20.2 Cutaways, inserts, and sequences

"B-roll" is the family name. Inside it live three specific tools you will call for by name on every shoot, and knowing which is which turns "grab some B-roll" into a precise plan. All three were previewed back in Chapter 7 (§7.6 promised them to this chapter); here they get their proper definitions.

A cutaway is a shot of something other than the main subject or action, still belonging to the world of the piece — a shot the editor can cut away to. In an interview, a cutaway is anything that is not the talking head: the subject's workshop, a listener nodding, the street outside, the thing they are describing, a wide of the room. Its defining feature is that it points the camera elsewhere — away from the A-roll — at something related. Cutaways are what you lay over interview cuts (FIGURE 20.1) and what you use to establish where you are.

An insert is a close-up shot of a specific detail within the main action, inserted into it — the hands on the tool, the finger on the button, the pen signing the line, the phone screen, the cup landing on the saucer. Where a cutaway looks away to something else, an insert looks closer at something that is right there in the scene. The order-counter close-up of the cup on the saucer you shot in Chapter 7 (FIGURE 7.6, shot 5) was an insert: a detail from inside the action, framed tight. Inserts show the thing that matters too small to read in the wider shot, and — like cutaways — they give the editor a place to hide a trim.

The distinction is real but not a wall: a tight shot of a detail inside the current action is an insert; a shot of something elsewhere is a cutaway. Many shooters use "cutaway" loosely for both, and nothing bad happens when they do. Know the precise meanings so you can plan deliberately, and do not lose sleep over a borderline case.

The third tool is the one this chapter cares about most, because it is where B-roll stops being single shots and starts being storytelling. A sequence is a coherent series of shots of a single action, covered from different sizes and angles — classically wide, medium, and detail — shot so that they cut together into one complete, compressed rendering of that action. A person making a cup of coffee, sharpening a knife, lacing a boot, throwing a pot: filmed as a sequence, that one action becomes a little self-contained scene — establish it wide, show the doing in a medium, punch into the detail, and the three cuts read as one fluid, unbroken action even though you may have shot each size separately. A sequence is coverage (Chapter 7) applied to a process instead of a conversation. It is the highest-value B-roll you can shoot, and §20.3 is entirely about how.

FIGURE 20.2 — The three tools inside "B-roll": what each one points at

                              ┌──────────────  B-ROLL  ──────────────┐
                              │      (everything that isn't A-roll)     │
        ┌─────────────────────┼──────────────────────┬────────────────┴────────┐
        ▼                     ▼                      ▼                          ▼
   A-ROLL (not B-roll)     CUTAWAY                INSERT                     SEQUENCE
   the interview /         a shot AWAY to         a tight detail             a single ACTION
   the main action         something else in      WITHIN the main            covered wide +
                           the world of the       action (hands, a           medium + detail,
   "the subject talks"     piece (the shop,       screen, the cup)           cut into one
                           a listener, the        "the thing, up close"      compressed scene
                           street, the result)                               "the whole action,
                           "look over here"       "look right here"           built from pieces"

   RULE OF THUMB:  cutaway = look ELSEWHERE.   insert = look CLOSER.   sequence = the whole
   action, told in wide→medium→detail.  All three are B-roll; all three cover cuts; only the
   sequence tells a mini-story on its own.

The taxonomy matters because it converts a vague instinct — "get some nice shots" — into a checklist you can actually shoot against. Standing in a location, you are no longer hoping for inspiration; you are asking three concrete questions: What can I cut away to (cutaways)? What details need to be seen up close (inserts)? What actions should I cover as full sequences? Answer those and you will not leave empty-handed.

The predictable way beginners waste a B-roll day is to answer none of those questions and instead shoot whatever looks pretty.

⚠️ Common Mistake: the beautiful shot of nothing. The classic beginner B-roll reel is a pile of gorgeous, unmotivated shots — a slow-motion push across some leaves, a lens flare, the coffee cup rotating on a turntable — none of which the story ever asked for. It looks like "cinematic B-roll," and in the edit it is useless, because it does not cover any specific cut or show any specific beat. Then the shot you actually needed — the subject's hands doing the exact thing they described — was never captured, because you spent the light on leaves. The fix is the rule that runs through this whole book: motivate every choice. Every B-roll shot should answer a story question ("this shows the craft she mentioned") or a technical one ("this covers the cut into her second answer"). Pretty is a bonus, never the brief. If you cannot say in one sentence what a shot is for, do not spend your daylight on it.

🔗 Connection: this is show-don't-tell, wearing work boots. In Chapter 17 (§17.4) you learned to convert a claim into a shootable beat — "she's dedicated" becomes "6:40 a.m., empty room, one light, already at work." B-roll is where you shoot those beats. The FIGURE 17.5 table of claims-turned-into-behaviors is, functionally, a B-roll list. And the A-story / B-story braid from FIGURE 17.6 tells you which B-roll matters: the A-story shots show the "what" (how the work is done), the B-story shots show the "why" (the worn tools, the private smile, the empty chair). Re-read §17.4–§17.6 as a shot-gathering guide and this chapter becomes the day you go get them.

🔄 Check Your Eye. From memory: 1. What is the difference between a cutaway and an insert — where does each point the camera? 2. Define a sequence. Which of the three tools tells a little story on its own, and why? 3. You are about to shoot "some nice B-roll" of a bakery. Give the one-sentence test that separates a useful shot from a pretty-but-useless one.

Check yourself

1. A cutaway points away from the main subject to something else in the world of the piece (the room, a listener, the result); an insert points closer, at a specific detail within the main action (the hands, the screen, the cup).
2. A sequence is a coherent series of shots of a single action, covered wide/medium/detail, cut into one compressed rendering of that action. The sequence tells a story on its own because it has a beginning (establish), a middle (the doing), and an end (the result/detail) — a cutaway or insert is a single beat.
3. "What is this shot for?" If the answer is a story beat it shows or a specific cut it covers, shoot it; if the only answer is "it looks cool," it is probably the beautiful shot of nothing.

20.3 Shooting a sequence (wide/medium/detail)

If you master one thing in this chapter, make it the sequence, because it is the move that turns a person doing an ordinary task into footage that looks and cuts like a film. The recipe is the master-and-coverage method from Chapter 7 (§7.5), pointed at an action instead of a conversation, and it has three obligatory sizes:

1. The wide establishes the action and its space: who is doing what, where. It is the sequence's establishing shot — the whole person at the workbench, the whole counter, the entire wheel and the studio around it. It answers "what am I looking at?" so the tighter shots that follow have a home.
2. The medium shows the doing: the hands and the tool and the body working together — waist-up or thereabouts, close enough to read the activity, wide enough to keep it legible. This is the working heart of the sequence.
3. The detail (a close-up or extreme close-up insert) isolates the one thing that matters most: the fingers on the clay, the blade meeting the stone, the espresso threading into the cup. This is the shot that makes a viewer feel the craft, and it is the one beginners forget and editors beg for.

Three sizes of one action, and — this is the part that makes them cut together — you shoot each size covering the same moments, so the action overlaps across them exactly as your scene coverage did in Chapter 7. With one camera, you simply have the subject repeat the action (most real processes repeat naturally — there is always another pour, another pot, another board to plane), running it once for the wide, again for the medium, again for the detail. Because the action is essentially the same each pass, the editor can cut from the wide to the medium to the detail and it reads as one continuous, fluid action — and you will hide the fact that they are three separate takes with a technique you already know: matching action (Chapter 9). Cut on the movement — as the hand descends in the wide, land in the medium mid-descent — and the motion carries across the cut so smoothly the seam vanishes.

FIGURE 20.3 — A sequence: one action (pulling a coffee), covered wide → medium → detail

   SHOT 1: WIDE                 SHOT 2: MEDIUM              SHOT 3: DETAIL (ECU insert)
  +----------------------+     +----------------------+   +----------------------+
  |   [==== counter ====]|     |     .-. steam        |   |        ~ ~ ~          |
  |    ( barista )   ☀win|     |    (hands on the     |   |     __||__  the pour  |
  |     at the machine   |     |     portafilter)     |   |    |cup   | threads   |
  |   whole action+room  |     |   the DOING, waist-up|   |   crema blooming, ECU |
  +----------------------+     +----------------------+   +----------------------+
   "where + who + what"         "the hands working"        "the one thing that matters"
   establish the action         the working heart          make them FEEL it

   CUT them on matching action (Ch.9): the down-press of the tamp in the WIDE lands mid-press in
   the MEDIUM; the stream starting in the MEDIUM continues in the DETAIL. Same action, three sizes,
   shot in three passes with one phone -> reads as ONE fluid, unbroken action. That is a sequence.

Now watch it become an edit. Here is the same coffee-pull rendered as a Described Sequence — the shot-by-shot table the book uses to show a cut — so you can see how three sizes of one repeated action assemble into a five-second scene that feels whole:

FIGURE 20.4 — The coffee-pull sequence, covered and cut        [constructed teaching example]
  #  | Shot (size / subject)              | Dur  | Audio (wild)               | Cut to next
  ---+------------------------------------+------+----------------------------+----------------------
  1  | WIDE: barista at the machine,      | 2 s  | machine hum, portafilter   | cut IN on the tamp,
     | window behind, whole action & room |      |  click                     |  on the down-motion
  2  | MEDIUM: hands lock in, tamp down   | 2 s  | the tamp's tap, a hiss      | cut on the press to
     |  (waist-up, the doing)             |      |                            |  the detail
  3  | ECU insert: espresso threads into  | 3 s  | the trickle, crema fizz    | hold, then cut back
     |  the cup, crema blooming           |      |                            |  wide to reset
  4  | ECU insert: steam curling off the  | 2 s  | steam hiss, spoon clink     | cut to the cup lift
     |  pitcher, rim-lit by the window    |      |                            |
  5  | MEDIUM: the finished cup set on the| 2 s  | saucer clink, "here you go"| out to the next beat
     |  saucer, slid forward              |      |                            |

Read the table as an editor: shot 1 establishes the whole action so we are oriented; shots 2 and 3 carry us into the doing and then the intimate detail (the thread of espresso is the beat that lands); shot 4 is a bonus insert — the steam, rim-lit by the café's window, pure atmosphere that also covers time; shot 5 resolves the action with the finished cup. Eleven seconds of screen time, built from a repeated ninety-second real process, and it reads as one seamless act of making coffee. You did not film "the barista making coffee." You filmed the pieces of it — and, just as with the order-counter coverage in Chapter 7, the pieces are what let the edit exist.

🎞️ Read This Sequence. Notice that FIGURE 20.4 obeys the same three rules as all coverage: at least three sizes; the same action overlapping across them; a clear establishing shot and a clear detail. If you can cover any action this way — coffee, clay, a knife on a stone, a hand-set line of type — you can make any process cinematic on any camera.

✂️ In the Edit: a well-shot sequence cuts itself. When you deliver wide, medium, and detail of one action with matching action across them, the edit is nearly automatic — the shots want to go in that order, the motion carries the cuts, and the whole thing assembles in minutes. When you deliver only a single locked-off wide of the same action, the editor has one shot, no way to compress it, no detail to punch to, and no cover for a trim: they are stuck watching ninety real seconds of coffee in real time, or making an ugly jump. The difference between a sequence that edits itself and footage you fight is entirely three extra passes on the day. In Chapter 30 you will lay sequences like this over your interview to build a whole documentary; everything you can do there is decided by whether you shot the pieces here.

🎬 On Set: shoot a five-shot sequence of one action. Pick any everyday action that takes 20–60 seconds and repeats naturally — pouring coffee, sharpening a pencil, tying a shoe, watering plants, wiping down a counter. Constraint: cover it in at least five shots including a wide (establish the action + space), a medium (the doing), and at least two details/inserts (the close moments that matter) — shot in separate passes, with the action repeated each time, and one shot grabbed on a matching-action movement so it will cut on the motion. Keep the camera steady (brace it; a phone on any surface beats a shaky handheld). Self-review question: drag the five shots into any editor, order them wide→medium→detail, and cut each on movement — does it read as one fluid action, or can you see the seams? If you can see a seam, find the frame where the motion doesn't match and re-cut on the movement. When it flows, you have shot your first sequence. (This is a rehearsal for the Production Checkpoint, where you will do it for your documentary subject.)

🔄 Check Your Eye. From memory: 1. Name the three obligatory sizes of a sequence and the job of each. 2. With one camera, how do you get three sizes of one action that still cut together? 3. Which Chapter 9 technique hides the fact that a sequence is three separate takes?

Check yourself

1. Wide (establish the action and its space — who/what/where), medium (show the doing — hands, tool, body working), detail/insert (isolate the one thing that matters most — the close moment that makes it felt).
2. Have the subject repeat the action and shoot it once per size, keeping the same action so it overlaps across the sizes — one pass wide, one medium, one detail.
3. Matching action — cutting on the movement so the motion carries across the cut and the three separate takes read as one continuous action.

20.4 Matching B-roll to the story beats

You can shoot flawless sequences and still make a shapeless film, because a pile of beautiful cover is not a story. The thing that turns B-roll from wallpaper into storytelling is that it is matched to the beats — every important shot exists to show, prove, or advance a specific beat of the piece. Story is the boss (Chapter 1, §1.3), and B-roll is one of the boss's most powerful employees; it works only when it takes orders from the story.

You already built the orders in Chapter 17. Your documentary has a spine — a hook, a middle, a resolution — braided from an A-story (the "what": how the work is done) and a B-story (the "why": what it means, who they are). Each beat on that spine implies a shot that would show it. The interview supplies the feeling on the audio track; the B-roll supplies the proof on the picture track; laid together, they are far more persuasive than either alone. When your subject says, on the A-roll, "I've done this thirty years and I still love it," you do not leave the viewer staring at a talking head — you lay over it the B-roll that earns the sentence: the hands moving with unmistakable ease, the wall of tools worn smooth, the small private smile when a piece comes out right. Words carry the claim; pictures carry the evidence.

That layering has a name and a shape you will meet formally in the edit, and it is worth seeing now so you know what you are shooting for. When the interview audio keeps running while the picture cuts to B-roll, the sound leads the picture across the cut — an L-cut. It is the most common professional cut in all of nonfiction, and B-roll is its fuel.

FIGURE 20.5 — B-roll over an interview: the sound leads, the picture proves (an L-cut)

  V1  [ interview: talking head ##### ][ B-roll: hands at work ############ ][ head ##### ]
  A1  [ interview sync audio: "...I've done this thirty years and I still love it..." ####### ]
  A2  [ wild sound: the workshop — a plane, a hum, birds outside .............................. ]
                              ^ picture cuts to B-roll here          ^ picture cuts back
                                (audio never stops — the L-cut)

   BEAT (Ch.17)              WHAT WE HEAR (A-roll)              WHAT WE SEE (B-roll proves it)
   ----------------------    -------------------------------    ------------------------------------
   hook / who they are       "I still love it"                  a sequence: hands working with ease
   A-story: how it's done    "every joint is cut by hand"       insert: the chisel entering the wood
   B-story: why it matters   "nobody makes them like this"      cutaway: a shelf of finished pieces;
                                                                 the empty chair; the worn tools
   resolution: the change    "and that's why I keep going"      cutaway: the shop sign; a customer's
                                                                 face receiving the finished piece

   The interview is the SPINE. The B-roll, matched beat-by-beat, is the FLESH. Because the audio
   runs unbroken beneath, every cut to B-roll is invisible AND it proves the sentence it plays under.

Study the mapping in the right two columns, because it is the entire craft of interview-driven video in one table. For every beat, there is a thing the subject says and a thing you must show. The subject cannot say "these hands have skill" convincingly — but a three-shot sequence of the hands working proves it beyond argument. The subject can claim the craft is dying, but a slow cutaway across a shelf of finished work and an empty apprentice's chair makes you feel it. Your job on the B-roll day is to walk down the beat outline from your Chapter 17 Production Checkpoint and, for every line the subject will say, shoot the picture that proves it.

✂️ In the Edit: B-roll does two jobs at once — hides the cut and proves the point. This is why matched B-roll is worth so much more than pretty B-roll. A shot of the subject's hands, cut in over the sentence "every joint is cut by hand," simultaneously (1) covers the interview edit underneath it, so you could have spliced two answers together at that exact frame invisibly, and (2) proves the claim the audio is making. One shot, two jobs. An unmotivated pretty shot does only the first job, and does it while saying nothing. When you gather B-roll matched to beats, you are loading the edit with shots that each pull double duty — which is exactly why an interview covered with the right B-roll cuts faster, shorter, and more persuasively than one covered with random footage. You will do this layering for real in Chapter 30 (B-roll-driven narrative); today you make sure the shots exist to do it with.

Now bring it home to the scene we have been building since page one. The Café Scene has been framed (Chapter 6), covered as a two-person order exchange on one side of the line (Chapter 7), given a push-in (Chapter 8), continuity-matched on the walk-in (Chapter 9), lit as an interior (Chapter 11), shot with the window as its natural key (Chapter 13), and recorded for clean order-counter dialogue and room tone (Chapter 15). Today we shoot its B-roll — the cutaways and inserts that would let an editor cover the order exchange and give the scene its texture. Nothing about the café changed. What changed, again, is what you now know to point the camera at.

Here is one of those cutaways rendered as a full Described Shot, so you can picture exactly what "B-roll of the café" means and see how it obeys the scene's locked geography — the big window is the room's key, camera-right behind the counter, so steam rising in front of it is rim-lit and glows:

FIGURE 20.6 — "The steam" (a café cutaway that will cover the order exchange)   [constructed teaching example]
  THE FRAME    Extreme close-up: a ribbon of steam rising off the milk pitcher, filling the frame against
               the dark of the machine; the counter's warm lamp glows soft and out of focus behind it.
  THE MOVE     Locked off, braced on the counter. The only movement is the steam itself, curling and
               breaking — a living detail in an otherwise still frame.
  THE LIGHT    The big window (camera-right, the room's key since Chapter 13) rakes across the steam from
               behind and turns it luminous — rim-lit, bright against shadow. Nobody added a light; the
               window that lights the whole room lights the steam too.
  THE SOUND    Wild sound, grabbed with the shot: the hiss of the steam wand, the tick of the machine.
               No dialogue — this is picture and texture, recorded clean so the editor can use it anywhere.
  THE CUT      Pure B-roll. In the edit it lays OVER the order exchange (Chapter 7's coverage): as the
               customer says "a flat white, please," we cut away to this steam, hiding the join between
               two takes and letting the café breathe. An L-cut — the order audio keeps running beneath.
  THE EFFECT   The frame narrows the whole café to one small, sensory, beautiful thing. It says "this is a
               real place where real coffee is made" without a word, and it buys the editor a free cut.
  THE LESSON   A cutaway is a gift to the edit AND a piece of storytelling. The steam covers a splice and
               sells the café at the same time — proof that the best B-roll always does more than one job.

That steam, plus a matching insert of the barista's hands on the portafilter, plus a cutaway to the café's sign on the door catching the morning light, plus the regulars at the window — those four shots are all the B-roll the order exchange needs. Lay them over the Chapter 7 coverage and the scene can be trimmed, tightened, and paced freely: cut to the steam to lose a fumbled line, cut to the sign to compress the wait, cut to the hands to cover the join between the wide and the reverse. The scene never changed — your command of it did. You will actually perform these cuts in Chapter 28 (J- and L-cuts on this exact order exchange); today you make sure the cutaways are in the can.

🔗 Connection: where this B-roll gets laid in. You are shooting cover today; the laying-in happens in the edit. The L-cut you see in FIGURES 20.5–20.6 is taught in full in Chapter 28 (§28.1, sound-leading cuts). Building an entire interview-driven piece by laying B-roll over a spine of interview selects — the technique this whole chapter feeds — is Chapter 30 (B-roll-driven narrative). And the specific mechanics of dropping a clip onto a second video track above your interview live in your editor of choice (in DaVinci Resolve, the book's reference, that is placing the B-roll on the track above and trimming its handles; see Appendix E). You do not need any of it yet. You need the shots to exist so those chapters have something to work with.

🔄 Check Your Eye. From memory: 1. What does it mean to "match B-roll to the beats," and which chapter's beat outline do you shoot against? 2. In an L-cut over an interview, which track keeps running — and why does that make the picture cut invisible? 3. Name the two jobs a single well-matched B-roll shot does at once.

Check yourself

1. It means shooting a specific picture to show, prove, or advance each beat of the story — you shoot against the A-story/B-story beat outline you built in Chapter 17's Production Checkpoint (the spine).
2. The interview sync audio (the A-roll, on the audio track) keeps running while the picture cuts to B-roll; because the sound is continuous, the eye accepts the change of picture and the interview's edit underneath is hidden.
3. It covers the interview cut (hides a splice) and proves the point the audio is making (show-don't-tell) — one shot, two jobs.

20.5 A B-roll shot list that covers you

Coverage that lives only in your head becomes coverage you forgot to shoot — you learned this for scene coverage in Chapter 7, and it is doubly true for B-roll, because B-roll is precisely the material you are most likely to under-shoot in the rush of a real day. The written form of B-roll coverage is the coverage plan: a B-roll shot list, drawn directly from your story beats, that names every cutaway, insert, and sequence the edit will need and lets you tick each one off as you get it. It is the same idea as Chapter 16's shot list and Chapter 7's coverage checklist, aimed now at the specific job of covering a documentary — and it is the difference between leaving a location done and leaving it hoping.

The coverage plan is built, not invented. You already have its source: the beat outline from Chapter 17 and the "what would I see if this were true?" question from show-don't-tell. Walk down your beats; for each one, write the shot (or the sequence) that shows it; group the shots by location and by subject so you shoot efficiently; and mark which are must-haves versus nice-to-haves so that when the light or the schedule runs out, you sacrifice the right shots. Here is a coverage plan for a short documentary profile — the craftsperson from FIGURE 20.5 — rendered as the page you would actually carry:

FIGURE 20.7 — A coverage plan (B-roll shot list) built from the story beats

  BEAT (from the Ch.17 spine)     | TYPE      | SHOT / SEQUENCE                    | PRIORITY | GOT?
  --------------------------------+-----------+------------------------------------+----------+-----
  HOOK: "I still love it"         | sequence  | hands working, WIDE+MED+DETAIL     | MUST     | [ ]
  establish: where we are         | cutaway   | WIDE of the shop; the sign outside | MUST     | [ ]
  A: "every joint by hand"        | sequence  | the joint being cut, W+M+ECU       | MUST     | [ ]
  A: the tools of the trade       | insert    | ECU pans across the worn tools     | nice     | [ ]
  B: "a dying craft"              | cutaway   | shelf of finished work; empty      | MUST     | [ ]
                                  |           |  apprentice's chair                |          |
  B: who they are                 | cutaway   | the private smile; a photo on wall | nice     | [ ]
  process / time passing          | sequence  | glue-up or finish, W+M+DETAIL      | MUST     | [ ]
  resolution: the result          | insert    | ECU the finished piece, turning    | MUST     | [ ]
  resolution: the human payoff    | cutaway   | a customer receiving the piece     | nice     | [ ]
  ATMOSPHERE (safety net)         | cutaway   | dust in the window light; the door | nice     | [ ]
  --------------------------------+-----------+------------------------------------+----------+-----
  WILD SOUND to grab everywhere:  the plane, the mallet, the shop's hum, the door, 60s of room tone
  ALWAYS: at least 3 sizes per action; brace every shot; get MORE than the MUSTs before any "nice."

Read the plan and you can see the finished film's picture track before a frame is shot: it will open on a sequence of the hands (the hook), establish the shop, prove the "by hand" claim with another sequence, deepen with the shelf and the empty chair, pass time with a finish sequence, and resolve on the piece and the customer. The plan also does the quiet, essential thing a list does: it asks you, row by row, whether you got each shot — so you will not drive home and discover in the edit that you have no establishing wide, no detail of the joint, no resolution shot. That is the whole reason it exists.

💡 Why It Works: the list converts memory into a guarantee. On a real shoot your attention is saturated — the light is changing, the subject is nervous, the battery is low, a phone is ringing. Human memory fails exactly under that load, which is why "I'll remember to grab the wide" is a promise you will break. A written plan offloads the remembering onto paper, so your saturated attention can be spent on making the shots good instead of remembering which shots to make. The plan is not bureaucracy; it is a prosthesis for the part of your brain that a busy set overwhelms. The shot you checked off a list is free; the shot you meant to get and forgot costs you a second trip or a hole in your film.

This is the last of the book's five throughlines showing its face in this chapter: fix it in pre, not in post. The B-roll you planned and captured is cheap; the B-roll you improvised and missed is a reshoot, a scheduling nightmare, or — most often — a compromise, an edit built around a hole. A page of rows, filled in before you leave the house, is the cheapest insurance in all of documentary production.

🔗 Connection: the plan's cousins and where to find the blank. The coverage plan is a specialized shot list (Chapter 16, §16.3), which is itself a written form of coverage (Chapter 7, §7.5–7.6). It is built from the beat outline and the show-don't-tell table of Chapter 17 (§17.4, §17.6). A blank coverage-plan template you can copy — along with shot lists, storyboards, and call sheets — lives in Appendix I. If your plan has a beat with no shot next to it, that empty cell is a hole in your film; find it now, on the page, where it costs a sentence to fix.

🔄 Check Your Eye. From memory: 1. What is a coverage plan, and what document from Chapter 17 do you build it from? 2. What one job does a written B-roll list do that a plan in your head cannot? 3. Why mark shots as "must-have" versus "nice-to-have"?

Check yourself

1. A coverage plan is a B-roll shot list drawn from the story beats, naming every cutaway, insert, and sequence the edit needs; you build it from the Chapter 17 beat outline (the spine) by writing the shot that shows each beat.
2. It guarantees the pieces by asking you, row by row, whether you got each shot — so you don't leave a location missing the establishing wide, the key detail, or the resolution shot; memory fails under a busy set's load, a list doesn't.
3. So that when light or time runs out, you sacrifice the right shots — you finish all the must-haves (the shots the story cannot exist without) before spending any daylight on the nice-to-haves.

20.6 Getting enough — the coverage mindset

Everything so far has been technique. This last section is a habit, and it is the one that most reliably separates people whose interviews cut beautifully from people who are forever one shot short. It is the coverage mindset: the working assumption that you can never come back, so you shoot more than you think you need, in more sizes and angles than feel necessary, with each shot's own sound — because footage is cheap on the day and priceless in the edit.

Start with the ratio, because it shocks beginners. Professional B-roll shooters routinely capture ten, twenty, thirty times more B-roll than ends up in the film. A three-minute documentary might use ninety seconds of B-roll and be built from forty-five minutes of it. That is not waste; it is insurance and choice. You cannot know on set which shot will cut best against which sentence, which angle the story will end up needing, or which beautiful accident will become the moment everyone remembers. So you gather generously and choose ruthlessly later. The rule of thumb — you first met a version of it in Chapter 1 — is shoot five, use one, and for B-roll specifically, err even higher. Running out of card space is a twenty-second problem; running out of B-roll is a hole in your film.

Generous does not mean random. The coverage mindset has four disciplines that keep the abundance useful:

• Variety of size and angle. For every subject, get a wide, a medium, and a detail at minimum — and then move: a higher angle, a lower one, the other side (respecting the line, Chapter 7), a different lens feel. Sameness is the enemy; ten shots of one subject from one spot is one shot, ten times. Ten shots from ten positions is a scene.
• Motion and stillness both. Get the locked-off version and, where it's motivated, a slow move — a gentle push, a reveal, a drift across the tools (Chapter 8). But get the static one first; a steady shot is always usable, and a bad move is not.
• Every shot gets its own sound. This is Theme 2 — sound is half the picture — applied to B-roll. As you shoot the hands, record the sound of the hands; as you shoot the steam, grab the hiss. This is wild sound (sometimes "nat sound"): audio recorded wild, not synced to a specific line, to be laid under the B-roll in the edit. It is the cousin of the room tone you learned to record in Chapters 14–15, and it is what makes B-roll feel alive instead of like a silent slideshow. A shop that sounds like a shop under your picture is worth a dozen music cues.
• The safety net. Before you leave, get the shots you will always need no matter how the edit turns out: an establishing wide of the location, a clean detail of the subject's hands, sixty seconds of room tone, and a couple of neutral cutaways (the door, the light, the sign). These are the shots that rescue an edit whose story changed — and in documentary, the story always changes.

That last point deserves its own weight, because it is where the coverage mindset meets the reality of nonfiction you learned in Chapter 17. You will discover the real story in the edit — the B-story that turned out to matter more than the A-story you pitched. When that happens, the only thing that lets you tell the truer story is having shot enough coverage to build it. The reader who shoots exactly their planned list and no more is trapped by their plan; the reader who shoots the plan and then keeps going — grabbing the surprising shot, the extra angle, the thing that caught their eye — is the one who can follow the story wherever it leads. Plan the coverage; then shoot past the plan.

🚪 Threshold Concept: "B-roll" is a terrible name for the most important footage you'll shoot. The "B" makes it sound secondary, optional, the stuff you get if there's time. Reverse that belief permanently. In an interview-driven piece, the B-roll is what makes the A-roll usable — it is the reason the interview can be cut at all, the thing that proves every claim, the material that carries the story visually. A film with a great interview and no B-roll is unwatchable; a film with a decent interview and rich, matched B-roll can be superb. The footage labeled "B" is doing the heaviest lifting in the edit. Once you stop treating B-roll as the leftover and start treating it as the main event of your shoot day — planned first, shot most, protected fiercely — your nonfiction work changes level. It is not B-material. It is the film.

🎒 Gear Note: B-roll rewards the shooter, not the camera. B-roll is the most gear-agnostic work in this whole book, because it has no dialogue to sync, no performance to nail in one take, and all the time in the world. A phone braced on a stack of books shoots a flawless locked-off insert; a phone's real wide lens gets you a clean establishing shot; your feet get you every size. What does help, in order of value: something to stabilize (a mini tripod, a beanbag, a table edge — steadiness reads as "intentional"); a way to grab wild sound (even the phone's own mic, held close to the source for a few seconds per shot); and, far down the list, a longer lens for compressed detail shots and a way to shoot slow motion (a higher frame rate — 60 fps or more — for a silky detail, which you met in Chapter 2 and will meet again in Chapter 25). None of it is required. The discipline of getting the sizes and matching the beats is free, and it is 95 percent of the result. There is no camera that shoots a sequence you didn't cover, and no camera that fails to shoot one you did.

⚠️ Common Mistake: not enough, and all the same. Two failures, one cause. The first: shooting too little B-roll, so the edit runs out of cover and the interview is forced back into visible jump cuts and long static holds. The second: shooting a lot of B-roll but all of it the same — same size, same angle, same locked-off framing of the same subject — so despite the quantity there is nothing to cut between. Both come from not treating B-roll as the main event. The fix is the mindset and the plan together: walk in with a coverage plan built from your beats, shoot every action in at least three sizes, keep moving your feet and your angle, grab wild sound on everything, and do not stop at "I think that's probably enough." It is almost never enough. Shoot past the point where it feels excessive; the edit will thank you and the reshoot you avoided will never happen.

♿ Accessibility & Inclusion: B-roll is the sound-off viewer's whole story — and the blind viewer's challenge. Two habits, both decided today. First, roughly one in five viewers watches with the sound off, and every deaf or hard-of-hearing viewer always does — for them, your B-roll is the film, because the interview audio is silent. So make sure your matched B-roll carries the story on its own: if the picture track, muted, still shows the craft and the beats, you have served those viewers. Second, the opposite viewer — someone who is blind or low-vision — gets the story from the audio, which means your beautiful wordless sequences are lost to them unless the piece has audio description or narration that conveys what the picture shows. The skill of describing a shot in words (the Described Shot you have practiced since Chapter 1) is exactly the skill behind writing that description. And a practical consent note: when your B-roll includes other people — customers, passersby, a subject's family — the same releases you get for on-camera subjects apply (we treat model and property releases formally in Chapter 38). Shoot inclusively, get consent, and build the piece so both kinds of viewer receive the story.

🔄 Check Your Eye. From memory: 1. Roughly what shooting ratio should you expect for B-roll, and why is the "waste" actually valuable? 2. What is wild sound, and which earlier practice is it the cousin of? 3. Why does the coverage mindset — shooting past your plan — protect you specifically in documentary?

Check yourself

1. Ten-to-one, twenty-to-one, or higher — you use a small fraction of the B-roll you shoot. The "waste" is insurance and choice: you can't know on set which shot will cut best or how the story will change, so you gather generously and select ruthlessly later.
2. Wild sound (nat sound) is audio recorded wild — not synced to a specific line — to be laid under B-roll in the edit; it is the cousin of room tone (Chapters 14–15) and it makes B-roll feel alive.
3. Because in documentary the real story is often found in the edit (Chapter 17) — the B-story that mattered more than planned — and you can only tell the truer story if you shot enough coverage to build it; shooting past the plan is what leaves you the footage to follow the story.

Production Checkpoint

In Chapter 17 you drafted the story spine of your 3-minute documentary short. In Chapter 19 you shot its interview — the A-roll, with clean two-mic audio and full-sentence answers. Today you gather everything the edit will lay over that interview.

Your task: shoot a B-roll list for your documentary — cutaways and at least one full sequence — matched to your story beats.

Work in this order, because the order is the lesson:

1. Build the coverage plan first. Take your Chapter 17 beat outline and, for every beat, write the cutaway, insert, or sequence that would show it (the FIGURE 20.7 format). Mark must-haves and nice-to-haves. This page is your guarantee.
2. Shoot at least one full sequence of your subject's core action — a real process they do — covered wide, medium, and detail (§20.3), with the action repeated for each size and at least one cut planned on matching action.
3. Shoot the cutaways and inserts the plan calls for: the establishing wide of the location, the details that prove the interview's claims, the atmosphere and the human beats, the resolution shot.
4. Grab wild sound on everything — a few seconds of each shot's own sound — plus sixty seconds of room tone.
5. Then shoot past the plan. Get the surprising shot, the extra angle, the thing that caught your eye. Aim to feel like you got too much.

Why this matters: with today's shoot, Project 2 is fully captured — you now hold the interview (Chapter 19) and all the B-roll to cover and prove it, which is the complete raw material of a documentary short. Everything from here is the edit: in Part VI you will build the spine from your interview selects and lay this B-roll over every cut (Chapters 28 and 30) to make the film your Chapter 17 index card promised. You have just done, for your own project, exactly what you learned to do for the Café Scene — and you did it as someone who shoots for the edit. Keep your coverage plan and your footage organized together; you will be very glad of it when the timeline opens.

Summary

Reference-grade recap — the vocabulary, the recipes, and the decision aids for B-roll and coverage.

The core definitions:

The four jobs of B-roll: cover your cuts (hide jumps when you trim/reorder) · show what's described (show-don't-tell) · compress and control time · add visual interest and control pace.

The sequence recipe (§20.3):

Shoot each size covering the same action (repeat it), so it overlaps; cut on matching action (Chapter 9) so three passes read as one fluid action.

The B-roll coverage checklist — before you leave any location, do you have:

When to use what: a cutaway when you need to look elsewhere (establish, compress, cover a cut to another beat); an insert when a detail is too small to read in the wider shot; a sequence when an action deserves to be a scene. Match every important shot to a story beat — it should hide a cut and prove a point. Write it all down as a coverage plan so you don't leave a piece behind — fix it in pre, not in post. And remember the ratio: shoot far more than you'll use; the "B" is doing the heaviest lifting in the edit.

Spaced Review

Retrieval keeps knowledge. These revisit Chapter 7 (coverage and shots) and Chapter 19 (the interview) — the two chapters this one stands directly on. Answer from memory, then check.

1. Chapter 7: define coverage in one sentence, and explain how the master-and-coverage method relates to the way you shoot a sequence in this chapter.
2. Chapter 7: what is the line (the axis of action), and why does it still matter when you shoot B-roll cutaways of two people in a room?
3. Chapter 19: in the interview setup, why does the subject look off-axis (just beside the lens) rather than into it — and how does that A-roll framing make B-roll more necessary, not less?
4. Chapter 19: you ran two mics and asked for full-sentence answers. How does a full-sentence answer make the B-roll you shoot today easier to lay in?
5. Integrative: you covered the Café order exchange in Chapter 7 (a wide, two mediums, a close-up, an insert). Today you shot café B-roll (the steam, the hands, the sign). Which is A-roll and which is B-roll here — and in the edit, which one covers the cuts in the other?

What's Next

With today's B-roll in the can, you have done something worth pausing on: you have run a complete production. Across Part IV you planned a documentary (Chapters 16–18), shot its interview (Chapter 19), and gathered its coverage (Chapter 20) — the whole raw material of a real film now sits on your cards, waiting for the edit. Part IV is finished, and so is the shooting half of this book's spine for Project 2.

Chapter 21 opens Part V — Genres and Formats, beginning with Documentary and Nonfiction: the modes from vérité to essay, how to build a truthful story from what's real, narration and archival, and the ethics of shaping reality without falsifying it. Everything you shot in Chapters 19 and 20 — the interview and the B-roll — is documentary raw material, and Chapter 21 is where you learn the form it's about to become. You have gathered the pieces. Next you learn the genre that assembles them into truth.