The Spy Show: editing a spy-history documentary series

How is a spy-history documentary series edited?

Start by measuring the reference channels, then pick a picture for what the narration says in each second: the place, the time of day, the object named. Keep every image true to its period, never let a stranger's face stand in for a real person, and log the source and license of every asset.

What is The Spy Show?

The Spy Show is a spy-history documentary series run by Vincent Meijer. Each episode is long-form and narrated: a voiceover tells a true story, and the picture has to carry it. I am editing the series, and it is still in production.

The documentary editing page has mentioned "a spy-history documentary series (in production)" for a while. This page is where that series gets its name. Episodes, the people in them and their topics are left out on purpose until they are published.

Spy history is a demanding genre to edit. Much of what the narration describes happened in secret, so there is often no photograph of the moment itself. The edit has to find honest ways to show a meeting nobody photographed, a message nobody kept, or a person whose few surviving portraits run out long before the story does.

What did the brief ask for?

Vincent's brief named two reference channels:

1. Main reference: Philip Thompson, "The Unbreakable Spy of WW2 | True Life Spy Stories".
2. Secondary reference: Librarian, "The Real James Bond: The True Spy You've Never Heard Of".

The instruction that came with them was the useful part: the goal was not to copy either channel, but to match their level of quality, rhythm and narrative. That is a good brief, and a hard one to hit by feel. "Match the level" can mean almost anything until someone writes down what the reference actually does. So before any cutting, both references were measured.

How were the two reference channels measured?

Both videos were analyzed frame by frame: how often the picture changes, how long shots last, how shots join, how much text is on screen, how much of the picture moves, and where the music sits. Philip Thompson and Librarian are reference channels, not clients. The figures below describe their public videos, not The Spy Show.

Both channels tell spy stories, and one changes the picture roughly six times as often as the other. Neither is wrong. That is exactly why the brief's "don't copy them" matters: a series cannot be both at once, so it has to decide which choices to take from which reference. The full numbers and method are in how often you should cut in a YouTube video.

What was taken from each reference?

From Philip Thompson, the rhythm. The series follows the main reference's pace: frequent picture changes and short shots, much closer to four seconds than to thirty. One detail from his video is worth knowing even if you never make a spy documentary: the pace is a curve, not a flat rate. His cold open runs at about 19 changes a minute and the closing at about 8.2, so the fastest cutting lands where viewers decide whether to stay.

On music, a different choice. Philip Thompson keeps music to the open and the close. The editing review asked for something different: music through the whole video, not only the open and close. So the series now runs a music bed from start to finish that changes with the emotion of each section. On sound, that puts it closer to Librarian, whose music runs the whole video and changes character as the story moves, while the picture keeps the main reference's rhythm.

What was left behind. Librarian's 30-second stills joined by two-second dissolves work because a steady voice at about 157 words a minute carries everything. That is a different product from the one the brief described, so its picture pacing was not taken.

What rules does the picture follow?

These are the rules applied to every episode of the series. They are written down so that episode ten looks like episode one.

1. Show what the narration says in that second. Not only who the story is about, but the place, the time of day and the object named. A line set at night gets a night image. A line that names a place shows that place.
2. When a person's real photos run out, change the kind of shot. Instead of repeating the same face, cut to a period detail: hands, a document, a radio, a suitcase. The viewer stays with the person without seeing the same portrait again.
3. No stand-ins. Recreation footage never shows a stranger's face playing a real person, and it never carries a name label.
4. Period accuracy. No modern objects, and no flags from the wrong era. An image with the right license and the wrong year gets rejected.
5. Real documents with their real dates. Archival documents and newspapers go on screen as what they are, dated correctly, so a headline shown for a moment in the story is a headline from that time.
6. Animated maps for journeys and locations. When the story moves from one city to another, the viewer sees the route instead of hearing it.
7. A license and source log for every asset. Each file goes into the log with its source page and license as it enters the timeline.

The first two rules do most of the work in a spy story. Narration tends to describe rooms, nights, journeys and objects far more often than it describes a face, and archives of secret work are thin. Matching the picture to the words in each second, and switching to period details when portraits run out, is what keeps a long episode from feeling like the same three photos on a loop. The general version of this workflow, from script to upload, is in how a faceless YouTube video gets edited. For claims, strikes and licenses on archival material, see archival footage and copyright in YouTube documentaries.

How does the music work across an episode?

The music bed runs through the whole video and changes with the emotion of each section. When the story turns, the cue turns with it, so the music tells the viewer how a section should feel without competing with the voice. On the measured channels, emphasis comes from changing the cue, not from pushing the bed louder under the narration.

That follows what the measured channels show. The two references handle music in opposite ways, one keeping it to the open and close and the other running a bed about 31 dB under the voice for the whole video, and neither puts music close to the narration. The measurements and how to set a bed in Premiere Pro or DaVinci Resolve are in how loud background music should be under a voiceover.

How are episodes reviewed?

Each episode goes to review before it is final. Revisions run as many rounds as needed, and each revised cut comes back within 24 hours at no extra cost.

A note on one moment is treated as a rule, not a one-off fix. The music request is the example: it was not about one scene, so it changed how every episode is scored. The same goes for picture notes. If an image in one scene does not show what the line says, the fix is to check every other place in the episode where the same thing could happen, and to carry the rule into the next episode.

What can other documentary channels take from this?

1. Name your references, and say what "match" means. "Not copy, but match the level" is a strong brief once someone turns it into numbers.
2. Measure before you cut. Picture changes a minute, shot length, how shots join, text on screen and where the music sits. Two spy channels can differ sixfold on cut rate.
3. Pick choices, not channels. Rhythm can come from one reference and the music approach from another.
4. Plan the picture for each second, not each topic. The place, the time of day and the object named matter as much as who the story is about.
5. Have a plan for when photos run out. Period details beat repeating a face, and a stranger's face never stands in for a real person.
6. Keep the license log from the first episode. A series multiplies every sourcing shortcut.

For a finished example of the same approach on a published video, see the case study of The Interpreter's Tenerife 1977 deep dive. Verified client reviews are on the reviews page.

What does this case study not claim?

The series is in production, so this page names no episode, person or topic, and it gives no views, retention figures or publication dates. It does not link a channel and quotes no client review. Every number on it comes from the two reference channels, which are not clients, and describes their public videos, not The Spy Show. This is a record of how the series is being edited, not a promise of results.

Sources

Project details (the client, the two named references, the "match the level, don't copy" instruction, the picture and music rules, and the review request for music through the whole video) come from the series brief and the editing work on it. Reference figures come from the frame-by-frame pacing measurements and the loudness measurements, taken in August and September 2026 on Philip Thompson's "The Unbreakable Spy of WW2" (full video) and Librarian's "The Real James Bond" (0:00–30:34). Exact numbers move a little with the method, which is why some are given as ranges.