Howdy Storytellers 🤠,

First, the recording of Movie Machine Madness week 2 is coming later today, but before we share the recording, I wanted to give you a special peek into our VIP Production Process.

Most production companies treat their process like a trade secret. We don't.

Below is the full playbook our VIP Production team uses on every project — six phases, refined over dozens of productions and thousands of hours of AI filmmaking.

The craft moves. The pricing math. The mistakes we catch so you never have to.

If you've ever wondered what you're actually paying for when you hire our VIP Production team, this is the unfiltered version.

Note: On making your movie yourself v.s. working with our VIP Production Team. Many of you are making your films on your own using the Movie Machine. We’re so grateful for you – however, the third party video and image models the Movie Machine uses are nascent. They don’t always give you the output you want, which requires many regenerations to get your desired output with consistency, continuity, and realism. You can spend this time (and money) on your own, or work with our VIP Production Team, where trained AI Directors make your film, saving you 10+ hours per finished minute and often saving you tons of money and stress too. In short, with where the technology is at right now, we highly recommend most folks serious about making movies with the Movie Machine work with our VIP Production Team. Let me share more on that and how we do what we do 👇

🎬 VIP Production at a glance

$250 per finished minute standard rate. $200 per finished minute during the Movie Machine 2.0 launch while early slots last

20% down to reserve your slot · 100% satisfaction guarantee (or your money back)

You own everything — film, characters, voices, IP. Forever.

Any length — 30-second trailer, 10-minute pilot, full feature. We've done all of them.

[coming soon] The Six-Figure Film Formula: an exclusive paperback book on how to market and distribute your film distributed exclusively to VIP Production Clients

No pressure. We'll tell you honestly whether VIP is the right move, or whether you'd be better off in the Movie Machine yourself.

Now, let's pull back the curtain.

Phase 1 — Story Intake & Creative Alignment

We read your story like filmmakers, not editors.

You hand us your book, script, prose, or treatment. Before we touch a single piece of technology, we sit with it. We're not looking for what's beautiful on the page. We're looking for what's filmable: the moments that need a close-up to land, the scene that only works on a wide, the line that has to hit a specific beat or the whole sequence falls apart.

What we actually do here:

  • Map your story's tone, emotion, characters, worldbuilding, pacing, and style — and pressure-test each one against what AI can credibly render right now

  • Identify the core conflict that drives your arc and the retention beats that will keep a viewer watching past minute one, minute three, and minute ten

  • Locate your comparable titles in the market — what's working in your genre, what audiences are searching for, where the demand-vs-supply gap actually is

  • Recommend a format and length that matches your story (sometimes that's a feature, sometimes a 10-minute pilot, sometimes a vertical short series — we don't push you into the wrong shape for our convenience)

The trap most projects fall into here: Writers hand over a story that reads beautifully and assume the AI will translate the prose into a film. The AI translates exactly what you give it. Vague prose makes vague films. Specific prose — a character's exact wardrobe, a room's exact light, a prop's exact placement — makes specific films.

You leave Phase 1 with: A locked script and aligned expectations on tone, style, format, and length.

Phase 2 — Director's Notes (Your Visual Bible)

This is the most important step in all of AI filmmaking. When this is done right, everything downstream gets smoother, cheaper, and more consistent. When it's skipped or rushed, you spend the next three weeks fighting your own film.

Continuity is the single hardest problem in AI filmmaking. Characters morph between shots. Wardrobes drift. Props vanish and reappear in different hands. The living room your character was sitting in ten seconds ago is suddenly a different living room. The eye color of your villain shifts from hazel to bright green halfway through a monologue because someone described them as "piercing" in the prompt.

So we build a visual bible. Before a single start frame gets generated, we lock down:

  • Character references — face-forward headshots for every character. Full-body shots actually reduce facial fidelity because the face takes up fewer pixels, so we're deliberate about how each character is captured.

  • Wardrobe consistency — every outfit a character wears, named and versioned, with notes on when in the story they change clothes and why

  • Setting palettes — the actual color, lighting, and feel of every location. Interiors with object placement (where the window is, where the couch sits) so the room stays the same room across cuts.

  • Props and recurring objects — the gun in scene 4 has to be the same gun in scene 11. The locket she wears in the opening has to be the locket she clutches in the climax.

  • Voice profiles — how each character sounds, with intonation notes and language for the audio model to lock onto

For continuity across multiple projects (sequels, episodic shows, expanded universes), we can import and merge characters across projects so the same character looks identical whether they appear in your pilot or your season finale.

The cost asymmetry that makes this phase non-negotiable:

  • Wrong wardrobe caught in Director's Notes → nearly free

  • Caught in a start frame → costs a regen (a few dollars)

  • Caught in a finished video shot → costs the original generation, the regen, and the rebuild of every shot in the scene around it (this could be hundreds of dollars)

The difference between a $200 film in generation costs and a $2,000 film in generation costs is almost always upstream. We solve continuity here, on purpose. Because of our dialed-in process, this is why our VIP Production Team can guarantee you a low-rate.

You leave Phase 2 with: A complete visual bible. Every character, prop, location, and voice locked. You approve every element before we render anything.

Phase 3 — Storyboarding (Your Film's Skeleton)

The Movie Machine generates a first-pass storyboard from your script. We tear into it.

A film is won or lost on the shot list before a single frame is generated. So before we render — before we spend a single coin on video — we walk through every shot, scene by scene.

What we actually do here:

  • Delete weak shots — anything redundant, slow, or unmotivated. Every unnecessary 6-second shot is roughly $2+ of generation cost. A 200-shot film with 30 unnecessary shots is $60 of pure waste before you count the time spent reviewing them.

  • Group shots into scenes with clear emotional shape — every scene has a beginning, a turn, and an end

  • Map emotion across the cut — where does the audience tense up, where do they breathe, where do they laugh

  • Adjust shot durations — a reaction shot can be 2 seconds. A driving shot can be 3. We right-size every shot instead of defaulting to 6 across the board.

  • Check the 180-degree rule on every cut — cross it and your characters flip across the frame. The scene feels wrong even when viewers can't say why. This is one of the most common failure modes in amateur AI films and one of the easiest to prevent at the storyboard stage.

  • Vary camera angles — variety is what keeps an AI film from feeling repetitive. Too many shots back-to-back with the same angle is the tell of an inexperienced editor.

  • Plan B-roll coverage — reaction shots, establishing shots, environmental cutaways. B-roll is how we smooth jump cuts, cover continuity issues, and give a film its rhythm. We plan it before we render, not as a panic fix when something goes wrong.

We walk through every shot asking a specific list of questions, every time: Who's in it? What's the framing? Is the action consistent with the shot before? Are the props right? Is the lighting consistent? Where are the characters looking? Did we cross the line?

If we don't ask these questions here, the AI doesn't either.

The genre lens we bring to this phase: A vertical short series paces differently than a feature. A romantasy book trailer paces differently than a thriller pilot. We tune the cut to match your genre's conventions and your audience's attention pattern — not just to "what looks cinematic."

You leave Phase 3 with: A locked, scene-by-scene shot list. Every shot motivated. Every duration right-sized. B-roll planned. Pacing mapped to genre.

Phase 4 — Start Frames (The Visual DNA of Each Scene)

Start frames are the literal first frame of every video shot in your film. If the start frame is wrong, the video is wrong. Full stop.

This is where most amateur AI films quietly fall apart. You can see beautiful start frames in someone's demo reel and still watch the finished film and feel that something is off — because the start frames were beautiful but didn't account for what had to happen next.

This is also where we slow down on purpose.

What we actually do here:

  • Generate the start frame for every shot and verify it against your visual bible — character face, wardrobe, props, background, lighting all consistent

  • Verify pre-action — every start frame has to be the moment before the action begins. If your character will draw a gun, the frame can't show them already holding it. If they'll speak, their mouth has to start closed. If they'll walk into a room, they have to be at the threshold, not mid-stride.

  • Check eye direction — viewers track where characters look. Two people in dialogue who aren't looking at each other reads as wrong instantly. A character staring at the camera in a scene that isn't meant to break the fourth wall ruins immersion in one frame.

  • Verify cohesion between adjacent shots — does shot 4 visually follow shot 3, or does it feel like a different scene? Are we maintaining the 180-degree line we set in the storyboard?

  • Use Edit for small fixes, Regenerate for major ones — this distinction matters enormously. Edit preserves the existing image and nudges it (different camera angle, slight position change, swap a prop). Regenerate creates a brand-new image from the characters, locations, and props in your Director's Notes — which means consistency can reset. We use Edit for the large majority of fixes because it's faster and preserves continuity with surrounding shots.

  • Fight style drift — if "photorealistic" starts trending toward "plasticky" or "animated," we re-prompt aggressively and remove the words that triggered the drift. Sometimes a single word in the prompt is the entire problem.

  • Use reference images for high-continuity shots — when a shot has to match a previous shot precisely, we feed the previous frame in as a reference so the model anchors to it. Sometimes we even take the end frame from one shot and copy it over as the start frame to the next shot. This is incredibly easy inside the Movie Machine.

If a start frame needs more than one regeneration to get right, we log it. Every flag is feedback we use to improve our Director's Notes upstream — because a start frame that keeps failing is almost always a symptom of an unclear visual bible, not a model problem.

You leave Phase 4 with: Every start frame approved. Every continuity issue caught before it gets locked into expensive video generation.

Phase 5 — Video Generation & Editing

Now we render. This is the most expensive step in the pipeline — which is exactly why we don't skip the previous four.

Model selection per shot:

  • Cinematic shots go to models like Wan, Veo, or Seeddance for B-roll, establishing shots, and atmospheric coverage — anything that doesn't need synced dialogue

  • Action shots go to models like Kling, Wan, Veo, or Seeddance for fights, chases, anything physical and fast where motion fidelity matters more than lip sync

  • Dialogue goes to models like WAN for talking heads, with audio generated in sync with mouth movement

Picking the wrong model for the wrong shot is the single most common waste of coins we see in AI films.

Then we refine. Every shot, every time:

  • Trim awkward frames — if a 6-second shot is great for 4 seconds, we cut at 4. Pacing is the difference between cinema and slideshow.

  • Regenerate weak shots economically — never more than 3 attempts on the same shot without some major troubleshooting. If three regens don't fix it, the issue isn't the model, it's the prompt or the Director's Notes. Brute force never works; targeted feedback does.

  • Cover problems with B-roll instead of forcing a fix that may never come. If 3 seconds of a 6-second shot are great and 3 seconds are weak, we cut, cover with a reaction beat, and keep moving.

  • Regenerate audio independently of video when intonation is off — we don't pay to re-render the whole shot when we just need a different read on a line. WAN with short dialogue under 3 seconds sometimes loops the line; we add a deliberate pause and double the dialogue to fix it. Small tricks, big savings.

  • Use lip-sync models (like Sync Reacts) on dialogue shots where the AI's native sync isn't quite landing

  • Polish with a lead editor — sound mixing, transitions, the last 5% that elevates a film from good to great. This typically happens in DaVinci Resolve or CapCut for advanced sound design and color work – increasingly the editor inside the Movie Machine is getting more and more powerful for these kinds of workflows.

  • Use transitions only when they mean something — a fade to mark time passing, a match cut for impact, a hard cut for shock. Never on every edit. Transitions used carelessly are the loudest tell of an amateur edit.

  • Score original music through tools like Suno when the project warrants it — we have a production account specifically for client work

One finished minute of film typically takes us a full day of work at this stage. Not because the rendering is slow, but because the refining, regenerating, and editing-down of raw footage into something that actually flows is where craft lives.

In VIP Production you don’t pay for regenerations or 6 seconds of dead space. You pay only for film on the final timeline that you are satisfied with.

You leave Phase 5 with: A polished, scored, sound-mixed draft film.

Phase 6 — Client Review & Final Delivery

We deliver a draft film for your review.

Then we open a structured revision window — typically 7-10 days, depending on length. You give us specific notes: shot-level, scene-level, music, pacing, dialogue, anything you want to change. This isn't a vague "I don't love it" loop that drags forever. We work through your notes in batches, deliver revised cuts, and converge on a final.

For most projects we resolve in 1-2 revision rounds. We're transparent about scope upfront so you know what's included.

Then we deliver the final master cut. Polished. Sound mixed. Ready to publish.

You own 100% of everything.

  • The film itself — distribute it wherever you want, monetize it however you want

  • The characters, voices, and visual assets — you can use them in sequels, prequels, spinoffs, merchandise, or any future project

  • Full IP rights — your story, your characters, your business

We don't keep distribution rights. We don't take a cut of your revenue (you can use Creatorwood’s Streaming Platform where you keep 80% of gross revenue or any other platform… more on that in a later email). We don't ask for a producer credit unless you want to offer one. The film is yours.

Optional add-ons available at this stage:

  • Localization — translation, dubbing, or subtitling into other languages and markets

  • Trailer cut — a 30-90 second promotional cut of your finished film

  • VIP Marketing – the Creatorwood Team can help you market your film, but more on that in a future email

You leave Phase 6 with: Your final master film, full ownership, and the option to come back for more whenever you're ready to scale.

Why we share all of this

Two reasons.

One: Storytellers deserve to know exactly what they're paying for with our VIP Production Service. It’s a ton of work and an intensive process that has taken a large dedicated team of folks thousands of hours to master – and we’re getting better every single day.

Two: This work is too hard and too important to gatekeep. Even if you never hire us, we'd rather you take this playbook and make a better film yourself than watch you ship something that breaks at minute three because nobody told you Director's Notes was the most important step. As time goes on and we learn more, we aim to keep sharing more and more publicly, right here.

We can make your film with you, for you, or alongside you. Whichever serves your story best.

Ready to make your film with our VIP Production Service?

On the call we'll walk through your story, recommend a length and style, and tell you honestly whether VIP is the right move.

Let us know any questions, whether you are making your film on your own or exploring doing it with our team. So grateful to have you all here.

And as always don’t forget…

Storytellers Rule the World,

Michael and the Creatorwood Team

P.S. A few of the films we’re already producing under VIP — feature horror film with one of the biggest horror authors working today, sci-fi trailers, romance microdramas, animated children’s cartoons. Different genres, different lengths, same six phases every time. The process is what makes the work reliable. The process is what we just handed you 😁 .

Keep Reading