Pre-Production Is the Film

There is a version of this post that is a war story, and I will spare you most of it. The short version: years ago, on a shoot I did not run, I watched a production burn a day and a serious amount of money because nobody had answered one question in advance. What does the hero shot actually need to say? Forty people on set, and the most expensive conversation of the project happened at 6 a.m. next to the craft services table.

I have been militant about pre-production ever since, and after two weeks of watching Olympic broadcast teams execute flawless live coverage of the most unpredictable events on earth, the conviction is worth writing down. Those teams look improvisational. They are the opposite. Every camera position, every contingency, every graphics cue was decided weeks ago. Preparation is what makes room for the lucky accident.

The budget shape most people get wrong

Ask most clients where a film budget goes and they picture the shoot: the crew, the camera, the location. So when money gets tight, they protect the shoot day and squeeze the weeks around it. This is exactly backwards.

Our film budgets are shaped differently. A disproportionate share goes to the phase before anyone books a truck: script iterations, boards, previz, casting depth, location scouts with the actual DP, and table reads where the idea gets to fail cheaply. The shoot itself is short and calm, because the film already exists in everyone's head and on paper. Post goes fast, because we shot a plan instead of shooting coverage and praying.

The paradox clients eventually feel in their bones: the calm set is the cheap set. Chaos photographs as chaos.

What previz is for now

Previsualization has changed more in three years than in the previous twenty. We can now stand up a rough moving version of the entire film in days: blocking, lensing, light direction, edit rhythm, temp sound. Generative tools drafted their way into this phase and genuinely earned their place there.

But the point of previz is widely misunderstood, so let me be precise about what we use it for.

• Killing shots. The most valuable previz outcome is discovering that the drone move everyone loved is a dead four seconds. Better to lose it in a viewport than at golden hour.
• Aligning the client on rhythm. A board can lie about pacing. A previz cannot. Sign-off on a moving sketch prevents the recut wars of March.
• Pressure-testing the idea. If the film does not work as an ugly gray-box version, no anamorphic glass will save it. Craft amplifies. It does not rescue.

What previz is not for is replacing the shoot. The texture of a real face in real light, the take where the actor does something nobody wrote, the weather that turns a good frame into the frame: that is why we still go. We prepare obsessively precisely so the set has slack for the unrepeatable thing.

The client's job in pre-production

One more opinion, offered with love. The best films we have made shared a trait that had nothing to do with us: the client did their deciding early. They fought hard in the script phase, pushed back on boards, and then, once the plan was locked, they let the plan be the plan.

The films we have regretted shared the inverse trait. Quiet in prep, opinionated on set, decisive only in the edit, where every decision costs ten times what it did on paper.

So we now say this out loud at every kickoff, and I will say it here: pre-production is not the paperwork before the film. It is the film, in its cheapest and most changeable form. Spend your courage there. The camera just records what you already decided to be.