Workflow

Post Supervising When the Studio Never Stops

For the past five years I have watched a simple post workflow turn into a constantly moving production pipeline.

What started as a couple of videos a week for a growing YouTube channel has become a full post-production operation supporting multiple shows, hundreds of terabytes of footage a year, a streaming platform, social cutdowns, sponsor deliverables, VFX, graphics, subtitles, QC and archive.

The challenge is not just getting one project finished. It is keeping several projects moving at once without letting the whole system collapse into just-in-time delivery.

In a traditional post environment, the workflow is often built around a defined project: one series, one delivery schedule, one finishing path, and a team scaled around that job. In an online-only studio, the work behaves differently. The studio never really stops. One episode is being offlined, another is waiting on VFX, another needs sponsor material inserted, another is in grade, another is being cut down for social, and another is already being checked for release.

To keep a pipeline like that moving, you have to make compromises. You have to switch projects intentionally so the right deliverables hit the right dates. You have to automate as much as you can safely get away with. You have to keep adapting the system as new demands arrive, because if you do not stay on top of it, the workflow can very quickly snowball into panic delivery.

That is the part of post supervision I find most interesting. It is not just knowing the software, or knowing what order the work should happen in. It is knowing where the pressure is building before it becomes visible to everyone else.

Here is how I think about the main stages of that pipeline day to day.

Ingest

Ingest is where the whole system either starts clean or starts accumulating problems.

Most of our filming still happens in production blocks, like any other production company. That gives crew notice, lets production prepare scripts, and allows post to plan for the volume that is coming. But we also produce content year-round as ideas come up, sponsors come in, and different teams need material turned around quickly.

Our narrative shows tend to sit inside those larger filming blocks, while tabletop gameplay, exclusive shows and sponsor-led pieces often get tacked onto the end, or fall outside the blocks altogether. That has forced us to think carefully about naming conventions, storage tiers and what we commit to archive.

We aim to capture the highest-quality raw footage we reasonably can, even though our final deliverables are usually H.265 UHD. That gives us flexibility, but it also means we churn through hundreds of terabytes a year. With a reasonably sized server, media has to keep moving through hot, warm and cold storage so the footage we need is available without keeping everything online forever.

We do not want to keep too much online, but keeping too little creates its own problems. Pulling material back from cold storage takes time, costs money and usually happens when everyone is already under pressure.

For a high-volume, fast-turnaround studio, I think the media cycle has to be treated as a controlled loop rather than a one-way path. The goal is not just to get footage copied. The goal is to make sure every piece of media has a known status, a known location, and a clear reason for staying online or moving into archive.

Editorial

In most post houses, you get one project, hire people around that project, and streamline the assets you need for it: title sequences, graphics, VFX look development, CG assets and finishing requirements.

That is not usually possible with the schedule I have been managing over the past few years. As more shows and different formats run concurrently, offline editors have to move between episodes and projects to keep up with delivery dates.

That could sound like a compromise, but it has actually become one of the healthier parts of the workflow. It gives editors more variety across the year. If we assigned one offline editor to a single series, they could be cutting the same kind of material for six months, or in some cases the whole year. I try to balance the schedule so editors can get ahead on complex episodes, hit delivery dates, and still have a range of different content to sink their teeth into.

The goal is not to keep every editor locked to one show. The goal is to keep the whole release calendar moving while keeping the work creatively alive.

We have also moved more of the social cutdown work back to the offline editors. In the past, that was handled by a dedicated social media editor. Over time we realised that the best person to create the cutdowns is usually the editor who has worked with the footage. They know what nearly made the cut, what landed in the room, and what can survive outside the context of the full episode.

The trick is separating selection from finishing. Editors can mark and plan cutdowns during offline, but the actual cutdowns should wait until online is complete. Otherwise VFX, grade, graphics and sound effects have to be rebuilt or relaid across multiple versions, which doubles the work at the worst possible point in the schedule.

Online

Online is where all the promises made earlier in the workflow become real.

By this point the episode is no longer just an edit. It is a delivery package. Sponsor material has to be correct. YouTube end-screen annotations need to point viewers to the right product, link, video or subscribe action. VFX need to be final or clearly versioned. Graphics need to match the show language. Platform requirements need to be checked. Anything that was casually left as “we’ll fix it later” has now arrived at later.

This is one of the trickiest parts of the workflow to keep efficient, correct and flowing. We constantly have tight deadlines around marketing material, midrolls, merchandise callouts, YouTube end clips, streaming platform prompts, complex edits that need VFX, and graphics work that has to be created, reviewed, revised and layered in before sign-off.

It is a delicate balancing act. You have to watch the whole delivery schedule, not just the episode in front of you. A complex episode that needs VFX, graphics, grade work and sponsor approvals cannot be treated the same as a simple dialogue-driven episode. If those differences are not identified early, online becomes a traffic jam.

For every episode, I want to know: what is missing, who owns it, when it is due, what it blocks, and whether the current version is safe to send up the chain. That sounds obvious, but when several shows are moving at once, those five questions are the difference between a controlled workflow and a pile-up.

Grading and QC

In this kind of environment, colour grading often gets folded into the role of the online editor. For a long time, that made sense for us. The online was where the episode was being finished, so it seemed like the most natural place to grade and QC at the same time.

I am less convinced of that now.

When the end of the pipeline is moving quickly, the person finishing the work is often the worst person to be the only person checking it. You are too close to the decisions. You know why a compromise was made. You remember the problem shot from two versions ago. You are also usually trying to get the episode out the door.

The danger is not just that mistakes slip through. It is that one mistake changes the way people watch the whole piece. Once someone notices a bad annotation, a missed graphic, a soft VFX shot or an inconsistent grade, they stop watching the episode and start looking for more problems. The suspension of disbelief is gone.

That is why QC has to become a layer in the workflow, not just a final pass at the end. Some of it can be automated. Some of it has to be human. But it cannot live entirely in the head of the person who is also trying to finish the episode.

The Schedule Is the Workflow

The more I have worked in this kind of environment, the more I have realised that the schedule is not separate from the workflow. The schedule is the workflow.

It is easy to think of a post schedule as a list of due dates. This episode locks here, this one goes to grade here, this one delivers here. But when you have multiple projects moving at the same time, the schedule becomes a map of dependencies. Every date is connected to something else. If an offline runs late, VFX starts late. If VFX starts late, online gets squeezed. If online gets squeezed, QC becomes rushed. If QC is rushed, mistakes make it to sign-off. If mistakes make it to sign-off, the final delivery becomes reactive instead of controlled.

That is how a workflow starts to break down. Not usually from one huge failure, but from a chain of small delays that nobody catches early enough.

For me, post supervision is largely about spotting those chains before they become obvious. I am constantly trying to work out which project needs attention now, which one can safely wait, which episode is quietly becoming a problem, and which deadline is only achievable if a decision gets made today.

This is also why I do not think a post schedule should only track the edit. The edit is just one part of the system. A useful schedule needs to show offline, online, VFX, graphics, grade, sound, QC, social cutdowns, sponsor requirements, platform requirements and archive status. If those things are tracked separately, the supervisor ends up becoming the only person who can see the whole picture. That might work for a while, but it does not scale.

The best workflow improvements I have made have usually come from making invisible dependencies visible. A missing graphic is not just a missing graphic. It might be blocking online. A late sponsor asset is not just a late sponsor asset. It might affect the final export, the YouTube end screen, the cutdown and the approval process. A piece of media sitting in cold storage is not just archived. It might be unavailable at the exact moment an editor needs to make a last-minute change.

When you can see those dependencies clearly, you can make better compromises. You can decide which episode needs another editor for half a day. You can move a simpler piece forward while a complex one waits on VFX. You can hold off on cutdowns until the online is actually ready. You can protect QC time instead of hoping there will be space for it at the end.

That is the real job. Not making a perfect plan at the start and expecting it to survive. The real job is constantly adjusting the plan while the work is already moving.