Video Editor

I start by ingesting and organising the footage before I let myself get creative: renaming clips, syncing audio, and building bins by scene or interview subject so nothing gets lost later. Then I do a rough assembly, working out the order of the story without worrying about polish, just to see if the structure holds together. Once the rough cut feels right I share it internally or with the client to confirm direction before I invest time in finishing work. From there I refine pacing, tighten transitions, and add lower thirds or graphics as needed. Colour and audio come near the end, once picture lock is agreed, because grading and mixing on a timeline that is still changing wastes time. I export a review copy first, then the final delivery files once everyone has signed off.

I watch the full recording once without touching anything, just taking timestamped notes on the moments that carry real energy or a clear point. Then I pull those selects into a new sequence and read them back to back as text before I even look at the footage again, because a highlight reel lives or dies on whether the words alone tell a story. I build around one throughline rather than trying to include every good line, which usually means cutting soundbites I personally liked but that pull the piece off course. For an event recording I also watch for reaction shots and ambient sound that can bridge two unrelated moments so the cut does not feel like a list of clips. I keep a rough draft under the target length from the start, since trimming down from something already tight is much easier than cutting from something bloated.

I keep an eye on release notes for Premiere Pro and DaVinci Resolve every update cycle, because small feature changes, like improvements to proxy workflows or new colour tools, often change how I approach a project. I also follow a handful of working editors on YouTube and in editing communities where people post breakdowns of trending formats, which is usually faster than reading a feature list. Beyond the tools themselves, I pay attention to what is working on the platforms I cut for: caption styles, pacing, the length of a hook on TikTok versus a YouTube video. I try things out on personal or low-stakes projects first, since testing a new plugin or workflow on a paid deliverable with a deadline is the wrong moment to learn it.

The brief is the foundation, so I read it properly before I touch any footage, including brand guidelines on tone, pacing, and colour palette if they exist. Within those boundaries I still look for a moment or a transition that lifts the piece, and I will usually build one version that follows the brief closely and flag a second option if I think a small departure truly serves the goal better. I present that as a suggestion with a reason attached, not as the only version, because the client knows their audience and their brand better than I do in a single meeting. Where I hold firm is on anything that would confuse the story or hurt watchability, since a beautifully on-brand video that nobody finishes watching has not done its job either.

On a corporate recruitment video, I had feedback coming in from HR, the marketing team, and two department heads, and each round contradicted parts of the last one: HR wanted more employee voices, marketing wanted tighter pacing, one department head wanted their team featured more heavily. I kept a single tracked document of every note with who made it, so I could spot conflicts early instead of discovering them mid-edit. Where notes conflicted, I went back to the original brief and the video's stated goal, attracting graduate applicants, and used that to decide which note actually served the piece. I also grouped small wording and graphic changes into one pass instead of re-exporting after every comment, which kept the team from watching six slightly different cuts and losing track of what had actually changed. The final video kept its structure intact and every stakeholder could see their note had been considered even where it was not used.

A client moved a product launch date forward by a week partway through a video project, and asked for three additional short-form cuts for social on top of the original long-form piece. I went back to them with a realistic list of what was and was not achievable in the new window, rather than agreeing to everything and quietly cutting quality later. We agreed to prioritise the long-form video and one strong short-form cut for launch day, with the remaining two delivered a few days after. To hit the new deadline I simplified some of the motion graphics I had planned, reused colour grading presets from an earlier project instead of building new looks, and worked from a locked shot list rather than experimenting with alternate cuts. The client appreciated the early honesty more than they would have appreciated a rushed set of five mediocre videos.

A creative director wanted to cut a full ten seconds from an emotional testimonial in a charity video to hit a tighter runtime target, and I felt the cut removed the pause that made the moment land. Instead of pushing back in the room, I exported both versions, the requested cut and my alternative with the pause restored, and sent them side by side with a short note explaining what I thought each version gained or lost. I made clear I would go with whichever the team preferred and that the runtime target still mattered to me too. They watched both and chose to keep the pause, trimming the runtime elsewhere instead. Having two real versions to compare made the disagreement about the footage rather than about opinions, which is a much easier conversation to have with a client.

I set up a consistent folder structure and naming convention before ingest, since chasing down a missing clip or an unclear file name costs far more time later than it takes to organise upfront. For team projects I work off shared storage or a cloud platform like Frame.io or Premiere Productions so everyone edits against the same source files rather than duplicating media locally, and I use proxies for anything shot in a high resolution or heavy codec so the timeline stays responsive. Project files get versioned with a date and short description in the name, and I keep at least one backup off the main working drive, ideally in a different location, because a single drive failure should never be able to lose a client's footage. On collaborative jobs I also make sure only one person has a sequence open for editing at a time to avoid two people saving over each other's work.

For colour I start with a corrective pass, matching white balance and exposure across shots so cuts between angles or cameras do not jump, and only after that do I move into a creative grade that sets the mood or matches brand colours. I reference a still or a look-up table early so I am not guessing at the tone shot by shot. For audio I start with dialogue: cleaning noise, levelling volume between clips, and applying light EQ so voices sit clearly in the mix. Music comes in under that, ducked automatically or manually so it never competes with speech, and I add sound design touches like whooshes or ambient bed tracks only once the dialogue and music balance feels right. Before final export I check loudness against broadcast or platform standards, since a video that plays too quiet or clips on a phone speaker undoes a lot of the polish elsewhere.

For short-form platforms like TikTok, Reels, or Shorts, I cut in 9:16, plan the first two or three seconds as a genuine hook since most viewers decide whether to keep watching almost immediately, and keep cuts fast with minimal dead air. Captions are burned in and styled to be readable at a glance, since a large share of that audience watches with sound off. For long-form YouTube content in 16:9, I can let a scene breathe a little longer, build toward a payoff instead of leading with it, and rely on pacing that rewards a viewer who has already chosen to sit with the video. Captions there are usually delivered as a separate file rather than burned in, so YouTube's own caption system and chapter markers can do the work. When I am cutting the same source material for both, I treat them as two separate edits rather than one edit resized, because what makes a cut work changes with the format, not just the frame.