How to Switch Video Tools Without Pausing Client Work
Migrations fail when they're planned as a cutover. A phased, week-by-week path for moving a video agency between tools - what to move, what to leave, and what to tell clients.
The most common reason agencies stay on a tool stack they have outgrown is not price, and it is not features. It is that nobody can see a safe week to move. Clients are mid-campaign, editors are mid-cut, and the person who would run the migration is the person who is already busiest.
So the decision gets deferred to a quieter month that never arrives. If you have said “let's revisit this after Q4” more than once, this article is for you.
The Mistake: Treating It as a Cutover
Almost every migration that goes badly was planned as an event - a weekend where everything moves and everyone starts fresh on Monday. That framing creates the problem it is trying to avoid:
- It needs a quiet week, and production agencies do not have quiet weeks.
- It moves every client at once, so every client experiences your learning curve simultaneously.
- It forces a decision on years of archive footage before anyone has confirmed the new tool is right.
- It has no rollback. If Monday goes badly, there is no half-step back.
The teams that switch smoothly do the opposite. They treat it as a gradual expansion that starts with one project and never has a big-bang day.
The Phased Path
Week 1 - One project, one client, real work
Pick a live project with a client who is patient and a deliverable that is not on fire. Not a test project with fake files - a real one, because test projects never surface the things that actually break.
Everything else stays exactly where it is. Nothing is cancelled. You are running two systems on purpose, and the cost of that overlap is the cheapest insurance in the whole process.
Week 2 - Add the people, not just the projects
Bring in two or three editors and one supervisor. The goal is the review loop: a cut goes up, notes come back, a version is approved. If that loop feels better than what you have, the rest of the migration is bookkeeping. If it does not, you have learned that for the price of one project instead of your whole business.
Weeks 3-4 - Expand brand by brand
Move whole brands, not scattered projects. A brand is the unit a client experiences, so moving one completely means that client has one place to go rather than two. Half-migrated clients are the single biggest source of confusion, and they are entirely avoidable.
Week 5+ - Retire the old tools one at a time
Cancel in the order of least risk. Usually the PM tool goes first (nothing is stored in it), then the scheduler, then the review tool, and the file storage last - or never, if it is also your archive.
What to Move, and What to Leave
The biggest source of migration paralysis is footage volume. A team with 40TB of archive assumes migration means moving 40TB. It almost never does.
Split your media into three buckets:
- Active production. Anything in flight or likely to be revisited in the next quarter. This moves. It is usually a small fraction of the total.
- Finished masters. Delivered work you may need to re-cut or re-license. Move these as you touch them, not upfront.
- Cold archive. Raw camera dumps from projects closed two years ago. Leave them. Keep them on the NAS, on LucidLink, or in the storage you already pay for, and link rather than upload.
Teams that keep a NAS or LucidLink for cold archive and use the new platform for everything in active production are not doing a workaround. That is a normal, supported end state.
Six Things to Decide Before You Start
- Who owns it. One named person, with real hours protected for it. A migration owned by “the team” is owned by nobody.
- What your stages are called. Do this before you import anything. Renaming statuses after 200 tasks exist is significantly more annoying than agreeing on them beforehand.
- Who sees money. Decide which roles see client price and margin before you enter any rates, not after an editor sees a number they should not have.
- What clients are told. One short message - here is your new link, you do not need an account, here is where to click approve. Send it per brand as you move them.
- What “done” means. Write the success criterion down at the start. “All active brands running, old PM tool cancelled” is checkable; “fully migrated” is not.
- Your rollback. Which tools stay paid, and until when. Deciding this upfront is what lets people commit to the pilot without fear.
The Client Conversation
Agencies dread this one and it is almost always a non-event - if you pick a tool where clients do not need an account. The message is short:
We've moved reviews to a new system. Same link-and-click as before - open the link, watch, leave notes or hit approve. No signup, no password. Everything waiting on you is on one page.
If a tool requires your client to create an account, that message gets much longer and your approval times get worse for a month. Weigh that heavily when choosing.
Signs It Is Going Well
- Editors stop asking where the latest version is - the answer became obvious.
- Client approvals arrive without someone chasing them by hand.
- Nobody has opened the old PM tool in a week, and nobody noticed.
- A new project gets created in the new system by default, without anyone deciding to.
That last one is the real finish line. A migration is done when the new system is the path of least resistance, not when the last file has moved.
