How Long Should Videographers Keep Raw Footage After a Project?

I delete raw footage from finished client shoots on a fixed schedule, right, and I know that sounds like heresy, because the standard answer to how long should videographers keep raw footage is keep everything forever since storage is cheap, and to be very honest that answer quietly bleeds money out of a business every month while you feel responsible about it.

I run a video editing agency and a personal branding agency, and we built PlayPause so creative teams could send cuts out for client review, so I see the tail end of a lot of projects, right, and the pattern barely changes. The drives fill up, nobody wants to be the person who deleted the wedding or the founder interview, and two years later you are paying to store terabytes of media nobody has opened since the day it was delivered.

So here is my short answer before the long one. Full card dumps get ninety days after final approval for most work, twelve months for weddings, and less for social content. The selects, meaning the takes that made the cut plus the obvious alternates, get a year, or two for weddings. The approved master, the project file and the record of who approved what get kept basically forever, because they are tiny and they are what clients actually come back for, and once you separate the light stuff from the heavy stuff the storage problem gets really really small.

How long to keep raw footage, and what videographers actually do

Read the forum threads where people argue about this and you mostly find habits rather than policies, right. Some people format the cards the moment the final file is approved, others keep every frame forever because one time a client came back after two years and they got to be the hero, and a surprising number have no real idea what they are keeping until a drive fails and takes three jobs with it.

The sensible middle, and the one I settled into after a lot of arguing with myself, is a window tied to the type of project. A wedding couple might genuinely come back for an anniversary edit, so their raw media earns a longer hold, while a corporate explainer that went live almost never gets reopened, and social content for a retainer client is stale within weeks, so holding its card dumps for years is basically paying rent on a room nobody will visit.

The catch here is that none of this works if it lives in your head, right, because the moment a client emails eighteen months later asking for the b roll, whatever you feel that day becomes the policy, so it has to be written into the one document the client actually signed.

Writing a video footage retention policy into your contract

A retention clause is three sentences, and it belongs next to your revision terms and delivery scope, which for most of us means the statement of work attached to the proposal. Mine reads roughly like this: "Raw camera media is held for ninety days after final approval. After that date it is permanently deleted and cannot be recovered unless extended archive has been purchased. Approved deliverables and the record of approvals are kept indefinitely." You change the number of days per project type, and the shape stays the same.

That third sentence is the one that saves you, because most of the panicked emails arriving a year later turn out to be asking for proof rather than footage, right. The client wants to know which cut went live, who signed off on the music, or whether that legal line was in MV3 or MV4, and if you kept the review trail you can answer in under a minute. I went through the mechanics of that in a piece on keeping the approval trail intact when a project closes.

A written window also gives you something to sell, because once there is a stated ninety day period, extended archive becomes an optional yearly fee, and in my experience clients either pay it without fuss or tell you plainly they do not need it. Closing a project, including deciding where its files go, is a recognised stage in any project management process, and at the end of the day archiving is just that closing stage done honestly.

What to keep: masters, projects and raw footage

Here is the split I use, and it decides everything downstream, right. The permanent tier is the approved master, a lighter web copy, the project file with its linked assets list, the licences for music and stock, and the review history. The medium tier is the selects, kept for the window in the checklist above. The heavy tier is everything else, every unused card dump, every safety take, every four hour multicam angle of a room where nothing happened, and so on, kept for the contracted window and then deleted on purpose.

The permanent tier is small, a handful of gigabytes for a corporate job and maybe twenty for a wedding, while the heavy tier for those same two jobs can easily run to six hundred gigabytes and two terabytes, so does that make sense, right, the decision you make on the day you finish is which tier each folder belongs in, and keeping or binning the whole project never really comes into it.

For instance, on a two camera corporate interview I keep the graded master, the web copy, the project file and the interview selects, while six hours of rolling b roll from the second camera goes into the heavy tier with a delete date. Whether the client gets the raw files at all is a separate question, and I laid out my position in a post about handing raw footage to clients, because if raws are a paid add on, your retention window gets a very clear commercial shape.

Storage math for a year of client work

Let me do the arithmetic out loud, because people avoid this topic mostly because they have never run the numbers. Say you shoot two projects a month and each generates four hundred gigabytes of camera media, which is conservative for multicam, so that is twenty four projects and about 9.6 terabytes a year. If the selects average sixty gigabytes a job, that is roughly 1.5 terabytes worth holding for twelve months, and the permanent tier at maybe five gigabytes a job comes to about 120 gigabytes for the whole year.

So the heavy tier is a little over eight terabytes, roughly eighty five percent of everything you shot, and it is the part almost nobody asks for again. That is illustrative math from the shape of my own work rather than a benchmark, but run it on your last twelve months and I'm pretty sure the ratio looks similar, which tells you to put the small permanent tier on fast reachable storage and the heavy one on cheap cold drives with a delete date.

Now line that up against the plans, which are priced flat per workspace rather than per member. On my example numbers, a year of permanent tier at around 120 gigabytes sits comfortably inside the 250 gigabytes on Agency, the 100 gigabytes on Creator would fill before the year ended, and the terabyte on Enterprise holds several years of finished work. You can compare the tiers on the PlayPause pricing page, and there is more on how workspace storage behaves on the media storage page.

Downgrading never deletes your content either, which removes the fear of moving up during a heavy quarter. If you are a one person shop shooting one project a month, the numbers land differently, and the page on PlayPause for freelance editors is the better starting point.

The archive pass I run at the end of every project

I do this in the same week the final file goes out, right, because the knowledge of what mattered evaporates fast, and in three months you will not remember which card folder had the good drone pass.

The step people skip is the last one, and it matters most because it makes the policy visible every time you open the drive. For instance, a folder called 2026 03 ClientName Wedding RAW DELETE 2027 03 tells you what it is and when it goes in one glance, without a spreadsheet or your memory getting involved.

Inside PlayPause I use a custom status for the same job, something like Archived with the raw delete month next to it, so anybody on the team can see which projects are closed, and nobody stacks a new version onto a card that was finished months ago.

Backing up finished projects to Google Drive

On the Agency plan and above there is a Google Drive connection, and it does two useful things for this exact problem. Viewers can save a file straight into their own Drive, which ends the era of uploading a master again because somebody lost the email, and there is a backup option, so the finished work can sit in a second place you control.

That second copy matters more than people think, right, because an archive living in one system only is a single point of failure wearing a nice interface. I compared the two common transfer options in WeTransfer against Google Drive for video delivery, which is worth a read if your handover still goes out as a zip file.

I went deeper on treating the archive as a system rather than a junk drawer in a piece about building a media archive that scales, because plain, consistent naming is what makes a twelve month old project findable at all.

Links and access after the project closes

The last piece of a retention policy is access, which people forget entirely, and then a link shared in March is still live in November on a device nobody controls. In PlayPause share links behave differently by plan, right. On Creator, links run for thirty days and the files expire with them, which is perfect for active review and the wrong home for anything you need next year. On Agency, links run for ninety days, and when a client comes back later you send a fresh link. On Enterprise, links never expire, which is the right setting when you are the permanent home for a brand's library.

You can revoke a link instantly and password protect the ones going to a wider group. I wrote about the timing side in a post on making a video link expire once the client has seen the cut, and about the limits of passwords on their own in whether video password protection is enough for client work.

Who watched analytics on Creator and up show who opened the link, when, and from which city, so before you close a project you can check that the final cut was actually viewed by the person meant to sign it off. Trust me on any level, that beats a hopeful assumption, because it stops you archiving a project that comes back with fresh notes three weeks after the raw media went to a cold drive, you see what I mean here.

Frequently asked questions

How long should I keep raw footage for a wedding?

Longer than corporate work, right. I hold the full card dumps for twelve months, the selects for two years, and the approved master indefinitely. Weddings are the one category where a late request is genuinely likely, because anniversaries and family events bring people back years afterwards, so offer extended archive as a small yearly fee rather than absorbing the cost, and put the end date in writing.

Does deleting raw footage put me at legal risk?

I am not a lawyer, so get your own terms reviewed properly, but in my experience the trouble comes from vague habits rather than from deleting. If your contract states a delete date and the client agreed to it in writing, deleting on that date is simply doing what you said you would. The messier situation is holding media indefinitely with no agreement, because then the client assumes you will always have it.

What should I keep permanently after a project ends?

The approved master, a lighter web copy, the project file with its linked assets and licences, and the record of what was approved and by whom. That whole set is usually a few gigabytes per job, which is nothing next to the raw media, and when a client comes back a year later it answers the question most of the time without a single card dump being touched.

Does downgrading my plan delete my archived projects?

No, downgrading never deletes content in PlayPause, so if you move up to Enterprise for a heavy quarter and then come back down, your projects and comment history stay where they are. What changes are the limits going forward, so plan your tier around your steady state rather than your busiest month.

Where should the heavy raw media actually live?

On cold storage that is separate from your working drive, with a second copy somewhere else, and the delete date written into the folder name so it stays visible every time you look. The review platform is for the permanent tier that clients and your team actually reach for, and keeping those two jobs apart stops your working setup filling up with material nobody opens.

If you want to try this shape of archive without rebuilding anything, put one finished project into a PlayPause workspace, masters and versions and the full comment history, and see whether you can answer a twelve month old question from it in under a minute. Every plan comes with a 7-day free trial, so pick the tier that matches your storage reality on the PlayPause pricing page and run one real project through it before you commit.

So yeah. That's my way of saying it.