Corporate Video and Media Asset Management FAQ

An enterprise video content factory is what you get when a company that isn’t in the media business decides to run its video work on purpose instead of by accident.
It comes down to four things: someone whose job it is to own video, a library where people can actually find what has been shot, storage sized to how often people go looking, and a way of working that gets a request from “we need a video” to “it’s live” without twenty emails.
It’s a way of operating rather than something you buy in a box.

Corporate video is everything a company makes to run its business: training, marketing, town halls, customer support, recruiting, product demos.
Broadcasters and studios sell video for a living, so they build their entire company around it and staff accordingly.
Corporate teams often turn out just as much material with a fraction of that support, because nobody planned for the volume they ended up with.
It comes down to four things: someone whose job it is to own video, a library where people can actually find what has been shot, storage sized to how often people go looking, and a way of working that gets a request from “we need a video” to “it’s live” without twenty emails.
It’s a way of operating rather than something you buy in a box.

If more than one team is regularly making video, you already have one, whether anyone calls it that or not.
The tell is usually scale, so count the people who touch video in a year, the freelancers on repeat business, the gear sitting in closets, and the subscriptions buried in the vendor list.
Most companies are surprised by the total, mainly because it has never been on one page before.

Hardly anywhere has a clear answer, and that is the actual problem.
Production usually lives in marketing or communications, storage decisions land with IT, and the budget is scattered across whoever happened to need a video that quarter.
Nobody is accountable for how the whole thing runs, so it never gets fixed at the level where it’s broken.
Naming an owner costs nothing and is usually the highest-value change available.

The moment you notice that digging up old footage takes longer than shooting something new.
Other giveaways: three copies of the same interview in three folders, files parked in someone’s personal account, editors waiting around on downloads, and nobody able to say what was shot last quarter.
Consumer file storage isn’t the villain here, it just was never built to answer questions about what’s inside a video.

Video doesn’t search itself, and a file called FINAL_v3_edit.mov tells you nothing about the two hours inside it.
So material stays findable only through the person who shot it, and only for as long as they remember.
That’s why so much content gets reshot, since it’s genuinely faster than hunting for something you’re not even sure still exists.

Look for two or more teams booking the same production company separately, more than one video platform on the vendor list, and similar content being made for different audiences.
Nothing looks wrong in any single budget, which is exactly why it survives so long.
It only becomes visible when somebody adds up every video-related line across the whole company, and most places have never done that.

A DAM looks after finished files and isn’t usually built to handle video.
A MAM looks after video that’s still being worked on along with everything around it, including the intake/ingest, editing, formatting, and delivery.
The practical difference is that a MAM can search inside a long recording, keep track of which raw footage a finished piece came from, and let people watch and review big files without downloading them.
Plenty of companies buy a DAM first and then discover the question they actually needed answered was “where’s the usable ninety seconds in these four hours,” not “where’s the approved final.”
Cloud MAM platforms such as Mimir are built for that second question.

Metadata is just the description attached to a file: what’s in it, who’s in it, where it was shot, and what you’re allowed to do with it.
Without it, a video library is a pile of files that only its creator can navigate.
A good deal of this can be generated automatically now through transcription and image recognition, but somebody still has to decide what gets recorded and hold people to it.

The price per terabyte you see advertised is only half the bill.
Most providers also charge you every time you pull data back out, and video files are big enough that those retrieval charges often beat the storage line.
The whole point of a library is that people take things out of it, so the more useful your system becomes, the harder that meter runs.
Providers differ enormously here, and some, such as Backblaze B2, charge far less for retrieval than the largest cloud platforms, which is worth modeling before you sign anything.

Give access to specific material for a set period and keep a record of who received it, rather than emailing files around or handing out an open folder link.
Links get forwarded, projects end while accounts stay live, and material keeps circulating long after anyone intended it to.
Access that expires on its own and is tied to a person rather than to whoever happens to be holding the link solves most of this without anyone having to police it.

Don’t build the case on storage savings, because storage is cheap and finance will spot it.
Build it on the content you paid to produce twice because nobody could find the first version.
Add the production time lost to people waiting on file transfers, plus the money two departments spent on the same thing without knowing.
Most companies can put rough numbers on all three within a few weeks, and the total is usually bigger than anyone guessed.

Almost never because the software was bad.
Usually nobody was made responsible for how material gets described, so the tagging holds up for a few months and then falls apart.
Or the system gets installed without anyone changing the habits that route around it.
The other classic is tipping an old archive into the new system without cataloging it first, which just relocates the mess somewhere more expensive.

A systems integrator gets paid to install and connect things, so the answer tends to involve buying things.
A consultant is hired to work out what you need in the first place, which sometimes means buying less, or fixing how people work instead of what they own.
Both are legitimate and you’ll probably want both eventually, just in that order.

Find out what’s actually happening today: who makes video, what they use, where it all ends up, and what the whole thing costs once you add it up.
Agree on what you want the operation to look like in two years, described in business terms rather than product names.
Choose technology last, because the same problem usually has three or four workable architectures and which one fits depends entirely on the first two answers.

Frequently Asked Questions