A UGC asset library for agencies that outlives platforms: six fields and one owner

The supplier you commission through will change. The footage will not, and neither will the terms attached to it. That asymmetry is why this page is about the record rather than the product, and why no ranking appears: the component that decides whether commissioned content is an asset or a liability is one you build, not one you buy.

What to take away

Why no ranking is offered

A ranking would imply these products are substitutes. They are not. A supplier that produces footage, a system that stores it, and a workflow tool that routes approvals solve three different problems, and agencies routinely need all three from different places. What the third of those is actually for, and what it should never be asked to hold, is separated in the guide to campaign management software.

The second reason is your own volume. An agency with a handful of assets a year should use a shared drive with a naming convention; at scale, that choice stops a campaign.

Any list naming a winner answers for one and misleads the other.

The six fields

None of these needs a specialist product. They need to be structured fields rather than a sentence in an attachment, because a term you cannot filter on is a term nobody will check. Payment does not settle any of this, which is a copyright question rather than a contractual style: the United States Copyright Office explains what copyright protects plainly enough to settle most internal arguments.

The fifth field stays blank when a platform handled the booking, and the asset's agreement is the platform's paperwork, not yours.

The guide to creator marketplaces sets out what that changes.

Structured against attached

The most common failure here is not a missing record. It is a record that exists as a PDF in a folder, so nobody can answer a question across the library.

The questions you will actually be asked are all cross cutting. What expires this quarter. What can run in a market you are entering. What features a person on camera. Which assets carry supplier library music. Every one of those is trivial with fields and impossible with attachments, and the difference is an afternoon of setup against a permanent tax.

Keep the attachment too. The agreement is the authority and the fields are the index, which is the ordinary discipline of records management applied to something most agencies treat as creative output.

Who maintains it

An unmaintained rights record is worse than no record, because people rely on it. Name one person, give them time, and give them a monthly job rather than a project.

The monthly job is short. What expired, what expires next quarter, which new assets arrived without complete fields, and which entries have a term that nobody can trace to an agreement. Twenty minutes if the fields are structured, and a lost afternoon if they are not.

The risk of leaving it unnamed is not that the work is skipped. It is that it gets done informally by whoever remembers, and then that person leaves. A rights library whose real state lives in one person's memory has a bus factor of one, and the loss is not visible until somebody asks a question that used to have an answer.

What to standardize and what to leave alone

Standardize the fields, the naming convention, and the location. Those are what make an account transferable and a handover possible, and they cost nothing to agree in advance.

Standardize the delivery specification as well. It removes most delivery failures before they happen, and what belongs in it is set out in the pillar guide to commissioning short form content.

Leave alone the creative folder structure each team has built, unless it actively prevents the fields from being filled. Reorganizing how people find their own working files buys little and consumes goodwill you will want for the fields.

Where a team has built something outside the standard, find out what it does before removing it. Unrecorded workarounds are the visible edge of shadow work, and they usually mark a genuine gap rather than a discipline problem.

Build against buy

For the library itself, a spreadsheet with the six fields, maintained, beats an unmaintained product every time. That is not a general endorsement of spreadsheets; it is an observation that the failure mode here is neglect rather than capability.

The point at which a spreadsheet fails is worth naming in advance: when two people need to edit it at once, or when the asset count passes what one person can eyeball. Put the cheap option in the comparison deliberately, since a comparison with no cheap option in it concludes with a purchase every time.

What the library does not do

It does not tell you whether an asset performed, which is a separate stack with its own traps, described in the guide to measurement tools.

It does not write the specification, it does not source creators, and it does not chase an approval. Each of those is a different purchase with different economics.

Expecting any of them from an asset system is the most common category error here, and it produces a product that does several things adequately and the one job badly.

Common questions

Can the supplier's own platform be our library?

Only if you can export it completely and read the export. The library has to survive the supplier relationship, and a record you can only see inside somebody else's product does not.

How far back should we backfill?

To whatever is still running, plus anything that might run again. Backfilling a dead archive is a project with no beneficiary.

Who should own the fields, creative or operations?

Operations, with creative filling them at delivery. Ownership by the team that produces the assets tends to produce fields that are complete for new work and empty for everything else.

What is the minimum viable version of this?

Six columns, one owner, and a monthly twenty minutes. That is genuinely enough for most agencies, and it is more than most have.