Pond5 Contributor Guide: Stock Footage Metadata That Sells
How Pond5's contributor-set pricing works and how to write footage metadata that matches how video editors actually search. Practical guide for 2026.
# Pond5 Contributor Guide: Stock Footage Metadata That Sells
Pond5 is one of the few large marketplaces where contributors have meaningful control over pricing on much of their catalogue. That is genuinely unusual, and it attracts contributors who want to position their work rather than accept a fixed rate.
But pricing only matters once a buyer has found the clip. For footage, being found depends on metadata that most contributors write as though they were tagging a photograph.
## Footage metadata has a dimension photos do not
A still image has a subject, a setting and a mood. A clip has all of those plus **time and camera behaviour**, and video editors search on both.
An editor building a sequence is rarely thinking "I need a video of a city." They are thinking "I need a slow aerial push over a city at dusk, 4K, at least eight seconds." That is a shot specification, and it is what they type.
So footage metadata needs to cover, in addition to the usual subject fields:
**What happens across the clip.** Not just the subject but the action and how it develops. A person sitting still and a person standing up and walking out of frame are different clips with the same subject.
**Camera movement.** Static, handheld, pan, tilt, tracking, dolly, crane, gimbal, aerial, drone orbit. This is one of the most-used filters and one of the most-omitted keyword categories.
**Shot type and framing.** Wide, medium, close-up, extreme close-up, over-the-shoulder, top-down, POV.
**Speed and treatment.** Slow motion, real time, time-lapse, hyperlapse, and the frame rate it was captured at.
**Technical specification.** Resolution, frame rate, duration, whether it is log or graded, whether there is an alpha channel.
Contributors who come to video from stills routinely write excellent subject metadata and omit every one of the movement and framing terms. The clip is then invisible to the exact search that would have licensed it.
## Technical fields are a filter, not a footnote
Editors filter before they browse. If your resolution or frame rate is not declared, or the clip is below 4K in a market where 4K is the working baseline, you are removed from the result set before the visuals are ever judged.
Upload the highest quality master you have, and declare the specification accurately. This is not metadata polish — it is qualification for the search.
## Pricing when you control it
Because Pond5 lets you set prices on much of your catalogue, you have a decision most agencies make for you. Two failure modes:
**Pricing everything at the floor.** Competing purely on price against a marketplace of millions of clips is not a strategy, and it caps the value of genuinely distinctive footage.
**Pricing everything high.** Generic B-roll priced like specialist work simply does not license.
The sensible approach is differentiated: price ordinary coverage competitively and price genuinely hard-to-obtain footage — difficult locations, expensive setups, rare events, specialist subjects — according to what it actually cost to get. The thing that makes this work is honest self-assessment about which of your clips are which.
## Video versus stills economics
Per licence, footage earns substantially more than stills across most agencies. Volume is lower and production cost per asset is much higher — a clip takes longer to shoot, longer to review, and vastly longer to upload than a photograph.
That changes the workflow calculus. With stills, the standard advice is volume, and metadata automation is what makes volume survivable. With footage, each asset carries more value, so the cost of a clip failing to surface because of thin metadata is proportionally much higher.
Our guide to [what Adobe Stock pays per download](/blog/how-much-does-adobe-stock-pay-per-download) covers the stills side of this comparison in detail.
## Handling footage across multiple agencies
Video metadata is more work per asset than stills, and agencies want it in different shapes. Some expect movement and shot type as keywords; others have dedicated fields. Resolution and frame rate declarations differ. Category systems differ.
Doing that by hand for every destination is where footage contributors lose their time — and unlike stills, you cannot compensate with volume. Rastock AI [generates per-agency metadata for video as well as images and delivers over FTP](/features), which keeps the technical and movement fields consistent across destinations. Plans by upload volume are on the [pricing page](/pricing).
## Summary
On Pond5, pricing control is the headline feature, but it only pays once your clips reach the right buyer. Write metadata the way an editor searches: subject *plus* movement, framing, speed and specification. Declare technical fields accurately, because they filter you in or out before anything else is considered. Then price differentially rather than uniformly.
This guide is part of our overview of [where to sell stock photos](https://rastock.ai/blog/where-to-sell-stock-photos), which compares every major agency on royalties, review standards and AI policy.
For the full picture, see our guide to the [stock photography workflow](https://rastock.ai/blog/stock-photography-workflow) — every stage from selection through metadata to delivery.
## Related reading
- [Footage Metadata Consistency: Tagging Video Series and Collections](https://rastock.ai/blog/storyblocks-consistent-footage-metadata)
- [Storyblocks Contributor Guide: How the Subscription Model Changes Strategy](https://rastock.ai/blog/storyblocks-video-collection-ftp)
- [VectorStock Contributor Guide: EPS, JPG and Vector Metadata](https://rastock.ai/blog/vectorstock-metadata-for-illustrators)
Frequently asked questions
How does Pond5 pricing work for contributors?
Pond5 is known for letting contributors set their own prices on much of their catalogue, rather than fixing them centrally. That gives you control over positioning, but it also means pricing becomes a decision you have to actively make rather than accept.
What metadata do stock video clips need?
Everything a photo needs plus the temporal and technical dimension: what happens across the clip, camera movement, shot type, duration, frame rate and resolution. Editors filter on those technical fields before they ever look at the visuals.
How do video editors search for stock footage?
Usually by shot specification as much as subject — aerial, slow motion, handheld, tracking shot, close-up, time-lapse — combined with the subject and a technical filter for 4K or frame rate. Metadata that only describes the subject misses most of that.
Does stock footage earn more than stock photos?
Per licence, generally yes — clips license at substantially higher prices than still images across most agencies. Volume is lower and production cost per asset is higher, so the trade-off is fewer, more valuable sales.
What resolution should I upload to Pond5?
Upload the highest quality master you have. 4K is the practical baseline for commercial work in 2026, and editors frequently filter for it. Delivering below it narrows your buyer pool before your content is even evaluated.
How many keywords should stock footage have?
Enough to cover subject, action, setting, mood, shot type and camera movement accurately. Because editors search on shot specification, omitting movement and framing terms is the most common gap in footage metadata.