You pay for creator footage by the deliverable. You use it by the second. Nobody compares the two.
Here's the audit I'd run once a quarter. It takes about an hour if your footage is in a searchable library, and a weekend if it isn't.
Step 1: minutes delivered, per creator
List every creator from the quarter and the total minutes they delivered. Raw footage, not edited clips.
If your footage is in BuildYourAds, this is the creator filter plus the duration column. If it's in Drive, it's a spreadsheet and a lot of right-click → properties.
Step 2: seconds used, per creator
Go through the ads you actually ran. For each one, note which creator's footage is in it and roughly how many seconds. An editor can do this from memory for the last month; for a full quarter you'll need the project files.
Let's say creator A delivered 40 minutes and 25 seconds ran. Creator B delivered 12 minutes and 90 seconds ran. B is the better creator for your brand, by a lot, even though A sent more.
Step 3: usable moments, per creator
This is the number the first two steps miss. Delivered minutes are not usable minutes. A creator who sends 40 minutes of self-talk and retakes is different from one who sends 12 minutes with 8 clean hooks in them.
With a segmented library: filter by creator, count the segments by type. Hooks, testimonials, demos, b-roll. That's the real inventory.
every hook from Maria
every testimonial line from Maria that mentions results
Without one: watch it. This is the step that makes the audit take a weekend, and it's the step most brands skip, which is why they never learn which creators are worth rebooking.
Usable moments per creator: count the segments by type instead of watching everything.
Step 4: the gap
Now you have three numbers per creator. The interesting column is usable minus used.
Quick Example:
| Creator | Delivered | Usable moments | Used in ads | What to do |
|---|---|---|---|---|
| B | 12 min | 8 hooks, 5 testimonial lines | 90 s | rebook; 10+ moments still unused |
| A | 40 min | 3 moments | 25 s | don't rebook; keep the 3 |
| C | 15 min | 11 moments | 0 s | nobody opened the folder; this is free money |
Creator C is the one this audit exists for. Good footage, paid for, never opened. Every brand I've looked at has at least one.
Step 5: turn the gap into the next test round
Every usable-but-unused moment is a variation you don't have to shoot. Put them in the next brief by link (see brief your editor with links).
In my experience this is the cheapest test round a brand ever runs, because the footage is already paid for and already good.
What should I change after the audit?
Three things, and they compound:
→ Rebook by usable moments per delivered minute, not by follower count or price. → Give creators the list of moments you actually used. They'll send more of those next time. → Stop paying for "3 videos" and start paying for "10 minutes of raw footage with these 8 moments in it". The brief gets specific and the footage gets usable.
Here's the line I'd put in the next creator brief:
Send raw footage, not an edit. We need, in your own words: the problem before the product (30s), your first reaction to it, three different ways of saying the main benefit, and a close-up of you using it. Say the lines more than once if you stumble. We cut.
The audit spreadsheet
If you're doing it by hand, these are the columns. Nothing else:
Creator | Delivered (min) | Usable moments (count) | Used (seconds) | Usable minus used | Rebook? (yes/no/why)
Fill it once per quarter. Sort by the second-to-last column. The top of that list is your next test round; the bottom is who not to rebook.
What this doesn't tell you
Why a moment worked. The audit shows which creators produce usable moments and which moments you haven't run. It doesn't know that the "skeptical" hook won because of the pause before the reveal. That's still you watching the ad and the results together.
It also can't judge footage quality beyond "usable or not". A tagged moment is a candidate. Whether it's a good candidate is a strategist's call.
Your first two weeks
Week one: run steps 1-3 on last quarter. If the footage is in a library, it's an hour. Write down the creator C's, the ones with usable moments and zero usage. There's usually more than one.
Week two: brief one test round from those moments only. No shoot. Then send your top three creators the list of moments you actually used from their footage, and the brief line above. Watch what the next delivery looks like.
FAQ
What counts as a usable moment?
A hook, a testimonial line, a demo, a b-roll action or a reaction that could go in an ad as is, with at most a trim. Retakes, self-talk and setup footage don't count, which is why counting by hand is slow and a segmented library is fast.
How often should I run this audit?
Quarterly for the rebooking decision, and a quick version (step 4 only) before every test round so the unused moments get used before new ones get commissioned.
Can BuildYourAds do this automatically?
The creator filter, the segment counts by type and the duration columns are there today. The "used in an ad" column is the part that still needs your project files; connecting ad performance back to the segments is on the roadmap.
What if a creator delivered a lot but nothing usable?
That's an answer too. Don't rebook, but look at the brief you sent them first. "Make a video about the product" produces 40 minutes of nothing. The brief line above produces moments.
Should I tell creators I'm measuring this?
Tell them what you used and why. Creators want to know which of their moments ran. Most will send more of exactly that, which makes the next audit shorter.
How do I count seconds used across many ads?
Roughly is fine. The decision is rebook or not, and the gap between creator A (25 s from 40 min) and creator B (90 s from 12 min) doesn't need to be precise to be obvious.