Most brands lose their creative learning constantly. An ad works, it fatigues, it is switched off, and six months later somebody briefs something similar without knowing the original existed. This guide is for whoever briefs creative. It covers what the library records, how to read it, and how to turn it into the next brief.
What is recorded
Every creative you have run, described by more than its file name.
The hook is the opening line or first few seconds. The angle is the persuasive approach: problem and agitate, relatable, social proof, authority, comparison. The format is the production style: user generated content, studio, static, carousel. Alongside those sit spend, return on ad spend, click through rate and hook rate, plus a status that summarises where the creative stands.
Recording angle and format separately from the hook is what makes the library useful. Knowing a specific video worked tells you to run it again until it fatigues. Knowing that problem and agitate hooks in user generated format consistently outperform studio comparisons tells you what to make next, which is worth considerably more.
Read patterns, not individual winners
The instinct is to sort by return on ad spend and look at the top row. That finds the best ad. It does not find what to do next.
Group by angle instead and look at the aggregate. If four of your five best performers share an angle, that angle is doing the work rather than any individual execution, and your next brief should start there. If your winners are spread evenly across angles, the variance is in execution and the useful question is what the individual hooks have in common.
Do the same by format. Many brands find one format carries almost everything, and it is frequently the cheapest to produce, which changes the economics of the whole creative programme.
Weight by spend
An ad with high return on ad spend and very little spend behind it is not evidence. It is a small sample that happened to land well.
Read the library with spend in mind, the same way the scale, wait or kill verdict does. A creative that sustained performance across substantial spend has proven something. One that looked excellent for two hundred dollars has not, and briefing more of it is acting on noise.
Fatigued is not failed
The status field distinguishes creative that never worked from creative that worked and stopped.
That distinction matters for what you do next. A creative that never worked should not be repeated. A fatigued creative earned its results and exhausted its audience, which means the angle is proven and the execution is spent. Those are excellent candidates for a new hook on the same angle, which is far cheaper than starting from nothing and more likely to work.
Fatigued creative also frequently recovers after a gap, because the audience turns over. Retiring something permanently that simply needs a rest is a common and avoidable loss.
Turning it into a brief
A good brief from this library names the angle, the format and the hook style that the evidence supports, and leaves the specific execution to whoever makes it.
Over specifying kills the thing that produces winners, which is variation. Under specifying wastes the library entirely. Naming the angle and format while leaving the execution open is the balance that uses what you know without repeating what you already ran.
What to check first
Sort by spend rather than by return on ad spend, and look at what your ten highest spending creatives have in common. That is where your money has actually gone, and whether those share an angle tells you immediately whether you have been iterating on a proven approach or producing unrelated work each time.