Clarity: what a creative dashboard has to show at hook/body/CTA level

What is missing from ad-level reporting in Ads Manager?

The unit is wrong.

Ads Manager reports on the ad. Spend, CTR, cost per purchase, ROAS, frequency, all attached to a whole asset. That is the correct unit for a media buying decision, because the buyer's question is "does this keep running." It is the wrong unit for a creative decision, because the creative question is "what part of this worked and can I use it again."

When ad A beats ad B, an ad-level report gives you the winner and withholds the reason. Was it the hook that bought the attention? Was it the way the body framed the offer? Was the CTA aligned with the stage, or did a purchase ask land on someone who had not seen a landing page yet? Three failure modes, one number covering all of them.

So the win is unrepeatable. You can duplicate the asset. You cannot rebuild what made it work, because you never located it.

What has to be broken out before a creative number means anything?

Clarity is the dashboard I built for this, and it does three things.

It breaks any ad into hook, body and CTA. It deduces the angle, the pain, the desire, the fear, the ICP and the market-sophistication level the ad was written against. Then it syncs the Meta Ads Manager metrics so every row carries both the decomposition and the outcome.

Each of those earns its place for a different reason.

Hook, body, CTA are three separate jobs inside one asset and they fail independently. A weak hook loses people before the body exists, so the body's quality is unmeasured. A strong hook attached to a mismatched body buys attention and then wastes it, which shows up as good CTR and bad cost per purchase, a pattern that gets misread as a landing page problem more often than any other. A CTA borrowed from the wrong funnel stage loses intent at the last step: a purchase ask on a top-of-funnel ad whose job was only to start a conversation.

Angle, pain, desire, fear, ICP are what makes an ad reusable. The angle field is the one people skip, and it does the most work.

The metrics sync is what stops the whole thing being a library of opinions. Decomposition without spend and outcome on the same row is a taxonomy. Decomposition plus spend, plus what happened, is a dataset you can score.

Why does the angle have to be a stored field rather than a note?

Because the angle is the thing you relaunch.

When an ad wins, most accounts duplicate it and raise the budget. What compounds instead is writing fresh hooks against the angle that won, because the angle is the strategic claim and the hook is one of ten to twenty ways of saying it. Angles outlive the assets that carry them.

If your dataset has no angle field, you cannot do that. You can count assets, and you can rank them, and you will end up with a list of winning ads and no view of which claims your market keeps responding to. Then a winner fatigues and you start again from zero, because the only thing you recorded was the asset.

Pain, desire and fear sit beside the angle for the same reason. Two ads can run the same angle and hit different fears, and they will perform differently in a way that looks random until those are separate columns.

What does market sophistication add to a creative dataset?

It separates a bad claim from a claim at the wrong level.

Eugene Schwartz's five levels are prior art, not my framework. The operational version: high-ticket "boring" products usually sit at level 4 or 5, where the audience has already seen every claim the category makes. At that level the claim cannot carry the ad, so what varies between a winner and a loser is the angle.

Without a sophistication field, a losing ad reads as a bad ad. With one, you can see that a run of losers were all pitched at level 2 into a level 5 market, which is a completely different fix. One says rewrite the copy. The other says the category's default argument is exhausted and everything built on it will keep losing.

ICP does similar work for a different reason. Metrics collapse across audiences, and the same asset on a warm retargeting audience and on cold prospecting is two different rows. Reading them as one is how a good cold ad gets killed for the sins of the warm layer, or how a warm-audience number gets mistaken for a whole account.

What can a dashboard like this not tell you?

Whether the field caused the outcome.

That is the honest limit, and it applies to every creative scoring system I have seen including mine. A structured dataset makes patterns visible earlier and narrows where to look. On small numbers of purchases it cannot separate an angle effect from a seasonality effect, an audience effect, or the fact that the offer changed that month. It also cannot say anything at all about the ads that were killed before launch, which on my process is most of them.

Two things follow. A creative test still needs four to six weeks before the read means anything, and no amount of decomposition shortens that. And a dashboard is evidence that someone can read an account, not evidence that the account will improve.

Clarity is my working tool. A client on the "+ Data Analysis" tier does get access to it, alongside conversion and event tracking audited end to end and a weekly read on what the numbers say. That access is a feature of the deliverable. What you are paying for is somebody telling you every week what the numbers say; the login is how that gets done.