The Pre-Launch Creative Scorecard: 5 Checks to Run Before You Spend

Most ad creative gets graded by the media budget. You launch, it runs, and a few days later the platform hands you a verdict you paid for. That's a slow, expensive way to learn something you could have caught at your desk.

This is the cheaper version: a scorecard you run before launch. Five checks, each scored 0–2, for a total out of 10. It won't tell you your conversion rate — nothing can, before launch — but it reliably flags the creative that was never going to work, which is where most wasted spend actually goes.

Print it, paste it into your review doc, or run it as a team. The point is to turn "this feels off" into a number you can act on before the spend, not after.

Why score creative instead of just eyeballing it

A checklist you read is easy to nod along to and forget. A checklist you score forces a decision on each line. Three things change when you put a number on it:

It makes weak spots specific. "The hook is fine" becomes "hook scored 1 of 2 — the opening line describes the product instead of the problem." Now you know exactly what to fix.
It removes the loudest-opinion problem. When two people disagree about an ad, a shared rubric settles it on the creative's merits, not on who's more senior in the room.
It sets a launch bar. A total score gives you a clean rule: above the bar it ships, below it goes back. No more shipping ads because the deadline arrived.

The five checks below are the same teardown a careful performance marketer runs instinctively — written down so anyone on the team can run it the same way every time. For the deeper "why" behind each one, the pre-launch creative testing playbook walks through the full process; this page is the fast, scoreable version.

How to score it

Each check is worth 0, 1, or 2 points:

2 — clears the bar. No changes needed on this dimension.
1 — partial. Recognizable but weak; there's an obvious, usually free fix.
0 — fails. This alone can sink the ad. Fix before launch, full stop.

Add the five checks for a score out of 10, then use the verdict table at the end.

Check 1 — The 1-second test (clarity)

What it tests: whether someone moving at full scroll speed knows what this is and who it's for before they're gone.

How to run it: cover everything except the first frame and the first line. Look for one second. Can you name the product category and the intended audience? If you hesitate, so does the feed.

Scoring:

2 — Instantly clear what it is and who it's for.
1 — One of the two lands; the other takes a beat.
0 — You have to study it to understand it.

Most common fail: a clever visual or a slow-building video that buries the point past the moment of attention. The fix is almost always free — lead with the clearest frame, not the most artful one.

Check 2 — Hook or description? (the opening line)

What it tests: whether the first line grabs a problem the buyer actually feels, or just announces the product.

How to run it: read only the opening line or first caption. Ask: is this hooking a tension the buyer already has, or describing what we sell? "We make X" is a description. "Still doing X the hard way?" is a hook.

Scoring:

2 — Opens on a real, felt problem or a genuine pattern interrupt.
1 — Relevant but flat; states a benefit without creating tension.
0 — Pure product description; assumes the viewer already cares.

Most common fail: "description openers" are the single most common reason good products run weak ads. The fix is usually a free rewrite — lead with the pain, not the feature.

Check 3 — Name the exact buyer (audience fit)

What it tests: whether the creative is built for a specific person, or for a vague "marketers / businesses / everyone" blur.

How to run it: name the exact buyer out loud — role, context, the specific frustration they have at 9am. Then re-read the entire ad as that person. Mark every spot where it stops fitting them.

Scoring:

2 — Clearly written for one specific person; nothing in it feels generic.
1 — Targeted, but a few lines drift toward "anyone."
0 — Could be for almost any audience; no one feels personally addressed.

Most common fail: writing to a segment instead of a person. The places the ad "stops fitting" your named buyer are your real weaknesses — and they're free to fix now, expensive to discover in-market. If you want to pressure-test fit against more than one buyer, an AI persona generator lets you read the same creative through several synthetic buyers and see where they diverge.

Check 4 — The "so what" + the proof (value and believability)

What it tests: whether the value is obvious in plain language, and whether there's any reason to believe the claim.

How to run it: state the payoff for the buyer in one sentence ("so what do I get?"). Then look for the proof — a number, a demo, a specific before/after, a credible source. A bold claim with no proof is a coin flip you're paying media to run.

Scoring:

2 — Value is obvious and backed by at least one concrete proof point.
1 — Value is clear but unproven, or proof exists but the value is fuzzy.
0 — Vague benefit, no proof. "Better results, faster" with nothing behind it.

Most common fail: claims that sound strong but give the viewer nothing to believe. Specific, mundane proof usually beats a bigger, vaguer promise.

Check 5 — One clear next step (the CTA)

What it tests: whether there's exactly one obvious action, and whether it matches where the click actually goes.

How to run it: count the calls to action. Then check that the CTA matches the destination — if the ad says "shop the sale," the click should land on the sale, not the homepage.

Scoring:

2 — Exactly one clear CTA, and it matches the landing destination.
1 — One CTA, but it's weak, buried, or slightly mismatched to the page.
0 — Zero CTAs (nowhere to go) or two competing ones (a fork that kills momentum).

Most common fail: the "two CTAs" fork ("Learn more" and "Buy now") or a CTA→landing-page mismatch that breaks the scent and tanks conversion after the click you already paid for.

The verdict: what your score means

| Score (of 10) | Verdict | What to do | |---|---|---| | 8–10 | Launch it. | Strong on every dimension. This is what media budget is for. | | 5–7 | Fix first. | One or two weak spots, almost always free to fix. Patch them, re-score, then launch. | | 0–4 | Rebuild. | A fundamental miss (usually Check 1 or 2). Don't spend behind it — rework the concept. |

The one hard rule: any single 0 sends the ad back regardless of total score. A clarity fail or a missing CTA can't be averaged away — it will sink an otherwise strong ad on its own.

Where this fits in your workflow

The scorecard is a screen, not a guarantee. It's the cheap test you run first, so the expensive in-market test only ever runs on creative that already cleared the bar. Used well, it sits at three points:

Concept review — score the idea before anyone builds the final asset.
Pre-launch gate — score the finished creative before it gets a budget line.
Post-mortem — when an ad underperforms, re-score it; the failing check usually explains the result.

For a fuller treatment of the underlying method, see how to test ad creative before you launch and the deeper question of what AI analysis can and can't tell you about an ad. If you're comparing the tools that automate this kind of review, the AI creative-testing tools compared hub maps the category honestly.

The scorecard is also the manual version of what POPJAM automates: run these checks against synthetic personas and pre-launch creative testing at the point where fixing the creative is still free. You can start with the free AI ad creative tools and see the same five-check logic applied automatically.

FAQ

What is a pre-launch creative scorecard? It's a structured ad creative checklist you run before spending media budget. You score five dimensions of the creative — clarity, hook, audience fit, value/proof, and call to action — from 0 to 2 each, for a total out of 10, then use the total to decide whether to launch, fix, or rebuild.

How is this different from A/B testing? A/B testing compares finished ads in-market with real spend, and it's great for choosing between strong options. The scorecard runs earlier and for free — it catches the weak creative before it ever reaches an A/B test, so you're not paying to discover that an ad was never going to work.

Can I really judge an ad before launch without performance data? You can't predict the exact conversion rate before launch — and you should be suspicious of anything that claims to. What you can do reliably is catch the structural problems (a buried message, a description-as-hook, no proof, a broken CTA) that account for most wasted spend. The scorecard is built to flag those, not to forecast numbers.

Who should run the scorecard? Anyone reviewing creative before it ships — the marketer who made it, a peer reviewer, or the whole team. Because it's a shared rubric, it's especially useful for settling disagreements about an ad on the creative's merits instead of on seniority.

Does a perfect 10 guarantee the ad will perform? No. A 10 means the creative is free of the common pre-launch failure modes — it's earned the right to a media budget. Performance still depends on offer, audience, and market timing. The scorecard's job is to stop you funding the ads that were doomed, not to promise the rest will win.