# Ad & Landing Page Conversion — Best-Practice Rubric

This native rubric applies established CRO and direct-response principles to **paid ads and landing pages specifically**. Related catalog references are [`heuristics.md`](./heuristics.md), [`emotional-design.md`](./emotional-design.md), and [`visual-hierarchy-layout.md`](./visual-hierarchy-layout.md). An ad has 1–3 seconds to earn attention before the click, and a landing page has one job (convert the click), not many.

Use catalog ID `reference.ad-conversion` as both the **scoring rubric** and **revision checklist** inside `/humble:ads`. Score each category, apply its weight, and sum to `/100`.

---

## 1. Message Match & Hook (Weight: 20%)

Does the first 1–3 seconds earn the next 1–3 seconds? The ad/headline must interrupt the specific
person it targets, and the landing page must **match** what the ad promised — a broken promise
between click and page is the single biggest paid-conversion killer (bounce before the fold
finishes rendering).

| Score | Criteria |
|---|---|
| 18–20 | Hook names the audience's specific situation/objection in the first line; landing headline restates the ad's exact promise within the fold |
| 14–17 | Hook is relevant but generic ("Plan your future together" vs. naming the objection); message match mostly holds |
| 8–13 | Hook is a category statement, not a hook ("Prenup services"); some drift between ad promise and page headline |
| 0–7 | No hook — opens with company name/logo; landing page headline unrelated to what was clicked |

## 2. Value Proposition Clarity (Weight: 15%)

Can a first-time visitor state, in one sentence and without scrolling, what this is, who it's for,
and why it's better than the obvious alternative? Benefit-led beats feature-led — "no surprise
invoice" beats "flat-fee billing."

| Score | Criteria |
|---|---|
| 14–15 | One sentence, above the fold, states the benefit (not the mechanism) and the differentiator vs. the default alternative |
| 10–13 | Value prop present but requires connecting 2+ elements to parse, or leads with mechanism over benefit |
| 5–9 | Value prop buried below the fold or spread across multiple competing statements |
| 0–4 | No stated value proposition — visitor must infer what the product does |

## 3. Single Clear CTA (Weight: 15%)

One primary action, stated as a verb, with friction visible before the click (not after). Every
competing CTA on the same view is a tax on the one that matters.

| Score | Criteria |
|---|---|
| 14–15 | Exactly one primary CTA in view, low-friction verb ("See your plan" > "Submit"), sets expectation for what happens next |
| 10–13 | One primary CTA but generic verb ("Learn more", "Submit") or next-step unclear |
| 5–9 | Two or more CTAs of equal visual weight competing for the click |
| 0–4 | No clear CTA, or CTA copy/placement actively discourages the click |

## 4. Objection Handling & Trust Signals (Weight: 15%)

The visitor's unspoken objection ("is this legit / is this going to cost me / will this make things
weird") is answered before they have to ask. Trust signals are specific, not decorative — a named
credential beats a generic badge.

| Score | Criteria |
|---|---|
| 14–15 | Names the top objection and answers it directly with specifics (a number, a credential, a named process step) |
| 10–13 | Generic trust signals present (logos, star ratings) but the actual objection is never named |
| 5–9 | Trust signals present but vague ("trusted by thousands") with no specifics |
| 0–4 | No objection handling, no trust signals — reads as a claim with nothing behind it |

## 5. Visual Hierarchy for the Format (Weight: 10%)

Ads and landing pages are read, not studied — the eye should land on hook → value prop → CTA in
that order without deliberate effort. See [`visual-hierarchy-layout.md`](visual-hierarchy-layout.md)
and [`gestalt.md`](gestalt.md) for the underlying mechanics.

| Score | Criteria |
|---|---|
| 9–10 | Clear focal point, contrast pulls the eye to the CTA, no competing same-weight elements |
| 6–8 | Mostly clear, 1–2 competing elements dilute the primary path |
| 3–5 | Multiple elements at equal visual weight, no obvious reading order |
| 0–2 | Flat, no hierarchy — CTA has to be searched for |

## 6. Audience Fit (Weight: 10%)

Does the creative speak the specific segment's language, or a generic average of everyone? A
tech-equity segment and a wedding-planning segment do not share a hook, a proof point, or a tone —
scoring this category requires the segment brief, not just the ad in isolation.

| Score | Criteria |
|---|---|
| 9–10 | Hook, proof points and tone are legible as written for this named segment specifically |
| 6–8 | Broadly on-audience but could run unchanged against an adjacent segment |
| 3–5 | Generic — could run against almost any segment in the category |
| 0–2 | Misaligned — speaks past the stated segment's actual concern |

## 7. Emotional Resonance & Persuasion Ethics (Weight: 10%)

Emotional pull earns attention; manufactured urgency or guilt burns trust the moment it's noticed.
See [`emotional-design.md`](emotional-design.md) — the visceral/behavioral/reflective split applies
to ad creative as much as to product UI.

| Score | Criteria |
|---|---|
| 9–10 | Genuine emotional stake (real anxiety, real relief) named honestly; any urgency is factual (a real deadline/limit) |
| 6–8 | Emotional angle present but generic; urgency present without being manipulative |
| 3–5 | Emotion used but shallow (stock-photo warmth with no textual backing) |
| 0–2 | Manufactured urgency/fear/guilt with no factual basis — a dark-pattern signal, cap this category at 2 regardless of the rest |

## 8. Mobile & Platform Fit (Weight: 5%)

Most paid traffic lands on a phone. Headline must survive a 3-second thumb-stop scroll at small
type; the format must fit the platform it's built for (YouTube pre-roll vs. static Display vs.
in-feed).

| Score | Criteria |
|---|---|
| 5 | Legible one-handed at native size, safe zones respected, format matches the stated platform |
| 3–4 | Legible but tight safe zones, or format is a generic export not tuned to the platform |
| 1–2 | Requires zoom to read the hook, or format mismatched to platform |
| 0 | Untestable on mobile as delivered |

---

## Weighted Score Calculation

```
Score = Hook (×0.20) + Value Prop (×0.15) + CTA (×0.15) + Objection/Trust (×0.15)
      + Hierarchy (×0.10) + Audience Fit (×0.10) + Emotional Resonance (×0.10)
      + Mobile/Platform (×0.05)
```

Each category is scored on its own point range above (already weight-scaled); sum directly to `/100`.

| Grade | Score | Meaning |
|---|---|---|
| A | 90–100 | Ready to run — the ≥90 bar `agent-11-ad-generator` iterates to |
| B | 75–89 | Close — one or two categories need a real revision, not a polish pass |
| C | 60–74 | Concept is sound, execution needs another round |
| D | 45–59 | Fundamental gap — usually message match or missing CTA |
| F | 0–44 | Not a viable concept as written |

---

## Output format (used by `agent-10-ad-scorer`)

```markdown
## Ad Score — [concept name / URL]
Date: [ISO 8601]

| Category | Score | Weight | Notes |
|---|---|---|---|
| Hook & Message Match | /20 | 20% | |
| Value Proposition | /15 | 15% | |
| Single CTA | /15 | 15% | |
| Objection Handling & Trust | /15 | 15% | |
| Visual Hierarchy | /10 | 10% | |
| Audience Fit | /10 | 10% | |
| Emotional Resonance | /10 | 10% | |
| Mobile & Platform Fit | /5 | 5% | |
| **Total** | | | **/100** |

**Top fix:** [the single highest-leverage revision — never more than one, never vague]
```

**Any hard block overrides the arithmetic:** a category-7 score of 0–2 (manufactured urgency/fear/guilt
with no factual basis) caps the **total** at 44 regardless of the other seven scores — a
manipulative hook is not something the rest of the ad can average away.