Meta Awareness Planner

---
name: meta-awareness-planner
description: Decide whether a Meta awareness campaign is worth running, pick the right performance goal, and size budget against reach and frequency before launch. Use when asked to plan a brand awareness campaign, decide between awareness and sales objectives, size a Meta budget, calculate reach and frequency, set a frequency cap, judge whether awareness fits a client, plan brand recall measurement, or someone says 'should we even run awareness'. For the creative the campaign runs, see meta-ads-creative. For creative at volume, see bulk-ad-variations.
metadata:
  version: 1.0.0
---

# Meta awareness planner

Most awareness budget is wasted before the first impression, because the campaign
should never have run. This plans it in the right order: decide, then pick the goal,
then size the money — and says **no** when no is the right answer.

Saying no is the highest-value thing this skill does.

> **Awareness buys memory, never buyers.** Judge it on reach, frequency and recall.
> Never on ROAS.

## Before you start

| Input | Why |
| --- | --- |
| Client situation — new brand, local shop, high ticket, impulse product | Drives the go/no-go verdict |
| Total budget and campaign length | The reach calculator needs both |
| Audience size | Below 500,000 rules out Ad Recall Lift entirely |
| Market (India / UAE / UK / US) | CPM differs by roughly 25× across these |
| Whether measurement is set up | Rule 6: if you cannot measure it, do not spend on it |

If the client cannot fund weeks of spend and wait, stop here and recommend a Sales
campaign. That is rule 1 and it ends the conversation.

## Step 1 — Go or no-go

Find the situation. The verdict decides whether anything else in this skill applies.

| Client situation | Verdict | Why |
| --- | --- | --- |
| Brand new brand or product | **YES** | The clearest case. Building memory from zero |
| Big brand, large scale | **YES** | Stay top of mind, around launches and seasons |
| New category nobody knows yet | **YES** | People must know it exists before they can want it |
| Local shop or service | **YES** | Nearby walk-ins, with a tight frequency cap |
| Expensive product (high ticket) | **SOMETIMES** | Introduce yourself, then follow up to close |
| Cheap, impulse product | **NO** | Run Sales. Awareness comes free alongside |
| App installs | **NO** | Use App Promotion. It is built for installs |
| "I just want sales this week" | **NO** | Sales campaigns scale that on their own today |

> Rule of thumb: if being remembered is not what wins the sale, awareness is the
> wrong objective.

On a **NO**, say so plainly and name the campaign type to run instead. Do not soften
it into "you could try both".

## Step 2 — Pick the performance goal

Do not default to Reach for everything. That is rule 4.

| If you want this | Use this goal | Frequency |
| --- | --- | --- |
| Reach the most new people, fast | Maximize Reach | 2 / 7 days |
| Drill one message into a small group | Maximize Impressions | 3–4 / week |
| To be remembered (audience 500k+) | Ad Recall Lift | 2–3 / week |
| The video actually watched | ThruPlay | 2–3 / week |
| Cheap, wide exposure (logo flash) | 2-second video plays | 2 / 7 days |
| Local walk-ins near a shop | Maximize Reach + 3–5 km radius | 2–3 / week |

Default for a brand that just wants to be remembered: Maximize Reach or Ad Recall
Lift, with a frequency cap.

**Ad Recall Lift needs an audience of 500,000+.** Below that, recall cannot be
measured — pick a different goal rather than running it blind.

## Step 3 — Size the money

```bash
python3 scripts/reach_calc.py --budget 30000 --days 30 \
    --audience 500000 --frequency 3 --market india
```

Do not compute this by hand. The script carries the market CPM table and the verdict
thresholds, and it is deterministic.

Read the verdict, not just the numbers:

- **Reach very small** — money is spread too thin. Cut audience or raise budget.
- **Reach near 100%** — the audience is too small for this spend. Widen it.
- **Frequency outside 2–4/week** — adjust length or budget until it lands in band.

## Step 4 — Sanity-check against benchmarks

Full table in [references/benchmarks.md](references/benchmarks.md). The three that
catch most bad plans:

| Metric | Target | If you miss it |
| --- | --- | --- |
| ThruPlay rate | 20% benchmark, above 15% strong | Below 5% means weak creative — fix the video, not the budget |
| Cost per person who remembers you | Under ₹85 (≈ $1) | Higher means refresh the creative |
| Frequency | 2–4 per person per week | Known brand lower, new brand higher |

Rule 5: **a weak result is usually a weak creative, not a small budget.**

## Step 5 — Confirm measurement before spend

Rule 6. If recall cannot be measured, the campaign cannot be judged, and it should
not run. Confirm the audience clears 500,000 for a Brand Lift study, or agree in
advance which proxy metric will stand in.

## Output

```
## Verdict
GO / SOMETIMES / NO — one sentence on why, naming the alternative if NO.

## Plan
Goal:            <performance goal>
Frequency:       <n> per person per week
Budget:          Rs <total> over <n> days
Market CPM:      Rs <cpm> per 1,000

## Projection
Unique reach:    <n> people (<n>% of audience)
Total frequency: <n> over the campaign
Cost per person: Rs <n>
Verdict:         <healthy / spread too thin / audience too small>

## Risks
<Any benchmark this plan is likely to miss, and what to change.>
```

## When it breaks

| What you see | What it means | The fix |
| --- | --- | --- |
| Client asks for ROAS on awareness | The objective was mis-sold | Awareness is judged on reach, frequency, recall. Reset the expectation before launch, not after |
| Reach is a tiny % of audience | Budget spread too thin across too large an audience | Narrow the audience or raise the budget |
| Reach near 100% | Audience too small for the spend | Widen targeting, or cut budget |
| Frequency above 4/week | Campaign length too short for the budget | Extend the flight or reduce daily spend |
| Ad Recall Lift unavailable | Audience under 500,000 | Pick a different goal. Do not run it blind |
| CPM far above the market range | Narrow targeting, feed-only placement, or sale season | Broaden targeting, add Reels; sale seasons can triple CPM |
| Reach looks great, nothing else moves | Cheap markets can carry 20–30% fake or bot views | Cheap reach is not always real reach. Check quality metrics |
| Numbers do not match the client's dashboard | About 25% of behaviour is hidden by privacy | Expect the gap. It hurts sales tracking more than awareness |

Never fix a failing campaign by simply adding budget. Diagnose against the benchmark
table first.

## Rules

- **No appetite, no awareness.** If the client cannot fund weeks of spend and wait,
  recommend a Sales campaign instead. Rule 1 overrides enthusiasm.
- **Never judge awareness on ROAS.** Reach, frequency and recall are the scoreboard.
- **Never default to Reach.** Match the goal to the job.
- **Never run Ad Recall Lift below a 500,000 audience.** It cannot be measured.
- **Fix the creative before the budget.** A weak result is usually a weak video.
- **No measurement, no spend.** Set up proof before launch, not after.
- Quote CPM as a range, never a single number. Ranges are 2025–2026 and move.
- If the client situation is not in the go/no-go table, say so and reason from the
  rule of thumb rather than forcing the closest row.

## Related skills

- **meta-ads-creative** — the creative this campaign runs; rule 5 sends weak results there.
- **bulk-ad-variations** — the creative volume an approved campaign will need.
- **creative-diversity-audit** — when results decay mid-flight, check the batch first.
- **offer-builder** — a NO on awareness often means the offer, not the medium.