# UGC ad creative

> Claude writes 30 ad-script variations, ElevenLabs voices them, Pika generates the b-roll, CapCut edits and adds captions.

Source: https://aistack.sh/stack/ugc-ads
Category: Marketing · Ads
Level: Intermediate
Last verified by a human: 2026-05-05

## The tools

### Claude — Script variations

- Pricing: Free · $20/mo Pro · $100/mo Max 5x · $200/mo Max 20x · Sonnet API $2/$10 per M tokens (input/output)
- Site: https://claude.ai
- Why this one: Best at ad-copy hooks; can generate 30 variants from 1 brief without all sounding the same.
- Swap in instead: ChatGPT
- More: https://aistack.sh/tool/claude

### ElevenLabs — Voiceover

- Pricing: $6/mo Starter · $22/mo Creator
- Site: https://elevenlabs.io
- Why this one: Pick from the marketplace voices for variety; pay only for characters used.
- More: https://aistack.sh/tool/elevenlabs

### Pika — B-roll

- Pricing: Free · $10/mo Standard
- Site: https://pika.art
- Why this one: Pika's stylized motion is more 'ad-shaped' than Runway's photoreal.
- Swap in instead: Runway
- More: https://aistack.sh/tool/pika

### CapCut — Edit + caption

- Pricing: Free · $9.99/mo Standard · $19.99/mo Pro
- Site: https://capcut.com
- Why this one: Built for vertical ad formats. AI captions are good enough to ship.
- More: https://aistack.sh/tool/capcut

## Monthly cost

### $26/mo — 10 ads/wk (small)

- Claude: $20
- ElevenLabs: $6 Starter
- Pika: $0 free
- CapCut: $0

### $67/mo — 30 ads/wk (medium)

- Claude: $20
- ElevenLabs: $22
- Pika: $10
- CapCut: $15

### $244/mo — Agency · 100 ads/wk (heavy)

- Claude: $80 API
- ElevenLabs: $99
- Pika: $50 Pro
- CapCut: $15

## Workflow

### 1. Brief to 30 hooks (Claude)

Claude turns one product brief into 30 hook variants across 3 angles.

**Prompt: 30 ad hooks across 3 angles**

```
Product brief:
"""
{{paste product brief: what it is, who it's for, what makes it different, 1 customer quote}}
"""

Generate 30 short-form ad hooks for vertical video ads (TikTok, Reels, Shorts), split evenly across 3 angles:

1. **Pain hook** — open with the audience's problem in their words.
2. **Solution hook** — open with the surprising mechanic or result.
3. **Social proof hook** — open with a specific outcome from a named (or anonymized) user.

For each hook:
- Format: "{HOOK LINE} (angle: pain/solution/proof)"
- Hook line: 8 to 14 words, spoken-word friendly, hard consonants up front
- No "Are you tired of", no "Imagine if", no "Stop scrolling"
- One concrete noun or number per hook

Output as a numbered list 1 to 30.
```

### 2. Voice (ElevenLabs)

ElevenLabs marketplace voice per persona. Render at 192kbps.

### 3. B-roll (Pika)

Pika generates 4–6s clips per beat. Stylized > photoreal for ads.

### 4. Cut (CapCut)

CapCut vertical 9:16, captions, music duck.

## What it produced

**DTC supplements brand**

30 AI-UGC ads tested in week 1. 4 winners; CPA dropped 38% vs the agency-shot UGC pack at $5k.

## Pitfalls

- **Voices sound the same across ads** — Force diversity by mapping persona → voice up front; never re-use the same voice across ads in the same creative test.
- **Compliance flags on health/finance** — Meta auto-flags AI-voiced health ads. Disclose 'created with AI' if you're in a regulated vertical.

## Recent changes affecting this stack

- **Scribe v2 Medical speech recognition model released** (2026-09-11, release) — Scribe v2 Medical is now generally available as a specialized batch speech recognition model for medical and clinical audio. It maintains the same billing rate as Scribe v2 and supports keyterm prompting, entity detection, speaker diarization, and no-verbatim mode. Pass scribe_v2_medical as the model_id parameter. Source: https://elevenlabs.io/docs/changelog
- **Agent response attachments in ElevenAgents conversations** (2026-09-07, feature) — agent_response WebSocket events now include an optional attachments field, allowing agents to send files with metadata including URL, filename, and MIME type. Source: https://elevenlabs.io/docs/changelog
- **Twilio outbound calls now support answering machine detection** (2026-09-07, feature) — Outbound calls through Twilio can now detect whether they reach a human or answering machine. Configure detection mode to trigger early detection or wait for voicemail greetings to complete. Results arrive via a new answering_machine_detection webhook event. Source: https://elevenlabs.io/docs/changelog
- **ElevenAgents WebSocket responses now support file attachments** (2026-09-07, feature) — Agent response WebSocket events include optional attachments with URL, name, and optional MIME type. Enables agents to send files and documents to users during conversations. Source: https://elevenlabs.io/docs/changelog
- **Twilio answering machine detection added to ElevenAgents outbound telephony** (2026-09-07, feature) — Outbound telephony configurations now support Twilio answering machine detection with two modes: early human-or-machine verdict or waiting for voicemail greeting to finish. Detection results are delivered via the new answering_machine_detection webhook event. Source: https://elevenlabs.io/docs/changelog

## Other ways to do this

- [UGC ads (CapCut-only)](https://aistack.sh/stack/ugc-ads-budget): Single-tool: Claude writes hooks, CapCut handles voice, b-roll, edit — $24/mo

---

Curated by tone-d on aistack.sh. Last updated 2026-05-05.