I turned product photos into UGC ads with Kling image-to-video — here are 3 workflows that actually work

UGC ads used to mean hiring creators, shipping products, and waiting weeks. Now the whole thing can start from one product photo: image-to-video turns the photo into a 5–10 second clip that looks like a real user made it — if you prompt it right.

I build these from the same production system behind 30+ shipped AI shorts. The three rules that matter most:

1. First second sells. The hook lands before the pitch. A slow fade or a logo first, and the scroll wins.
2. One real photo anchor. The product photo is the anchor. Describe the product the same way every time, or it drifts between clips and trust dies.
3. Text stays out of the frame. Hooks, on-screen text and CTAs are added in your editor, never generated in frame.

Below are 3 complete workflows from my new pack (20 workflows total). Replace the [brackets] and run them.

Workflow 1: Selfie testimonial

• Goal: trust in the first second with a real-person recommendation.
• Input: 1 photo — creator holding the product, natural light, casual home background.
• Kling settings (starting point): image-to-video · 9:16 · 5s.
• Prompt: Front-camera selfie style, [creator] holds [product] up beside their face, talks to camera with a relaxed smile, natural window light, casual home background, slight handheld feel, no text
• Camera: static front camera, tiny handheld sway, face and product both sharp.
• Hook (first second, in post): "I stopped buying the expensive one after trying this."
• On-screen text (in post): "[Product] — 30 days later"
• CTA (end card, in post): "Link in bio — try it yourself."

Workflow 2: Desk unboxing

• Goal: anticipation — the box opening is the ad.
• Input: 1 photo — sealed product box on a clean desk, top-down.
• Kling settings (starting point): image-to-video · 9:16 · 5s.
• Prompt: Top-down desk shot, two hands slowly unbox [product], soft light, clean background, product centered, texture close-ups, no text
• Camera: top-down, slow and steady, close-up on the reveal beat.
• Hook (in post): "Unboxing the thing all over my feed."
• On-screen text (in post): "First look"
• CTA (in post): "Link in bio for yours."

Workflow 3: Gadget demo

• Goal: function first — show it working, not just looking good.
• Input: 1 photo — gadget on a desk, screen visible.
• Kling settings (starting point): image-to-video · 9:16 · 10s.
• Prompt: [gadget] on a dark desk, screen lights up, a hand taps and swipes through the main feature, rim light tracing edges, tech feel, no text
• Camera: three-quarter angle, slow orbit feel, hold on the screen beat.
• Hook (in post): "It does this in one tap."
• On-screen text (in post): "One tap. Done."
• CTA (in post): "Link in bio to try it."

The full pack

These 3 are a sample. The full pack has 20 workflows in 7 groups — testimonials, unboxing, before/after, GRWM/beauty, product demos, problem/solution, seasonal/promo — each with goal, input photos, settings, full prompt, camera notes, hook, on-screen text, CTA and a variation, in Markdown + PDF + JSON.

• Full pack ($9.99): https://zzstudio.itch.io/kling-ugc-ad-prompt-pack
• The complete prompt system behind it ($4.99): https://zzstudio.itch.io/ai-video-prompt-pack-vol1
• Free starter (9 templates): https://zzstudio.itch.io/ai-video-prompt-starter-pack
• Free methodology chapter: https://github.com/zz583/ai-video-prompt-pack

Unofficial guide, not affiliated with Kuaishou or Kling AI. Settings are starting points; results vary by model version and input photo quality. What ad format should the next volume cover?