Evidence · first-party tested/creatify-ai

The delivery lacked an informal creator-style tone and personalization, so it reads more like a polished spot than a believable UGC testimonial.

⚠ Struggled🧾 artifact-verifiedinput + output shownTest date not recordedSynthesia
What was measured
Script quality & persuasiveness

Whether the script has a strong hook, clear value proposition, natural CTA, and sounds like UGC rather than a brochure.

decisive for this rankingtransformation

The whole point of a UGC-style ad is a hook, value prop, and CTA that feel native to social ads, so script quality is central to the job. (3 of 3 judges)

What was given, what came back

Test input: Nike Pegasus 41 – Physical Product · text
Input — what we sent
The exact prompt
Product URL: https://www.nike.com/t/pegasus-41-mens-road-running-shoes-LMhrRG/FD2722-002

Exact Script
Hook: These are my new favorite running shoes.
Script: I've been wearing the Nike Pegasus 41 for my daily runs, and they've been incredibly comfortable from day one. They're lightweight, well-cushioned, and great for everyday training. If you're looking for dependable running shoes, they're definitely worth considering.

Output Requirements
- Vertical (9:16)
- 20–30 seconds
- UGC-style AI presenter
- Natural AI voice
- Captions enabled

A short-form UGC-style testimonial ad prompt for a physical consumer product, designed to test product presentation, believable endorsement, and ad generation quality for e-commerce footwear.

Why this input is hard
  • · physical product understanding
  • · UGC testimonial quality
  • · product presentation
  • · avatar realism
  • · voice quality
  • · commercial ad readiness
Provenance
Observation
ca8cb322-0ec5-4dea-abb8-0094bdde69b3
Evidence run
f91a51ec-b400-46fe-97f5-7e779fe17532
Study
Generate UGC-Style Video Ads With AI Avatars
Research task
86baxegpg
Tested at
not recorded
Source
first-party
Evidence state
verified
Proof shown
input + output shown
Cost / latency
not captured
Repeat run
not captured
Tester
not captured

The last three rows are honest blanks, not placeholders — our capture has no field for them yet.

Query this
get_evidence({
  tool: "synthesia"
})
MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 5 other tools
measured on Script quality & persuasiveness
From the same study (page rebuilt from a later run)
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com