Split-screen chaos-to-clean transformation. Oldest trick in DR advertising because it still converts.
IMAGE 1 — BEFORE (GPT Image, vertical 9:16): [BEFORE SCENE — e.g. "a desk buried in crumpled receipts and tangled cables"], photorealistic, shot from a fixed tripod at chest height, 35mm feel, harsh greenish overhead fluorescent light, desaturated muted grade, slightly underexposed, stressful clutter but every object physically plausible. No text, no people, no warped objects. IMAGE 2 — AFTER: the same exact desk, same camera height, angle, lens and framing — nothing moved but the state: surface immaculate, [PRODUCT — e.g. "the FileFox app open on a laptop"] as the visual centerpiece, warm 9am window light from the left, one coffee cup, calm minimal styling, gently lifted warm grade. Repeat the camera language word-for-word from Image 1 so the two frames register perfectly. ASSEMBLY (CapCut): 0.0–1.5s HOOK — hold the BEFORE full-frame with a faint fluorescent buzz; 1.5–2.0s — hard vertical wipe left-to-right with a whip-whoosh sound revealing the AFTER; 2.0–5.0s — hold the AFTER over a soft ambient chord; 5.0–6.0s — text overlay only now: "me before [PRODUCT] / me after". One transition, no zooms, no extra effects.