All posts
Case StudyJuly 6, 20267 min readLumineer Editorial
🛒

AI video for ecommerce — what's actually working in 2026

For DTC brands, AI video stopped being a novelty in early 2026. Here is what the winning teams are actually shipping.

01

The three shots every product page now needs

A 3-second orbit loop, a 5-second lifestyle hero, and a 6-second in-use micro-demo. Photography still shoots the flatlays; AI increasingly shoots the motion. Three brands we tracked all converged on this same trio independently.

02

The orbit loop

Generated in Veo 3.1 image-to-video from a clean packshot. Prompt: 'slow clockwise orbit, 360 degrees, subject centred, studio white, subtle rim-light'. Cost per shot: around $0.30. Replaces roughly $200 of turntable rig time.

03

The lifestyle hero

Generated by first producing a hero frame in Flux 2 Pro or Nano Banana 2 that composes the product into a lifestyle setting, then animating in Kling 2.5 or Sora 2. This is the shot that carries brand mood on collection pages.

04

The in-use micro-demo

Sora 2 with native audio wins here. A 6-second clip of the product being used, with light foley (zip sound, click, poured liquid) — dramatically more engaging than the same shot silent.

05

The numbers we saw

Across three brands over sixty days:

  • PDP add-to-cart lift: +6% to +14% vs static hero
  • Reel/TikTok CTR: +22% average vs static-image ads
  • Cost per completed video asset: $2-$6 (vs $200-$800 outsourced)
  • Production time per asset: 15-40 minutes (vs 3-10 days)
06

The traps

Two: over-generating (twenty variations of the same shot when three would ship better), and skipping the frame-first step (pure text-to-video produces off-brand colour and composition drift). Both fixable with a stricter brief template.

ecommerceDTCcase study
Try it now

Build your ecommerce stack

Put this into practice in the studio — under a minute to your first result.

Build your ecommerce stack →

Keep reading