The Photo Shoot Is Optional Now

The problem

Photography shoots are expensive and slow. For a brand producing 71+ campaigns per quarter, not every shot category justifies a full production day — but the assets still have to exist. Product flatlays in particular were eating shoot budget and calendar space that higher-priority work needed.

The idea

I started asking which categories actually needed a photographer in the room and which ones had consistent enough quality standards that AI could handle them. Flatlay product shots — where the frame, lighting logic, and brand standards are well-defined — turned out to be the clearest answer.

I brought the hypothesis to our development team and we built a pipeline together: prompt frameworks, quality review checkpoints, and a process for flagging which SKUs were candidates for AI generation vs. which ones still needed a shoot. The goal was never to replace photography entirely — it was to protect shoot time for the work that genuinely needs it.

Where it landed

80–90% of new flatlay photography is now AI-generated. 100% of DB flatlay photography is AI. 50% of model photography uses AI — either fully generated or as modifications to existing shots. Shoot budget and production days that used to go to these categories now go to hero campaigns, seasonal shoots, and celebrity collaborations.

What it takes to do this well

The technical side is the easy part. The harder part is maintaining brand standards when the camera isn't in the room — building the review process, knowing when something isn't quite right, and making the call to reshoot versus refine. That judgment is what makes this a creative decision, not just a production one.