E-Commerce Product Photography System

The Challenge

A D2C e-commerce brand was spending $15k/month on product photography—hiring photographers, renting studios, and waiting weeks for delivery. They needed 500+ product photos per month across multiple SKUs, with consistent lighting, angles, and brand aesthetic.

The traditional approach wasn't scalable. As they launched new products, the photography bottleneck slowed everything down.

The Solution

We built a Midjourney prompt library that generates production-ready product photos in minutes, not weeks. The system maintains brand consistency across thousands of variations while reducing costs by 90%.

The Prompt Library

Here's an example of our structured prompt system:

/imagine product photography of [PRODUCT], studio lighting setup with --ar 1:1, soft shadows, white background, professional ecommerce style, consistent brand aesthetic, high resolution, Canon EOS R5 look --v 6.0 --style raw

We created 15 core prompt templates covering:

• →Hero product shots (white background)
• →Lifestyle context (in-use scenarios)
• →Detail close-ups (texture, materials)
• →Seasonal variations (holiday themes)

The Code

To automate batch generation, we used the Midjourney API with a custom Node.js script:

import { Midjourney } from 'midjourney-api'

const mj = new Midjourney({
  apiKey: process.env.MIDJOURNEY_API_KEY,
})

async function generateProductPhotos(products: Product[]) {
  const prompts = products.map(p =>
    `product photography of ${p.name}, studio lighting, white background, --ar 1:1 --v 6.0`
  )

  const results = await Promise.all(
    prompts.map(prompt => mj.imagine(prompt))
  )

  return results.map(r => r.imageUrl)
}

Style Consistency

To maintain brand aesthetic, we:

• →Created a style reference library from existing brand photos
• →Used --sref parameter to anchor all generations to brand identity
• →Built a QA pipeline that filters out off-brand outputs

/imagine [product], --sref https://brand-library.com/ref-001.jpg --sw 50

The Results

After 2 weeks of implementation:

• →500+ photos generated per day (vs. 20 per week before)
• →$13.5k monthly savings (90% cost reduction)
• →3-hour turnaround from concept to final asset
• →100% brand consistency across all outputs

The client now launches new products in days instead of months. Photography is no longer a bottleneck—it's a competitive advantage.

Key Takeaways

• →Prompt engineering is a skill. Good prompts = good outputs
• →Batch automation scales. Don't manually generate one at a time
• →Style references are critical. They maintain brand identity
• →QA is non-negotiable. Filter out edge cases before publishing