Skip to main content

What is the best way to roll out Looksy?

Start with a representative product set, verify the full shopper journey, establish a baseline, and expand based on observed quality and use. There is no universal conversion or return-rate result.

Choose a useful starting set

Include products with clear imagery, meaningful traffic, and reliable order/return data. A smaller set makes it easier to diagnose theme, image, targeting, and measurement issues before a broad rollout.

Make the action understandable

  • Use literal button copy such as Try On or See it on you.
  • Maintain readable contrast and a usable tap target.
  • Use a clear corner of the main product image without covering garment details, gallery controls, or sale badges.
  • Test the actual theme on desktop and mobile.

Set accurate expectations

Describe a generated Looksy result as a visual preview of a product on a shopper’s photo. The generated image does not predict size, guarantee fit, reproduce fabric behavior perfectly, or replace size charts and product details. If you enable Size Recommendations, explain that Looksy compares measurements entered by the shopper with your reviewed product chart. Keep the sizing guidance separate from claims about the generated image and never promise a perfect fit.

Improve input quality

  • Prefer product images that clearly show the item and important details.
  • Review different products rather than assuming one successful preview generalizes.
  • Ask shoppers to use a clear photo of one person with the relevant body area visible.
  • Treat generation errors and weak previews as evidence to adjust the product, imagery, or shopper guidance.

Measure before expanding

Track adoption and assisted behavior using Looksy’s definitions, then compare with Shopify outcomes over a stable period. Change one major variable at a time and document promotions, theme changes, stock changes, and tracking gaps.
Use “associated with” for observational analytics. Use causal language only when a controlled test supports it.

Review regularly

  • Check enabled products and collections after catalog changes.
  • Re-test theme integration after publishing a new theme or modifying product templates.
  • Review credits before promotions or peak periods.
  • Update shopper guidance when repeated support questions reveal confusion.
  • Remove unsupported claims when product behavior or commercial terms change.

Image Guidance

Review product and shopper-photo quality factors.

A/B Testing

Test one change against a stable baseline.