For brands·5 min read

Shade finder vs. shade quiz vs. AR try-on: which one actually reduces returns

All three get pitched as “the fix” for wrong-shade returns. Only one of them removes the actual guess, and it’s worth knowing which before you spend a quarter integrating the wrong one.

The short version

  • Quizzes and AR try-on both get marketed as return-reducers. Neither removes the actual guess that causes wrong-shade purchases.
  • The only method that reliably reduces wrong-shade returns is matching by measured colour against a shade the shopper already owns.
  • See how that works on Shopify, or get early access.

You’ve almost certainly been pitched all three: a shade quiz vendor promising personalization, an AR try-on vendor promising engagement, and a measured shade-matching option promising fewer returns. They all use similar language, “find your shade,” “reduce returns,” “personalized match,” which makes them sound interchangeable when they’re solving genuinely different problems.

Before committing engineering time and a quarter of roadmap to one of them, it’s worth reading the full returns playbook alongside this, so you know which tool actually touches your return rate, and which ones just look like they do.

What each tool actually solves

Shade quiz

Asks the shopper a handful of questions, often including undertone, and recommends a shade. It adds structure and personalization to the shopping experience, but the recommendation still rests on the shopper accurately self-reporting things (like their own undertone) that are notoriously hard to judge by eye.

AR / virtual try-on

Renders a foundation “look” onto the shopper’s photo or live camera feed. It’s genuinely good at engagement and styling preview, it answers “do I like how this looks,” but screen rendering and ambient lighting vary too much between devices for it to verify an actual colour match.

Measured colour matching

Compares a shopper’s real, measured colour, either scanned or inferred from a shade they already own and trust, against your range, and only surfaces a match when the difference is smaller than a person could notice. This is the only one of the three that verifies colour rather than recommending or rendering it.

Quiz and AR try-on both improve the shopping experience. Only measured matching improves the purchase decision in the specific way that prevents a wrong-shade return. Conflating the two is the most common reason a shade tool gets installed and return rate doesn’t move.

Why return rate doesn’t move for two of them

A wrong-shade return happens when a shopper’s belief about their shade doesn’t match reality. A quiz can change which wrong belief they hold. AR try-on doesn’t address the belief at all. It answers a styling question the shopper wasn’t actually stuck on. Neither closes the gap between belief and reality, which is the only gap a return rate actually responds to.

What to actually measure before you decide

  1. Isolate wrong-shade returns specifically

    Overall return rate mixes in sizing, quality, and preference returns. Tag or estimate the shade-specific slice before you can tell if any tool is moving it.

  2. Check where mismatches concentrate

    If wrong-shade returns skew toward your deepest or lightest shades, that’s often a range gap, not a discovery-tool gap, worth knowing before you invest in matching technology instead of catalog breadth.

  3. Pilot before a full rollout

    A single-product pilot with a measured-colour tool, watched against wrong-shade return rate for a few weeks, tells you more than any vendor's case study.

The things nobody tells you

AR try-on and measured matching solve different problems well, consider both

This isn’t necessarily an either-or. AR try-on can genuinely help engagement and browsing time; it just shouldn’t be relied on to reduce returns. Some brands run both, for the jobs each is actually good at.

Integration weight varies a lot between these three

AR try-on tends to be the heaviest technically, camera access, rendering, often a slower page load. A well-built measured-matching embed can be lightweight and load asynchronously without touching your theme, which matters if page speed is already a concern. See what the actual integration looks like.

The cost comparison isn’t just the subscription price. It’s subscription cost against prevented returns. See how to actually calculate that ROI for your store.

How do you know it’s working?

Watch wrong-shade return rate specifically, before and after, over a few weeks of real traffic, not engagement metrics, quiz completion rate, or try-on session length. Those measure adoption, not whether the underlying purchase decision actually improved.

Engagement is not accuracy. Only one of these three actually measures the colour.

For you
See it on Shopify

How measured-colour matching works on a real product page, in a live demo.

See it live
For brands
Get early access

Design partners get preferential pricing and hands-on onboarding as we roll out.

Join early access

Why doesn't a shade quiz reduce foundation returns more than it does?

Because it still asks the shopper to make the same unreliable judgment call, usually about their own undertone, that they'd otherwise make unaided. A quiz adds structure to the guess; it doesn't remove the guessing.

Isn't AR try-on the most advanced option, so shouldn't it work best?

It's the most visually impressive, not the most accurate for colour. Try-on renders a look on a face through a screen, and screen rendering and lighting vary too much to verify an actual shade match. It's built to answer a styling question, not a colour-matching one.

What should we actually track to know if our shade-matching tool is working?

Wrong-shade return rate specifically, isolated from other return reasons, before and after rollout. Overall return rate can mask whether the actual cause, shade mismatch, is improving.

Skintone matches foundation shades by measured colour across 100+ brands. No sponsored placements, just the closest colour.