How to Measure Virtual Try-On Results on Your Store

Most stores install a new app, glance at the dashboard for a week and then argue with themselves about whether it worked. That is a bad way to spend a subscription. A plain spreadsheet and some discipline will give you a real answer.

Which metrics should you track for virtual try-on?

Four numbers cover the whole path from browsing to a parcel that stays kept. They can move independently, so a single headline figure hides more than it shows.

• Return rate: the share of orders on try-on products that come back. This is the cost the feature exists to cut, and it is the only one of the four that maps directly onto money you can see leaving.
• Conversion rate: visitors who complete a purchase. Useful, but noisy, since traffic mix and promotions push it around.
• Cart-add rate: visitors who add a product to the cart. It often shifts first, because try-on answers the fit question before the shopper reaches checkout.
• Average order value: what a completed order is worth. Shoppers who feel sure about fit sometimes add a second item, and sometimes they just stop ordering two sizes of the same one.

Add one adoption number alongside them: how many shoppers actually start a try-on. If nobody clicks the button, the other four have nothing to reflect. Corlen counts the try-ons your store uses each month, which gives you a rough adoption read without any extra setup.

How do you set a fair baseline?

Before you switch anything on, write down the four numbers for the products you plan to enable, covering the last four to eight weeks. Use the same products, not your whole catalog. A store-wide average blends in categories the feature never touches, and it will drown any real change.

Keep a note of anything unusual in that window: a sale, a paid campaign, a viral post, a stock-out. These are the things that make a spreadsheet lie convincingly. For context on what a normal return rate looks like in fashion, see our sourced return statistics.

How do you run the comparison step by step?

1. Choose a test group: a set of products that get the try-on button, ideally ones with a known fit problem and steady traffic.
2. Keep a comparison group: similar products without the button, if your setup allows it. Otherwise, compare the same products before and after.
3. Pull the four metrics for both groups from Shopify Analytics, and from GA4 if you use it, by product or by product page.
4. Wait one full purchase cycle for the category. Returns lag orders by days or weeks, so return rate needs the longest wait.
5. Compare like with like: the same length of period, no sale weeks, and the same traffic sources where you can.

One warning on returns: they arrive late. If you check after ten days, you are counting orders whose returns have not shown up yet, and the number will look better than it is. Give it time, and check again.

How do you read the results honestly?

Be suspicious of your own good news. Shoppers who bother to try a garment on are probably more serious buyers to begin with, so higher conversion among try-on users does not prove the feature caused it. That is why the product-level comparison matters more than a comparison of users against non-users. Published studies have the same limit, which we cover in does virtual try-on increase conversion rate.

A quick sanity check follows. If returns fell but adoption was tiny, something else probably changed. If adoption was high and nothing moved, look at photo quality and button placement before you blame the technology.

Once you have your own numbers, plug them into the ROI calculator to see what the change is worth in revenue. It uses your visitor count and order value instead of a borrowed percentage, and the walkthrough of that calculator explains each input. If you run Shopify, the Corlen Shopify app adds the button as a theme block, so you can test it on a handful of products first. Or try Corlen yourself and see what your shoppers would see.