Virtual Try-On: Do Returns and Emissions Really Drop?

Virtual try-on (VTO) can reduce returns and emissions, but the size of the effect depends heavily on the technology, the product category, and how the retailer implements it. For fashion items like eyewear and cosmetics, VTO can cut return rates by 20–40% and lower the carbon footprint of shipping and reverse logistics by a similar margin, but for clothing with fit issues the impact is smaller and less consistent.

Formula

This calculator estimates the potential reduction in returns and the associated emissions savings when a retailer implements virtual try-on. You enter the current return rate, the expected reduction from VTO, the number of shipments per return (outbound and return), the average distance per shipment, and the CO2 emissions per kilometer per package. The calculator outputs the new return rate, the number of returns avoided per 1,000 orders, the distance saved, and the CO2 saved in kilograms.

The formulas are:

• New return rate = Current return rate × (1 − Reduction from VTO / 100)
• Returns avoided per 1,000 orders = 1,000 × (Current return rate / 100) × (Reduction from VTO / 100)
• Distance saved = Returns avoided × Shipments per return × Distance per shipment
• CO2 saved (kg) = (Distance saved × CO2 per km per package) / 1,000

All inputs are percentages or positive numbers. The default values represent a typical online clothing retailer with a 30% return rate, a 25% reduction from VTO, two shipments per return (one outbound, one return), an average shipment distance of 500 km, and 100 g CO2 per km per package.

Worked Example

Using the default values:

1. Current return rate = 30%
2. Reduction from VTO = 25%
3. New return rate = 30 × (1 − 25/100) = 30 × 0.75 = 22.5%
4. Returns avoided per 1,000 orders = 1,000 × (30/100) × (25/100) = 1,000 × 0.3 × 0.25 = 75 returns
5. Distance saved = 75 × 2 × 500 = 75,000 km
6. CO2 saved = (75,000 × 100) / 1,000 = 7,500 kg

So with the defaults, VTO would cut the return rate from 30% to 22.5%, avoid 75 returns per 1,000 orders, save 75,000 km of shipping, and reduce CO2 emissions by 7,500 kg.

Background: Why Returns and Emissions Are Linked

E-commerce returns are a major environmental problem. Every return involves at least two shipments: the original delivery and the return trip. In many cases, returned items are also restocked, inspected, repackaged, and potentially shipped again to another customer. Each shipment consumes fuel and produces greenhouse gases. The carbon footprint of a single returned garment can be several kilograms of CO2, depending on distance and transport mode. Reducing returns by even a few percentage points can have a measurable impact on a retailer’s carbon footprint.

Virtual try-on technology aims to reduce returns by giving customers a better sense of how a product will look or fit before they buy. The technology ranges from simple size recommendation tools to augmented reality (AR) overlays that show a product on a live image of the user. The effectiveness of these tools varies widely by product type and implementation.

How Virtual Try-On Works

Virtual try-on systems generally fall into two categories:

• Size and fit recommenders: These use body measurements, past purchase history, and fit preferences to suggest a size. They often rely on algorithms that analyze thousands of size charts and customer feedback.
• Augmented reality (AR) overlays: These use a smartphone camera to superimpose a product—like glasses, shoes, or makeup—onto the user’s live image. Advanced systems can track facial features or body movements to make the overlay realistic.

For eyewear and cosmetics, AR overlays are highly accurate because the product’s appearance is the primary factor. For clothing, fit is more complex, and many systems combine AR with size recommendations. Some retailers also use 3D body scanning, where users enter their measurements to generate a virtual avatar.

Evidence of Impact on Returns

Several retailers and technology providers have reported reductions in return rates after implementing VTO. For example, eyewear retailers have seen return rates drop from around 20–30% to under 10% when customers can virtually try on glasses. Cosmetic brands have reported similar improvements for color-matched products. For clothing, the results are more mixed: some studies show a 10–20% reduction, while others show negligible change, especially for items where fit is highly variable.

A key factor is whether the VTO is used before purchase. If customers only use it out of curiosity but still order multiple sizes, returns may not drop. Retailers that integrate VTO into the size selection process and show a clear recommendation tend to see the biggest gains.

Emissions Savings: The Math

The emissions saved by VTO come primarily from avoiding return shipments. Using the calculator above, you can estimate the impact for a specific scenario. For a large retailer with 1 million orders per year, a 30% return rate, and a 25% reduction in returns, the savings would be:

• Returns avoided: 75,000 per year
• Distance saved: 75,000 × 2 × 500 = 75 million km
• CO2 saved: 75 million × 100 g/km = 7.5 million kg = 7,500 metric tons

That is equivalent to the annual emissions of about 1,600 passenger cars in the US. The actual savings depend on the shipping distance, mode (truck vs. air), and the weight of the package. Air freight has a much higher carbon intensity per km than ground shipping.

However, VTO itself has an environmental cost: the servers that run the AR algorithms and the energy used by customers’ devices. These are typically small compared to the shipping savings, but they are not zero.

Common Problems and Limitations

Virtual try-on is not a magic bullet. Common issues include:

• Inaccurate size predictions: If the algorithm is poorly trained, it may recommend the wrong size, leading to frustration and returns anyway.
• Poor AR rendering: Lighting, skin tone, and movement can make the overlay look unrealistic, reducing customer trust.
• Limited product coverage: Some retailers only offer VTO for a few products, so customers may not use it consistently.
• Customer behavior: Even with VTO, some customers deliberately order multiple sizes (bracketing) to try at home, negating the benefit.

For clothing, fabric drape and stretch are hard to simulate. A virtual avatar may show how a shirt looks but not how it feels or whether it wrinkles. These limitations mean that VTO is more effective for products where appearance is the main purchase driver.

How to Choose a Virtual Try-On Solution

Retailers considering VTO should evaluate:

• Product type: Is the product appearance-based (glasses, makeup) or fit-based (jeans, dresses)? AR works well for the former; size recommenders are better for the latter.
• Integration: Does the VTO integrate with the product page and checkout? Can it be embedded in a mobile app?
• Data requirements: Does the system require users to enter measurements? If so, is that a barrier to adoption?
• Accuracy metrics: Does the vendor provide data on return rate reduction from their other clients?
• Cost: VTO solutions can be expensive, with licensing fees and integration costs. The ROI depends on the volume of returns avoided.

Start with a pilot on a few product categories, measure the return rate before and after, and then scale up if the results are positive.

Alternatives and Complementary Strategies

Virtual try-on is not the only way to reduce returns. Other effective methods include:

• Better product descriptions: High-resolution photos, videos, and detailed size charts.
• Customer reviews: Photos and comments from other buyers about fit and quality.
• Free returns: Paradoxically, free returns can increase returns, but they also increase sales. Some retailers use a mix of free returns and VTO to maximize net profit.
• Return prevention at checkout: Reminders about fit and size can reduce impulse purchases.

For emissions, other strategies include consolidated shipping, using eco-friendly packaging, and optimizing delivery routes. VTO should be part of a broader sustainability plan, not a standalone solution.

Frequently Asked Questions

Does virtual try-on work for all clothing items?

No. It works best for items where fit is relatively simple, like t-shirts or outerwear, and for appearance-based products like glasses. For complex garments like tailored suits or jeans, the technology is improving but still not fully accurate.

How much does virtual try-on cost for a retailer?

Costs vary widely depending on the vendor and the complexity. Basic size recommendation tools can start at a few thousand dollars per month, while advanced AR solutions can cost tens of thousands. The ROI depends on the volume of returns avoided.

Can virtual try-on reduce emissions enough to matter?

Yes, if the return rate reduction is significant. As the calculator shows, a 25% reduction in returns for a large retailer can save thousands of metric tons of CO2 per year. But the exact amount depends on shipping distances and modes.

Do customers actually use virtual try-on?

Adoption rates vary. Some retailers report that over 50% of customers use the feature when it is prominently displayed, but others see lower usage. Making the VTO easy to access and showing a clear recommendation increases engagement.

Is virtual try-on just a gimmick?

No, but it can be if implemented poorly. When done well, with accurate rendering and integration into the purchase flow, it has been shown to reduce returns. When done as a standalone feature without guidance, it may not change behavior.

Related

• Hercules Power Tools at Harbor Freight: Worth It?
• Most Versatile Ryobi Power Tools
• Organize Screenshots Automatically with Power Automate

in t | Current return rate (%) | % | 30 | 0 | 100
in r | Reduction from VTO (%) | % | 25 | 0 | 100
in s | Shipments per return | - | 2 | 1 | 10
in d | Distance per shipment (km) | km | 500 | 1 | 20000
in co2 | CO2 per km per package (g) | g | 100 | 1 | 5000
out new_rate | New return rate (%) | % | t * (1 - r/100) | 1
out saved_returns | Returns avoided per 1000 orders | - | 1000 * (t/100) * (r/100) | 0
out saved_km | Distance saved per 1000 orders (km) | km | saved_returns * s * d | 0
out saved_co2 | CO2 saved per 1000 orders (kg) | kg | (saved_km * co2) / 1000 | 1