A fashion ecommerce team sees a convincing virtual try-on demo. The shopper looks natural, the jacket looks desirable, and someone asks the question that could unlock the budget: will this reduce returns?
It is a reasonable question, but an incomplete buying brief. A jacket returned because the shopper disliked its appearance presents a different problem from one returned because the shoulders were too tight. A late delivery presents another problem altogether. Before attaching a savings target to virtual try-on, decide which uncertainty the proposed experience can actually resolve.
Virtual try-on can contribute to fewer returns, but a visual preview alone is not evidence of accurate sizing or a guaranteed commercial result. For buyers evaluating iKawn Mirror, the useful test is whether better discovery produces more purchases customers keep, at an acceptable total cost.
Why this question belongs in the budget meeting
The National Retail Federation and Happy Returns estimated that 19.3% of US online sales would be returned in 2025. That is a cross-retail estimate, not an apparel benchmark, an Indian market estimate, or a forecast for your brand. It establishes the scale of the commercial issue; your own category and return-reason data must establish the opportunity. Source: NRF, October 15, 2025.
Budget pressure makes that distinction consequential. In the BoF–McKinsey State of Fashion 2026 executive survey, 46% of respondents expected industry conditions to worsen in 2026. McKinsey also forecast low single-digit industry growth. These are dated expectations, not confirmed full-year results. Our interpretation: a try-on proposal will be stronger if it identifies a recoverable commercial loss instead of treating engagement as sufficient justification. Source: BoF–McKinsey, November 17, 2025.
What public retailer evidence does—and does not—prove
On June 22, 2026, Zalando reported return-rate reductions of up to 40% in recent Virtual Fitting Room pilots. Its description matters: the experience uses body measurements to build a 3D avatar for comparing sizes. The same article separately reports that its broader size-and-fit solutions prevented 8% of size-related returns overall in 2025. Those figures describe different scopes and should not be combined. Source: Zalando, June 22, 2026.
This is a retailer-reported pilot result, not an independently transferable uplift. The article does not provide enough experimental detail to reconstruct the 40% comparison, and describes achieving similar impact at scale as an ongoing objective. Ask a vendor for the category, denominator, comparison group and return window behind any comparable claim.
Another public example illustrates a different job. Walmart’s September 15, 2022 announcement described its Be Your Own Model apparel experience as letting shoppers view clothing on their own photograph, with entry through an enabled product page. The announcement demonstrates a historical discovery workflow; it does not establish a measured reduction in returns. Source: Walmart, September 15, 2022.
The buying implication: compare the customer uncertainty each system addresses before comparing its headline results. Neither retailer is presented here as an iKawn customer, and neither result is an iKawn Mirror performance claim.
Start with the returns you could plausibly influence
Ask CX and merchandising to review return reasons alongside a sample of customer comments and products. Treat the following as a proposed triage framework, not a claim about the distribution of your returns:
- “I did not like how it looked on me.” A faithful visual preview is a plausible intervention. Test whether shoppers can judge the style and silhouette more usefully before ordering.
- “The size was wrong.” Establish whether the product includes validated size advice. A realistic image should not be sold internally as a measurement or fit guarantee.
- “The colour, print or garment was different.” Review product truth. An attractive preview that changes defining details could create new expectation gaps.
- “The fabric felt wrong,” “it arrived late,” or “it was damaged.” Do not assign savings to visual try-on without a specific, testable mechanism. Fabric feel, logistics and product quality need their own interventions.
Ambiguous labels such as “not suitable” need investigation before they enter the business case. If most of the addressable problem is sizing, evaluate size advice alongside visual try-on. If it is discovery and appearance confidence, a visual experience may be the more relevant starting point.
Make the business case survive a zero-uplift scenario
Finance should first price a pilot that produces no benefit. Include software, integration, usage, catalogue preparation, support and measurement; add hardware and staff time for an in-store deployment. A fixed budget and exit decision make this an evidence purchase rather than an open-ended rollout.
Then calculate the improvement required to cover those costs. The example below is illustrative arithmetic, not an industry benchmark, a quote or a forecast for Mirror. Assume one item per order, an unchanged order volume, and only avoidable return-handling savings:
- 10,000 eligible monthly orders, with a 25% baseline return rate.
- A hypothetical decline to 23%: two percentage points, or an 8% relative reduction.
- 200 fewer returns × ₹350 of avoidable handling cost = ₹70,000 monthly savings.
- ₹90,000 in total monthly programme cost leaves a ₹20,000 shortfall on handling savings alone.
At those assumptions, break-even requires approximately 258 avoided returns, or a 2.58-percentage-point reduction across the 10,000 orders. The test must therefore detect an effect large enough to matter financially. If your eligible volume is much smaller, a short pilot may establish usability without settling the savings question.
Retained sales could add value, but model them separately and reconcile the calculation with finance. Avoid counting the same retained margin twice, treating a refunded order’s full selling price as profit, or claiming staff savings that do not release capacity or expense. Replace every illustrative input with your own contribution economics.
Design a pilot that answers the returns question
Choose a category and a mechanism. For example: “A visual preview will help shoppers reject unsuitable styles before purchase.” Select representative products with stable availability, enough eligible traffic and reliable return-reason records. Agree which garments and shopper situations the experience must handle before launch.
Compare access to the experience, not enthusiastic users with everyone else. Randomly offer eligible shoppers the experience or the existing journey, where feasible. People who volunteer to use try-on may already be more engaged. Keep shoppers in their original comparison group even if they decline or abandon the feature; that measures the effect of offering it.
Keep the commercial ledger complete. Track eligible shoppers, successful previews, purchases, delivered units, returned units and retained contribution. Define whether a return rate is measured by items, orders or sales value, and keep that definition consistent. A lower percentage alongside more sales can still produce more returned units.
Wait for comparable return windows. Set the purchase attribution period and returns-maturity rule before reading results. Compare cohorts with equivalent follow-up. Have the analyst estimate required sample size from the baseline and the minimum worthwhile effect; a convenient campaign end date is not proof that the experiment is conclusive.
Pre-agree the decision. Scale when retained contribution clears the agreed cost threshold and customer-experience checks pass. Extend only when the remaining uncertainty and cost of learning justify it. Stop or revise when adoption is too low, the garment preview misleads shoppers, or the plausible benefit cannot cover the cost. Report an inconclusive result as inconclusive.
Objections worth resolving before procurement
A shopper who declines a photograph or camera should still have a useful shopping journey. Ask what images are stored, where processing occurs, who can access them, how deletion works, and what notice the shopper sees. Obtain answers for the proposed deployment rather than assuming all try-on workflows handle data identically.
Also test the awkward garments and situations, not only the best demo. Changed prints, missing details, long waits and failed previews can undermine the confidence the project is meant to build. Agree a fallback to ordinary product discovery and a clear owner for catalogue updates.
For store or activation use, a successful interaction is not automatically a return-reduction result. If purchases and subsequent returns cannot be linked appropriately, evaluate discovery, assisted-selling handoff and operating cost instead. The case may still be worthwhile, but it answers a different budget question.
What to bring to an iKawn Mirror evaluation
iKawn Mirror supports visual discovery and preselection across ecommerce and physical retail journeys. Its public positioning does not guarantee conversion gains, fewer returns or physical-fit certainty. Evaluate it against the uncertainty your shoppers actually face.
Bring your priority category, recent return reasons, eligible traffic, contribution assumptions and intended next action after try-on. Use the Mirror workflow guide to frame the journey, and the pricing guide to identify what a scoped quote needs to include.
A useful first conversation should leave your team with a testable commercial question, a realistic catalogue scope and a decision rule. If appearance confidence is a credible part of your returns problem, book a live Mirror demo to assess the experience and discuss a focused pilot.
Published September 21, 2026. Public retailer examples are attributed evidence, not endorsements. Hero image is an AI-generated editorial illustration. Financial assumptions and the evaluation framework are original iKawn analysis.