Evaluate a virtual try-on vendor by the retail journey it can deliver, the evidence it can show and the obligations it will put in writing. Before signing, ask for an agreed scope, an acceptance demonstration, a complete cost schedule, a data-handling explanation and a workable exit plan. A convincing visual demo is the beginning of that evaluation.
Imagine an ecommerce director, a retail head and a finance lead reviewing the same proposal. One expects a website experience, one expects a staffed kiosk and one assumes the quoted fee includes both. This illustrative buying problem is avoidable: make the proposal describe exactly what shoppers and staff will receive, who maintains it and what happens when something fails.
This checklist is for fashion founders, ecommerce, retail, innovation and technology leaders comparing suppliers, including iKawn Mirror. It focuses on the evidence and commitments needed to approve a purchase. The framework is an editorial recommendation, not a vendor ranking or a claim about measured iKawn outcomes.
Why the proposal deserves as much attention as the demo
The BoF–McKinsey State of Fashion 2026, published November 17, 2025, forecast low single-digit global fashion growth in 2026 and reported that more than 35% of surveyed executives already used generative AI in areas including customer service, content creation, search or product discovery. These are broad industry findings, not virtual try-on adoption or ROI benchmarks. Our buying implication: a new customer-facing investment should have a specific commercial job and a cost that the business can defend.
Trust is also a relevant adoption barrier. DHL’s 2026 E-Commerce Trends Report announcement, published June 2, 2026, identifies consumer privacy and trust concerns alongside businesses’ plans to expand generative AI use. Its research covered 29,000 shoppers and 5,800 businesses across 29 countries. This is wider ecommerce context; it does not establish how your shoppers will respond to a camera or photo-based fashion experience. Put that question into the evaluation.
1. Make every supplier answer the same buying brief
Write a short brief containing the priority channel, customer hesitation, product category, intended next action and person accountable for the result. For example: “Help store visitors compare selected occasionwear and request a physical garment from an associate.” That is a proposed use case, not a reported deployment.
Ask each vendor to separate what is available now, what needs configuration or integration, what requires custom work and what is only planned. Require the proposal to name supported devices, deployment locations, catalog units and exclusions. “Omnichannel” is insufficient if the quote leaves the website and store experiences undefined.
Define the product’s role precisely. In its June 22, 2026 explanation of its size-and-fit technology, Zalando describes a layered system of size advice, customer body measurements and a 3D-avatar virtual fitting room. This public retailer example shows why a size recommendation, a fit-oriented avatar and a visual styling preview should not be treated as interchangeable deliverables. Ask suppliers which uncertainty their proposed experience addresses and which it leaves unresolved.
2. Turn the demonstration into an acceptance record
Request a demonstration using representative products and the intended shopper flow, with the device and connection conditions you expect to support. Record what was tested, what passed, what failed and what was excluded. A prepared recording can introduce the product; the acceptance review should let your team inspect the proposed experience and ask for ordinary actions to be repeated.
Include the full journey: start a session, choose a product, change the selection, reach the correct next action and end the session. Observe delays, interruptions and recovery. Ask what happens when a product becomes unavailable, the connection drops or a shopper declines a camera or photo step. Agree which failures block launch and who approves the correction.
Keep product representation as a separate brand-approval gate, owned by merchandising. Use your pilot product shortlist to define the sample; do not let a demonstration of a narrow selection imply acceptance of the entire catalog. Record supported scope in the proposal before calculating a wider rollout.
3. Request one complete cost schedule
Ask the supplier to itemize setup, catalog onboarding, hardware where relevant, integration, software usage, support and ongoing changes. Confirm taxes, delivery, installation, replacement equipment and any third-party charges where applicable. An item can be included, optional or excluded, but it should not be ambiguous.
For usage-based pricing, establish exactly when billing starts and stops, whether idle time or retries count, what happens at the allowance limit and how your team can check consumption. Obtain the overage rate, expiry rules and the price of increasing capacity in writing. Ask for a worked invoice example using the same expected usage for every vendor.
Compare the cost of the initial pilot and the cost of the rollout you would actually approve. A low entry fee may be acceptable if future costs are explicit; a broad fixed fee may be acceptable if its inclusions match your needs. Use the iKawn Mirror pricing guide as a starting reference and request a scope-specific proposal. Do not assume a public starting price covers your particular configuration.
4. Resolve data handling before customer access
Ask for a plain-language account of what happens to camera input, uploaded photos, generated output, session records and contact details, if collected. Request the processing locations, service providers involved, retention periods, deletion process, access controls and whether any material is used for model improvement. Have your privacy and security owners review the answers against your requirements.
Distinguish a feature demonstrated today from a policy or control merely promised. Ask the vendor to show how a shopper ends a session and how the next visitor is prevented from seeing the previous visitor’s content. Confirm who provides the customer notice and handles a deletion request. These are procurement questions; this guide does not assert that every vendor, or every Mirror deployment, has the same controls.
A pilot can reveal whether the explanation is understandable to shoppers. It should not be used to defer unanswered data-handling questions until after launch.
5. Name the owner when the experience stops working
Request a responsibility list covering catalog changes, device setup, network issues, software faults and customer-facing recovery. For each item, record the retailer’s owner, supplier contact, support hours and escalation route. Distinguish acknowledgment of a support ticket from restoration of service.
Ask what support is included during your actual trading or event hours, how incidents are reported, and which fallback keeps the buying journey usable. Confirm whether replacement hardware and on-site assistance are included or separately charged. Require notice and an approval process for changes that materially affect the agreed experience.
The tradeoff is practical: a more managed service can reduce internal work while increasing supplier cost; a lighter service can suit a capable internal team. Compare the staffing obligation as well as the subscription fee. Neither option removes the need for a named retailer owner.
6. Agree the evidence for renewal—and the way out
Define the pilot readout before signing: eligible exposure, completed sessions, the agreed next action, operational failures and the commercial outcome you can measure responsibly. Ask who provides each data point, in what format and how your analyst can check it. A product-page visit, a garment request and a paid order are different events.
If a supplier cites a conversion or returns result, request the source, period, category, sample and comparison method. Decide whether it applies to your journey. A higher purchase rate among people who voluntarily use try-on does not by itself establish that try-on caused the difference. Keep projected benefits separate from observed outcomes and agree what evidence justifies expansion.
Also confirm the pilot end date, renewal notice, minimum commitment and cancellation process. Ask which retailer-owned catalog assets and agreed reports can be exported, what happens to shopper data, who removes the installation or integration, and what offboarding costs apply. An exit plan is a way to make a limited test genuinely limited.
Use a decision sheet that exposes unresolved work
For each of the six areas, record the requirement, evidence supplied, responsible owner, unresolved point and decision deadline. Mark it demonstrated, documented but untested or unresolved. This is a working review method, not a research-derived score.
Make three decisions separately: is the experience commercially relevant, is the deployment operable, and are its obligations acceptable? Do not let a strong demo average away an unanswered launch-critical question. Conversely, a feature outside the agreed first use case need not block a focused pilot if the exclusion is explicit.
“Can we approve a pilot before the full rollout is specified?” Yes, if the pilot’s deliverables, spending limit, owners, data handling and exit are clear. Record the questions that still need answers before expansion.
“Should we choose the lowest quote?” Compare quotes against the same scope and responsibility list first. The useful comparison is the total cost and work required to deliver the agreed journey, including the effort your own team must contribute.
Bring your buying brief to an iKawn Mirror demo
iKawn Mirror offers live and photo-based virtual try-on for fashion discovery across ecommerce, stores and activations. Its role is discovery, comparison and preselection; physical fit, fabric feel and final sizing judgment remain separate needs. Evaluate it against your intended journey and representative products.
Bring your priority channel, product shortlist, operational constraints and proposal questions. Use the demonstration to identify what is available, what needs scoping and what evidence your team requires before approving a pilot. Confirm pricing, support, integrations, data controls and exit terms for the specific deployment; this article is not a statement of contractual inclusions or a promise of sales uplift.