Virtual Try-On Technology Transitions from Experimentation to Practical Application

For years, virtual try-on technology lingered in a state of anticipation, continually viewed as an "almost" solution. The concept of allowing customers to visualize how clothing or makeup would look on them has been appealing, yet the implementation proved challenging. Brands struggled with the costly 3D product data requirements and faced issues with poorly lit selfies, leading to results that felt more odd than helpful.

However, this landscape is shifting significantly. Generative AI is now capable of transforming standard product images into 3D models that accurately simulate elements such as cut, drape, and fabric texture in real time. This advancement eliminates the need for the cumbersome 3D modeling processes that previously defined the technology. Retailers like ASOS are leading the way, allowing customers to upload photos or create digital twins based on their body dimensions and preferences. In Germany, Breuninger has integrated Google's virtual try-on capabilities into its mobile app, while Maybelline offers users the option to experiment with lipstick shades through uploads, digital models, or live camera functionality. These implementations are not just experimental trials; they are fully operational and actively delivering measurable results.

The significance of this transition—moving from AI pilots to fully integrated infrastructure—makes virtual try-on a key reference point within the retail industry. Online fashion retail has grappled with two persistent issues: hesitation in purchasing due to uncertainty about product appearance and high return rates resulting from size discrepancies. In the U.S., the National Retail Federation projected a return rate of 19.3% for online fashion purchases by 2025, with Gen Z shoppers averaging eight returns each. Arnold Pötsch, a leading author in the BVDW working group on 3D applications in e-commerce, asserted that advancements in computer vision and real-time rendering have provided forward-thinking retailers a distinct edge by creating a seamless connection between physical products and their digital counterparts, influencing personalized shopping experiences, reducing return rates, and contributing to sustainability goals.

The implications extend beyond just fashion. The trend of AI scaling beyond pilot projects to become directly aligned with profitability is emerging across various sectors. In fintech, AI is shifting from optional enhancements to essential components of underwriting and fraud detection processes. In healthtech, models for diagnostics and triage are crossing the threshold of trust that had previously kept them in limbo. Companies in deeptech are successfully moving foundational research into viable products within realistic timeframes. Meanwhile, SaaS providers are increasingly evaluated on measurable outcomes like customer retention and margins rather than an extensive list of AI features. Additionally, sustainability is evolving from a mere compliance requirement to a genuine efficiency driver, as AI helps minimize waste across industries, including fashion and technology infrastructures.

This evolution aligns with DMEXCO's chosen theme for 2026: "Scaling Intelligence." This concept emphasizes the transition from experimenting with AI to creating tangible value. It accurately reflects the current market conditions where the initial gains from basic AI implementations have diminished. The focus is now on developing AI solutions that not only work at scale but also earn user trust and demonstrate quantifiable results that matter to investors and operators.

Scheduled for September 23–24 in Cologne, DMEXCO 2026 will convene decision-makers from agencies, commerce, technology, and media to explore these pressing challenges. Founders, operators, and investors seeking insights into where AI is translating to revenue and customer retention rather than mere headlines will find an agenda that addresses various sectors, including fintech, healthtech, deeptech, and sustainable innovation. Attendees will have the opportunity to evaluate which aspects of AI hype have matured into functional infrastructure and which are still in the pilot phase.

The event's format is intentionally diverse. The Expo area juxtaposes scaled solutions from established organizations with early-stage technology in the Start-up Area, providing attendees a chance to compare working applications against emerging innovations. The conference sessions will feature insights from operators detailing their experiences with moving models from pilot to production and the initial points where return on investment materialized. This is particularly valuable for a European tech audience, as the conversation around AI scaling is often dominated by U.S. platforms and Chinese manufacturing. DMEXCO’s agenda is notably abundant in European innovators across various domains, addressing similar challenges under distinct regulatory frameworks and capital landscapes.

While it took a decade for virtual try-on technology to mature from novelty to essential tool, the advancements in vertical-specific AI applications across fintech, healthtech, deeptech, and SaaS are poised to progress more swiftly. Nonetheless, they will face a similar challenge: proving their functionality at scale, with real users, and delivering numbers that hold up. This is the core dialogue at DMEXCO 2026.