On October 16, 2025, our team spent an incredible afternoon in London at the #Digital Shelf Summit Europe 2025. We were thrilled to connect with industry leaders and immerse ourselves in the conversation about the future of the digital shelf.
The summit centered on the profound impact of AI on the commerce landscape, moving from traditional e-commerce to "Agentic Commerce." Agentic Commerce is shopping powered by intelligent AI agents capable of anticipating, personalizing, and automating every step of the process to create frictionless, proactive experiences. In addition to the impact of AI on shopping behavior, AI is also driving brands to implement significant changes in technology, data strategy, and organizational structure.
VML’s Gemma Spence moderated the panel "From Algorithms to Anticipation," which addressed the accelerating pace of commerce. One key message was, "Change has to happen at the speed of commerce," underscoring that traditional algorithms are no longer sufficient. The panelists stated that it is crucial to "Break down company silos, you all need to be successful." When the question was raised how the success of AI adoption can be measured, the collective answer was that current Digital Shelf KPIs will remain equally relevant,but the numbers will shift as AI will create more efficiency. However, new KPIs are required to measure e.g., LLM bias. As a final key insight, the panelists shared, "To be successful, you need to lead and show and create space for failure."
Here are our key takeaways from the rest of the sessions:
1. The rise of Agentic Commerce:
- Evolution: Shopping is moving from in-store (Era 1) and traditional e-commerce (Era 2, characterized by distinct steps like search, SERP, PDP, checkout) to "Agentic Commerce" (Era 3).
- Single Interface: Agentic Commerce involves a single, conversational interface (via chat or voice) where an AI agent handles all "jobs to be done" for the shopper, from product discovery to checkout.
- AI's Disruption: AI is disrupting core commerce tech stack components, including UI/UX, search, Product Detail Pages (PDPs), personalization, dynamic pricing, promotions, and merchandising.
- Data Challenge: For AI agents to effectively sell products, they require a "crap load more information" than traditional e-commerce. This includes detailed product data, use cases, scenarios, trends, and brand guidelines. LLMs will "make stuff up" if not fed with accurate, comprehensive data. The shift is from keyword optimization to comprehensive product knowledge.
- Content is King (but evolving): While content remains crucial, it needs to be optimized for AI agents. This means clear, rich product titles, maximizing metadata, embedding answers to conversational questions, and leveraging enhanced content. PDPs are foundational as they train these AI models.
2. Reinventing the organization for modern digital commerce:
- Outdated Structures: Current organizational structures, often built around in-store shopping or siloed e-commerce teams, are not sustainable for growth in the AI era. Adding headcount for e-commerce is not scalable.
- Omnichannel Imperative: The future organization needs to be "channel-less," with every role being "omni-focused." The "eCommerce" department as a separate entity will disappear.
- AI-Enabled Future Org: A proposed model features a "Category Leader & Orchestrator" at the center, supported by various AI agents (e.g., Content Briefing Agent, Demand Planning Agent, Legal Contract Review Agent) and human roles working harmoniously.
- Mindset Shifts: Key shifts include deploying shared goals across the company, creating joint commercial leadership, focusing on omnichannel talent development, identifying how AI can free up human time, recentering around the consumer, and balancing short-term pressure with long-term vision.
Breaking Silos: Cross-functional teams are essential. Traditional offline account teams and brand teams need to integrate more deeply with digital strategies.
3. Innovation and Data-Driven Decisions:
- Consumer Insights: Innovation must be grounded in deep consumer insights. Examples included rebranding based on packaging issues and launching digital-first brands.
- Cracking Algorithms: For new product launches, it's crucial to understand and "crack the algorithms" of platforms like Amazon. Experiential marketing, social media, and influencers are vital, especially with limited budgets.
- Retail Media: Becoming increasingly important, requiring gold-standard content and imagery.
- Data Utilization: Web scraping programs can inform product roadmaps and transform product language. Reviews are critical for insights and keyword tailoring.
- Measuring AI Success: While traditional KPIs (e.g., sales, efficiency) will still apply, new metrics might be needed for AI-specific challenges like LLM bias. The focus is on the cost of doing business and creating better talent.
