:::module orientation commercetools is among the most frequently mentioned platforms when DACH region companies consider transitioning to Composable Commerce. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Technology is rarely the success factor. Decision readiness is. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience commercetools positions itself as the leading platform for "Composable Commerce". The vendor explicitly addresses enterprise companies with complex requirements: multi brand, multi market, multi channel. The target audience definition implies: mid market and SMB are not the primary segment. The product complexity and pricing model confirm this orientation. Source: commercetools.com, as of January 2025 Architecture & Delivery Model API first, Headless, Cloud native: commercetools is designed as a pure SaaS solution. There is no on premise option. The entire platform is accessed via APIs. MACH Principles: commercetools is a co founder of the MACH Alliance and embodies its principles (Microservices, API first, Cloud native, Headless). The architecture is fully decoupled. Event based Architecture: All state changes are persisted as events, enabling audit capability and integration into event streaming architectures. Source: commercetools Technical Documentation, MACH Alliance Core Scope & Product Logic commercetools delivers core commerce functions as independently usable modules: Product Information Management, Cart & Checkout, Order Management, Customer Management, Inventory, Payments. What commercetools explicitly does not deliver: Frontend, CMS, PIM (in the extended sense), OMS (complete), Search (native). These components are supplemented via partner ecosystem or custom development. Source: commercetools Product Documentation Ecosystem & Partner Landscape Integration Partners: Partnerships with leading best of breed providers such as Algolia (Search), Contentful/Contentstack (CMS), Akeneo (PIM), Fluent Commerce (OMS). Implementation Partners: Large system integrators (Accenture, Valtech, EPAM) as well as specialised MACH agencies. Source: commercetools Partner Directory, Connect Marketplace Licensing/Pricing Model Logic commercetools uses a GMV based pricing model. Licence costs scale with the revenue processed through the platform. Additionally, API call volume factors into costs. Experience reports speak of entry costs from approximately EUR 150,000/year for medium sized implementations (as of 2024/2025). Source: commercetools Pricing page, market reports Roadmap Signals Active development in: AI/ML Integration, Composable Frontend ("Frontend as a Service"), B2B Extensions, Connect Platform expansion. Source: commercetools Blog, Elevate Conference 2024 Reference Types Global Brands (Audi, Danone, Sephora, Volkswagen), D2C Brands with complex requirements (Bang & Olufsen), B2B/B2C Hybrids (AT&T). Notable: Most references are large enterprises with specialised tech teams. Mid market references in the DACH region are less prominently communicated. Source: commercetools Case Studies ::: :::module transition From here, it is no longer about vendor logic, but about typical deployment realities and decision patterns. ::: :::module reality Business & Industry Fit B2C (simple): Overengineered for simple use cases. B2C (complex): Strengths in multi brand, multi market. B2B: Features in development, not yet market leader. D2C: Good if complexity justifies investment. Multi Brand: Core competency. Complexity & Product Logic commercetools is suited for high catalogue complexity with many variants, complex pricing logic and configurable products. The flexible data modelling supports unusual product structures. For simple catalogues with standard pricing, the platform is oversized. Internationalisation Strong native support for multi currency, multi

commercetools Deep Dive: Vendor Claims vs. Reality Check

<! MODULE 0: ORIENTATION :::module orientation Shopware Deep Dive Vendor Claims vs. Reality Check Shopware is the most deeply rooted e commerce platform in the DACH region. This article analyses the platform without vendor bias. What this article does not do: It provides no recommendation. It does not compare with other platforms. It provides structure for your own decision. ::: <! MODULE 1: STATEMENT I :::module statement Technology is rarely the success factor. Decision readiness is. ::: <! MODULE 2: VENDOR PERSPECTIVE :::module vendor Vendor Perspective Positioning & Target Audience Shopware positions itself as "the leading Open Commerce Platform" in the European market. The combination of open source core and enterprise functionality, as well as particular suitability for the German Mittelstand, are central to its communication. According to vendor information, the target audience ranges from small merchants (Community Edition) to larger mid market companies and enterprises (Rise, Evolve). This dual strategy creates tensions. Source: shopware.com, as of January 2026 Architecture & Delivery Model Shopware 6 is based on the Symfony framework and written in PHP. The architecture was fundamentally modernised compared to Shopware 5. Shopware 6 follows an API first approach and exposes core platform functions through well defined APIs (including Admin API, Store API, Sales Channel APIs). The architecture enables headless approaches, decoupling of frontend and backend, and integrations via APIs and webhooks. Shopware positions this approach as the foundation for flexible commerce setups and modular extensibility. Unlike pure SaaS providers, Shopware offers flexibility in deployment: Self Hosted with full control over infrastructure, PaaS via Shopware Cloud as managed hosting, or Enterprise Cloud with dedicated infrastructure for large customers. The platform can be operated headless but is not a "headless first" architecture. The traditional storefront template remains a core component. The extension system is based on a plugin concept with clear extension points. Source: Shopware Developer Documentation Core Scope Native functionality includes Product Information Management, integrated Content Management with "Shopping Experiences", Cart and Checkout with Flow Builder, Order Management, Customer Management with B2B functions, Rule Builder for flexible business logic, and SEO tools. What Shopware does not natively deliver: Complete OMS for complex fulfilment scenarios, enterprise PIM for very large catalogues, native marketplace functionality. Source: Shopware Product Documentation Ecosystem & Community The Shopware Store contains over 4,000 extensions. Quality varies considerably. The agency landscape in the DACH region is strong with over 1,200 certified partners, but fragmented. Technology partner integrations exist with common systems such as SAP, Microsoft Dynamics, as well as best of breed tools like Algolia, Nosto, Klaviyo. The open source community is active, and regular Community Days foster local anchoring. Source: Shopware Store, Partner Directory Licensing & Pricing Logic The tiered editions model includes the free Community Edition (open source, self hosted), Rise from approx. EUR 600/month, Evolve from approx. EUR 2,400/month, and Beyond with individual enterprise contracts. Additional costs for plugins, hosting and agency services apply. Total costs depend heavily on the degree of customisation. Compared to commercetools or Salesforce Commerce Cloud, licence costs are significantly lower. Source: Shopware Pricing, as of January 2026 Roadmap Signals Communicated development priorities: AI Copilot for content creation and admin support, Spatial Commerce for AR/VR product presentation, extended B2B Suite, Headless Improvements, Sustainability Features. May 2026 update: Shopware 6.7.9.0 (released on 17 April 2026) introduces a native Agentic Commerce sales channel as a built in component. The platform's posture is shif

Shopware Deep Dive: Vendor Claims vs. Reality Check

:::module orientation Shopify is the world's most widely adopted e commerce platform and, with the introduction of the Universal Commerce Protocol (UCP), a central player in the Agentic Commerce space. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Simplicity is not a lack of depth. It is the result of focus. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience Shopify positions itself as "Commerce Infrastructure for Everyone", claiming to serve both small merchants and large enterprises. The platform emphasises simplicity and fast time to market while maintaining scalability. According to vendor information, the target audience ranges from solo entrepreneurs and D2C brands to enterprises with Shopify Plus. This range requires different products within the ecosystem. Source: shopify.com, as of January 2026 Architecture & Delivery Model Shopify is a pure SaaS platform. There is no self hosted option. The entire infrastructure is operated by Shopify and automatically updated. The checkout system is a central differentiating factor. Shopify controls checkout to ensure security, performance, and conversion optimisation. Customisations are limited, expanded with Plus. Hydrogen and Oxygen: Shopify's headless solution enables decoupled frontend development with a React based framework. Hosting occurs via the Oxygen infrastructure. Source: Shopify Developer Documentation Core Scope & Product Logic Native functionality includes Product Information Management, integrated Content Management with Online Store 2.0, Cart and Checkout (with Shop Pay), Order Management, Customer Management, Inventory Management, integrated Payments (Shopify Payments), and Marketing and Analytics. What Shopify delivers natively but with limitations: B2B functionality (complete only with Plus), multi store setups (requires separate shops or Markets), complex promotions (app dependent). Source: Shopify Product Documentation Ecosystem & App Store The Shopify App Store contains over 8,000 apps. Quality varies, but Shopify's review process is stricter than on other platforms. The partner ecosystem comprises over 40,000 Shopify Partners worldwide. The density of specialised agencies is higher in English speaking markets than in the DACH region. Technology partner integrations exist with Klaviyo, Gorgias, Recharge, and many other best of breed tools. Source: Shopify App Store, Partner Directory Licensing & Pricing Logic The tiered model includes Basic from EUR 27/month, Shopify from EUR 79/month, Advanced from EUR 289/month, and Plus from approx. EUR 2,300/month with individual enterprise contracts. Additionally, transaction fees apply when not using Shopify Payments. App costs and agency services are additional. Compared to commercetools or SAP, licensing costs are significantly lower. Compared to Shopware, they are at similar levels with different feature sets. Source: Shopify Pricing, as of January 2026 Roadmap Signals Communicated development priorities: Universal Commerce Protocol (UCP) as foundation for Agentic Commerce, Sidekick AI for shop management, expanded B2B functions, Hydrogen 2.0 improvements, Shop Pay expansion. The UCP announcement in January 2026 positions Shopify as central infrastructure for agent driven commerce. May 2026 update: With the Shopify Agentic Plan, Shopify is for the first time systematically opening up agentic distribution to brands beyond its own platform. Products can be syndicated into AI conversations and purchased there without running a conventional Shopify online store. Wider context is provided in the AI Commerce Status Report Q2 2026. Source: Shopify Newsroom, Unite 2025, UCP Announcement January 2026, Shopify "Agentic Plan", 2026 Reference Types Highlighted customer types: Global brands like Allbirds, Gymshark, Heinz, D2C brands like MVMT and Kylie C

Shopify Deep Dive: Vendor Claims vs. Reality Check

:::module orientation Salesforce Commerce Cloud (formerly Demandware) is one of the most established enterprise commerce platforms in the market and part of the comprehensive Salesforce ecosystem. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement An ecosystem is not an argument. It is a dependency that must be understood. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience Salesforce positions Commerce Cloud as the central commerce platform within the Salesforce Customer 360 ecosystem. The claim: seamless connection of commerce with CRM, marketing, service, and data platform from a single source. According to vendor information, the target audience comprises mid market and large enterprises in the B2C and B2B segments. The promise is a unified customer view across all touchpoints, driven by native integration with Sales Cloud, Service Cloud, Marketing Cloud, and Data Cloud. Source: salesforce.com, as of January 2026 Architecture & Delivery Model Commerce Cloud is a pure SaaS platform. The infrastructure is fully operated by Salesforce. Two technically distinct product lines exist: B2C Commerce (formerly Demandware) and B2B Commerce (formerly CloudCraze), which are built on different technical foundations. Composable Storefront (formerly PWA Kit) enables headless frontend development based on a React based architecture. This is Salesforce's response to the trend toward decoupled frontends and follows MACH principles (Microservices, API first, Cloud native, Headless). The Commerce APIs provide programmatic access to catalogue, cart, checkout, orders, and customer data. API coverage has improved significantly in recent years but remains behind specialised API first platforms in depth. Source: Salesforce Developer Documentation, Composable Commerce Architecture Core Scope & Product Logic Native B2C functionality includes Product Information Management, Content Management via Page Designer, Cart and Checkout, Order Management (as a separate product), Customer Management, Promotions and Campaigns, integrated Search and Merchandising, and Einstein AI for personalisation. In B2B: Account based pricing, customer specific catalogues, contract pricing, approval workflows, quick orders, and self service portals. What Salesforce delivers natively but with limitations: Order Management is a separate, paid product. Full Unified Commerce requires additional licences. Marketplace functionality only through partners. Source: Salesforce Commerce Cloud Product Documentation Ecosystem & AppExchange The Salesforce AppExchange is the largest enterprise app marketplace with thousands of extensions. Commerce specific apps are a subset thereof. The partner ecosystem includes global system integrators such as Deloitte, Accenture, and Capgemini, as well as specialised commerce agencies. The density of qualified implementation partners is high in the enterprise segment, lower in the mid market. Native integration with other Salesforce Clouds (Marketing Cloud, Service Cloud, Data Cloud) is the central ecosystem advantage. For companies already using Salesforce, integration complexity is substantially reduced. Source: Salesforce AppExchange, Partner Directory Licensing & Pricing Logic The pricing model is primarily based on Gross Merchandise Value (GMV), the revenue processed through the platform. Typical rates range from 1% to 3% of GMV, depending on edition and negotiation. B2C Commerce Starter from approx. 1% GMV (limited to 1 site, 2 price books). B2C Commerce Growth from approx. 1 2% GMV (up to 5 sites, 10 price books). B2C Commerce Plus from approx. 2 3% GMV (unlimited sites, advanced AI features). B2B Commerce Starter and Advanced follow similar GMV based models. Additionally, costs for Order Management, Marketing Cloud, Service Cloud, and Data Cloud apply if used. The total cost

Salesforce Commerce Cloud Deep Dive: Vendor Claims vs. Reality Check

:::module orientation Adobe Commerce (formerly Magento) is one of the most established e commerce platforms in the market. What began as an open source project has become part of the Adobe Experience Cloud following the acquisition by Adobe. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Flexibility is not a feature. It is an obligation that must be funded. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience Adobe positions Adobe Commerce as an experience driven commerce platform for mid market and large enterprises seeking maximum customisability combined with the power of the Adobe Experience Cloud. The claim: technical flexibility at enterprise level, paired with native integration into Adobe's marketing, analytics, and content ecosystem. The platform addresses both B2C and B2B scenarios on a shared codebase. Adobe emphasises the ability to natively support complex catalogue structures, customer specific pricing, and multi storefront architectures. Magento Open Source remains available as a free variant but offers a significantly reduced feature set. Source: business.adobe.com, Adobe Commerce Product Documentation, as of January 2026 Architecture & Delivery Model Adobe Commerce is available in two variants: as Managed Cloud (Adobe Commerce on Cloud Infrastructure, formerly Magento Commerce Cloud) and as a self hosted on premise variant. The cloud variant runs on AWS and includes hosting, automated deployments via Fastly CDN, and an integrated CI/CD pipeline. The platform is built on PHP (Zend/Laminas Framework) with a MySQL/MariaDB database and Elasticsearch/OpenSearch for search. The architecture is modular and allows deep customisation through an extensive module system. GraphQL and REST APIs enable headless scenarios. Adobe has introduced App Builder (based on Adobe I/O Runtime), a serverless extensibility framework that enables customisations outside the core, thereby reducing upgrade conflicts. Source: Adobe Commerce Developer Documentation, Adobe Experience League Core Scope & Product Logic Native functionality includes catalogue management with complex product types (configurable, bundle, grouped), native B2B module with company accounts, contract pricing, request workflows, and shared catalogues, Page Builder for visual content management, Live Search (AI powered via Adobe Sensei), Product Recommendations (AI powered), customer segmentation and targeting, multi store, multi language, and multi currency natively, staging and preview for content changes. Adobe Commerce includes, compared to Magento Open Source, additionally: B2B functionality, advanced segmentation, Visual Merchandiser, Adobe Sensei AI features (Live Search, Product Recommendations), and cloud hosting. Source: Adobe Commerce Feature Comparison, Adobe Experience League Ecosystem & Adobe Marketplace The Adobe Commerce Marketplace offers thousands of third party extensions. Quality varies considerably, as the marketplace historically stems from the Magento ecosystem and encompasses both enterprise extensions and community modules. The partner ecosystem is global and includes large system integrators (Accenture, Deloitte, Merkle) as well as specialised agencies. The community base from the Magento era remains active but has shifted since the Adobe acquisition. Parts of the community have migrated to Shopware, Shopify, or headless native platforms. Integration with the Adobe Experience Cloud (Adobe Experience Manager, Adobe Analytics, Adobe Target, Adobe Real Time CDP) is the strategic ecosystem advantage for companies already invested in the Adobe ecosystem. Source: Adobe Commerce Marketplace, Adobe Partner Directory Licensing & Pricing Logic Adobe Commerce's pricing model is not publicly available and is based on individual quotes. Typical factors include annual revenue (GMV), the

Adobe Commerce Deep Dive: Vendor Claims vs. Reality Check

:::module orientation SCAYLE is an enterprise commerce engine that emerged from the technological infrastructure of Hamburg based fashion retailer ABOUT YOU. What was developed as an internal system to operate one of Europe's fastest growing online retailers has been offered as a standalone B2B product since 2022. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Retail DNA becomes platform product. The question is whether the origin becomes an advantage or a limitation. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience SCAYLE positions itself as the fastest growing enterprise commerce engine in the world, built by retailers for retailers. The claim: a platform that understands the complexity of large retail businesses because it was born from precisely that complexity. The platform primarily addresses B2C scenarios in the upper mid market and enterprise segment, with a particular focus on fashion, lifestyle, and retail. SCAYLE emphasises the combination of API first architecture and pre composed enterprise functionality, which is intended to enable faster time to market than purely headless native platforms. Source: scayle.com, SCAYLE Company Communications, as of February 2026 Architecture & Delivery Model SCAYLE is built on a MACH architecture (Microservices, API first, Cloud native, Headless). The platform operates as a SaaS solution, with hosting entirely managed by the provider. Customers bear no infrastructure responsibility. The Commerce Engine forms the core and encompasses catalogue, shopping cart, checkout, order management, and customer management via APIs. Additionally, SCAYLE offers a Storefront Accelerator, a pre configured frontend framework based on Nuxt.js, designed to accelerate launch without precluding the flexibility of a fully custom frontend. API coverage includes REST and GraphQL. The admin panel (SCAYLE Panel) provides a browser based user interface for operational teams. Source: scayle.dev Developer Documentation, SCAYLE Technical Architecture Core Scope & Product Logic Native functionality includes catalogue management with complex product structures and variant logic, promotions and voucher engine, checkout with express payment options and peak load stability, order management for multi channel scenarios, multi shop, multi language, and multi currency natively, Storefront Accelerator as a pre configured frontend, integrated search and filter functionality, customer management with segmentation. SCAYLE emphasises performance under load, referencing the processing of over 12,000 orders per minute, 100% uptime, and over 67 million total orders processed. These figures relate to the entire platform including ABOUT YOU's own shops. Source: SCAYLE Product Documentation, scayle.dev Ecosystem & Marketplace SCAYLE offers an add on marketplace with over 115 extensions covering integrations for payment, logistics, analytics, marketing, and further areas. The partner network includes specialised agencies and system integrators; the number of certified partners is growing but remains modest compared to established platforms such as Shopware or Shopify. The ecosystem benefits from its retail origin: many integrations have emerged from real operational requirements and are correspondingly battle tested. At the same time, the relative youth of the platform as a B2B product means the ecosystem breadth is not yet comparable to established market leaders. Source: SCAYLE Add on Marketplace, SCAYLE Partner Directory Licensing & Pricing Logic SCAYLE's pricing model is not fully publicly documented. Industry sources describe an individual licensing model typically based on transaction volume and feature scope. The enterprise positioning suggests entry costs above standard SaaS solutions like Shopify but below the licence costs of Adobe Commerce or

SCAYLE Deep Dive: Vendor Claims vs. Reality Check

:::module orientation BigCommerce is a US based SaaS e commerce platform headquartered in Austin, Texas. Founded in 2009, it serves merchants ranging from SMBs to large enterprises and positions itself as an Open SaaS alternative to closed ecosystems. Bain Capital took the company private in 2024 for roughly $6.7 billion. This article offers no recommendation. Its purpose is to provide structure, drawing a clear line between vendor claims and independent assessment. ::: :::module statement Open SaaS as a promise. The real question is whether that openness holds up in practice as convincingly as it does in the positioning. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience BigCommerce bills itself as the leading Open SaaS e commerce platform for growing and established brands. The central proposition is straightforward: combine the flexibility of an open system with the operational ease of a managed SaaS solution. The platform caters to both B2C and B2B scenarios, placing particular emphasis on merchants who need headless commerce architectures and multi storefront operations. BigCommerce Enterprise is aimed at mid market and larger organisations with specific requirements around API access, multi storefront management, and customisable checkout flows. A notable differentiator the platform highlights is its zero transaction fee policy, setting it apart from Shopify. Source: bigcommerce.com, BigCommerce Company Communications, as of February 2026 Architecture & Delivery Model The platform is built on an API first architecture that supports both hosted storefronts (Stencil themes) and fully headless frontends. Hosting is entirely managed by BigCommerce as part of the SaaS delivery model. Since 2024, BigCommerce has offered Catalyst, a modern headless storefront framework built on Next.js and React Server Components. It is available as an open source project on GitHub and integrates with the visual editor Makeswift. The classic Stencil theme engine continues to run alongside it. API coverage spans REST and GraphQL. The GraphQL Storefront API has been steadily expanded and now underpins most headless scenarios. Operational teams interact with the platform through a browser based admin panel. Source: BigCommerce Developer Documentation, catalyst.dev Core Scope & Product Logic Out of the box, BigCommerce offers catalogue management supporting up to 600 SKUs per product, multi storefront from a single instance, native B2B capabilities (price lists, customer groups, quote management), an integrated promotions engine, checkout customisation via APIs, native multi currency and multi language support, built in search and filtering, and abandoned cart recovery. The platform's zero transaction fee policy is worth underscoring: unlike Shopify, BigCommerce levies no additional fees on payment transactions, irrespective of the payment gateway used. Source: BigCommerce Product Documentation, bigcommerce.com Ecosystem & Marketplace The BigCommerce App Marketplace lists over 1,200 applications and integrations spanning payment, shipping, marketing, analytics, and beyond. A global partner network of certified agencies and technology partners complements the offering. That said, the ecosystem does not match Shopify's breadth. In Europe specifically, partner density falls short of locally entrenched platforms such as Shopware. The open API architecture does, however, allow for custom integrations that bypass marketplace dependency altogether. Source: BigCommerce App Marketplace, BigCommerce Partner Directory Licensing & Pricing Logic BigCommerce operates a tiered pricing model: Standard (from $29/month), Plus (from $79/month), Pro (from $299/month), and Enterprise (custom pricing). Enterprise rates are not publicly documented and depend on revenue volume and feature requirements. Each tier comes with revenue caps: Standard up to $50K, Plus up to $180K, and Pro up to $400K in online revenue.

BigCommerce Deep Dive: Vendor Claims vs. Reality Check

:::module orientation SAP Commerce Cloud, born from the Hybris platform acquired in 2013, ranks amongst the most established enterprise commerce solutions worldwide. Particularly in organisations with an existing SAP landscape, it is frequently regarded as the natural choice for digital commerce. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Within the SAP universe, integration is the strength. The question is whether integration alone can carry a commerce strategy. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience SAP positions Commerce Cloud as the commerce solution for organisations already anchored in the SAP ecosystem. The platform addresses B2B, B2C and B2B2C scenarios, emphasising seamless integration with SAP S/4HANA, SAP CPQ and the broader SAP Customer Experience Suite. The target audience encompasses large enterprises and the upper mid market with complex business processes that understand commerce as part of an integrated enterprise landscape. Source: sap.com, SAP Commerce Cloud Product Pages, as of February 2026 Architecture & Delivery Model SAP Commerce Cloud is built on a Java/Spring architecture with a modular structure. The platform is available as a cloud solution (CCV2, hosted on Microsoft Azure) or on premise, though SAP is actively pushing migration to CCV2. End of on premise maintenance for version 2205 is announced for 31 July 2026. The platform offers the Composable Storefront (formerly Spartacus), an Angular based frontend framework enabling headless scenarios. OCC REST APIs form the central integration layer. Source: SAP Commerce Cloud Technical Documentation, SAP Help Portal Core Scope & Product Logic Native functionality encompasses product catalogue management with complex classification systems, B2B capability including organisation management, order approvals and individual price lists, multi site and multi channel capability, integrated order management, promotions and coupon engine, search and merchandising functionality, and content management via SmartEdit. SAP emphasises deep integration with S/4HANA for real time stock information, pricing calculations and order processing. Source: SAP Commerce Cloud Feature Documentation, SAP Community Ecosystem & Marketplace SAP Commerce Cloud benefits from SAP's extensive partner ecosystem, including global system integrators (Accenture, Deloitte, Capgemini, EPAM) and specialised SAP Commerce agencies. SAP BTP (Business Technology Platform) extends customisation possibilities through standardised interfaces. The number of specialised SAP Commerce developers is declining compared with newer platforms, reflected in availability and day rates. Source: SAP PartnerEdge Directory, SAP Community Licence & Pricing Logic SAP Commerce Cloud pricing is typically embedded within the overall SAP CX agreement and individually negotiated. Costs vary considerably depending on contract scope, SAP modules in use and transaction volumes. Industry sources report annual licence costs ranging from approximately 100,000 to over 500,000 euros, depending on the overall agreement. Source: Industry analyses, SAP partner reports, as of 2025/2026 Roadmap Signals Communicated development priorities include the integration of SAP Joule (generative AI) into commerce workflows, expansion of the CX AI Toolkit for personalised customer experiences, further development of the Composable Storefront, and deeper S/4HANA integration for Unified Commerce scenarios. May 2026 update: SAP's Q1 2026 Business AI release notes (published 14 April 2026) confirm that Joule is moving from copilot to a cross suite agent framework, with autonomous agents for SAP Customer Experience entering general availability and being progressively woven into Commerce Cloud workflows. Wider context is provided in the AI Commerce Status Report

SAP Commerce Cloud Deep Dive: Vendor Claims vs. Reality Check

:::module orientation Since its founding in 2014, Spryker has positioned itself as one of the most ambitious commerce platforms in the enterprise B2B segment. In the Gartner Magic Quadrant for Digital Commerce, the company was recognised as a Leader in 2024 and as a Visionary in 2025. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement B2B commerce is not simplified B2C. It is an ecosystem with its own rules. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience Spryker positions itself as the leading composable commerce platform for enterprise B2B, marketplace and self service scenarios. The company addresses industrial companies, wholesalers and organisations seeking to digitalise their B2B sales channels and extend them with marketplace models. The positioning is deliberately B2B focused. Whilst the platform technically supports B2C scenarios, the strategic emphasis lies firmly on the specific requirements of business to business commerce. Source: spryker.com, Spryker Company Communications, as of February 2026 Architecture & Delivery Model Spryker is built on a PHP/Symfony architecture with a modular structure. The platform follows a composable approach in which over 900 modules can be independently used, customised or replaced. The core distinguishes between a backoffice layer (Zed), an API layer (Glue API) and an optional frontend layer (Yves). Since 2024, Spryker offers the App Composition Platform (ACP), enabling third party integrations without core customisation. The platform is available as PaaS (cloud hosting by Spryker) or as SaaS. Source: Spryker Technical Documentation, docs.spryker.com Core Scope & Product Logic Native functionality is optimised for B2B scenarios and encompasses company account management with organisational structures and roles, individual price lists and contract based pricing models, multi level approval workflows for orders, native marketplace module for B2B marketplaces, self service portals for reordering and account management, request for quote management, product configurators for complex items, and multi store and multi currency support. Spryker emphasises the ability to serve commerce, marketplace and self service from a single platform without requiring third party solutions for core functions. Source: Spryker Product Documentation, Spryker Feature Releases 2025 Ecosystem & Marketplace The Spryker ecosystem is growing continuously. The App Composition Platform encompasses integrations for payment, logistics, search and further areas. The partner network includes specialised B2B commerce agencies and increasingly larger system integrators. Partner density is highest in the DACH region. Internationally, the network is growing but does not yet match the breadth of established platforms. Source: Spryker Partner Directory, Spryker ACP Documentation Licence & Pricing Logic Spryker's pricing model is licence based and individually negotiated. Industry sources report annual licence costs from approximately 100,000 euros for mid sized B2B implementations. Total costs depend significantly on the scope of required modules and the degree of customisation. Source: Industry analyses, OMR Reviews, as of 2025/2026 Roadmap Signals Communicated development priorities include AI driven B2B workflows and self service portals, expansion of the App Composition Platform, enhanced marketplace functionality, improved analytics and reporting for B2B scenarios, and further development of the composable architecture. The December 2025 funding round led by TCV and One Peak strengthens the investment base for these initiatives. Source: Spryker Press Releases, Spryker Blog 2025/2026 Reference Types Highlighted customer types include industrial companies (Siemens, Bosch), automotive suppliers, wholesalers (Aldi, Metro), precision parts manufacture

Spryker Deep Dive: Vendor Claims vs. Reality Check

:::module orientation OXID eShop ranks amongst the longest serving e commerce platforms in the DACH region. Since 2003, OXID eSales AG in Freiburg has been developing a shop system that has built a loyal user base, particularly within the German Mittelstand. Compared with the major international platforms, OXID operates in a niche defined by high customisability, open source availability and deep roots in the German speaking market. This article provides no recommendation. It provides structure and precisely distinguishes between vendor claims and independent assessment. ::: :::module statement Longevity is not a quality mark in itself. It is an indication that a platform serves needs others have overlooked. ::: :::module vendor Vendor Perspective: based on publicly available information Positioning & Target Audience OXID eShop positions itself as the shop system for demanding e commerce projects with tight deadlines. The platform addresses the upper mid market in the DACH region, focusing on B2B and B2C scenarios that require industry specific customisation. OXID emphasises the combination of open source flexibility and enterprise grade functionality. The target audience encompasses retail companies, industrial firms with both end customer and dealer business, and organisations requiring a highly customisable platform without the budgets demanded by global enterprise solutions. Source: oxid esales.com, OXID Product Communications, as of March 2026 Architecture & Delivery Model OXID eShop is built on a PHP architecture that, since version 7, has transitioned to Symfony components and Composer based package management. The platform is available as an open source Community Edition, a Professional Edition and an Enterprise Edition. Since 2025, OXID additionally offers a cloud variant (OXID Cloud) as a managed hosting solution. The architecture follows a modular approach with Unified Namespace Classes, enabling structured extension of the core system. The current version 7.x supports PHP 8.1+, MySQL 5.7 / MariaDB and employs Composer as its central dependency management tool. Source: OXID eShop Developer Documentation 7.x, docs.oxid esales.com Core Scope & Product Logic Native functionality encompasses a complete B2C storefront with basket, checkout and customer account management, B2B features with company accounts, individual price lists and tiered pricing, an integrated CMS for content pages, multi shop capability with multiple storefronts from a single installation, multi language and multi currency support, a comprehensive SEO toolset with speaking URLs, a promotions and voucher system, and flexible product management with variants and selection lists. The Enterprise Edition adds extended multi tenancy, sub shops with individual configurations and granular access rights management. Source: OXID eShop Feature Documentation, oxid esales.com/en/shop system/oxid eshop features Ecosystem & Extensions OXID maintains a marketplace with several hundred modules for payment, logistics, marketing and industry solutions. The partner network comprises specialised OXID agencies, predominantly in the DACH region. The community is smaller than Shopware's or Magento's but stable and technically proficient. Industry specific solutions exist particularly for technical wholesale, pharmaceuticals, food retail and B2B distribution. Source: OXID Forge, OXID Partner Directory Licence & Pricing Logic The Community Edition is free and open source (GPL v3). The Professional Edition starts at several thousand euros per year. The Enterprise Edition sits in the five figure range annually, targeting more complex installations with elevated support requirements. OXID Cloud is offered as a managed hosting package combining licence and infrastructure. Source: OXID pricing communications, industry analyses, as of 2025/2026 Roadmap Signals Communicated development priorities include further development of OXID Cloud as a hosted solution, continuous modernisatio

OXID eShop Deep Dive: Vendor Claims vs. Reality Check

AI Commerce Status Report Q2 2026 From Promise to Division of Labour This is the second instalment in a series begun in February 2026. It now extends the stock take first published on 5 May 2026 with the decisive moves through the end of the quarter. It covers only those initiatives publicly announced or rolled out in production between 1 April and 28 June 2026 whose technical substance can actually be verified, and whose mechanics rest on genuine generative or agentic AI rather than a generously rebranded recommender system. Nothing from the Q1 Report is repeated for its own sake. Earlier themes reappear only where the second quarter has produced something materially new to say about them. An overview of all reports published so far is available on the Status Reports hub. Executive Summary The second half of the quarter has sharpened the picture considerably beyond what the May stock take suggested. Two developments now define the period, and neither sits comfortably under the old vocabulary of polite consolidation. First, the architecture of agentic commerce has visibly split in two. Discovery and transaction are coming apart, and they are increasingly operated by different parties. Walmart disclosed in March that its native conversion rate inside ChatGPT ran roughly three times below its own website, after which OpenAI quietly retired Instant Checkout in its first form. The June relaunch no longer runs on OpenAI's own payment stack but on Shopify's agentic infrastructure. Perplexity has chosen, with some deliberation, to remain a discovery layer with no transaction of its own. Shopify moved the Universal Commerce Protocol and the Catalog API into self service with Spring 2026, Salesforce acquired the AI native search engine Cimulate, and BigCommerce found itself functionally reduced to a feed provider for external discovery surfaces. Visa and Mastercard, on the same day, 10 June, took up the trust and identity layer for agentic transactions in what looked suspiciously like a coordinated manoeuvre. Europe's first production agentic payment was settled on 2 June in the Netherlands by Worldline, ING and Mastercard. Second, the first documented cull has begun. Ssense dissolved its in house photography and make up teams in June and named AI explicitly as the substitute. OpenAI's first Instant Checkout failed visibly and had to be rebuilt on someone else's infrastructure. BigCommerce is defending an agency channel position that has slid, structurally, from full stack provider towards data plumbing. commercetools posted a recovery quarter while quietly retiring the word composable and re emerging under Autonomous Commerce. The two developments are linked. Anyone who did not, in Q2, decide whether they are a platform, a product or a distribution layer will find that question answered for them in Q3, along the line that now divides discovery from transaction. The long promised breakthrough of fully autonomous consumer agents has, again, declined to arrive. What has hardened instead is a division of labour in which brands must decide separately about visibility in the discovery layer and margin control in the transaction layer. Methodological Note The overview below is drawn from publicly documented announcements, press releases, peer reviewed studies and reporting by established trade media. Only initiatives whose technical architecture stands up to inspection, and whose AI component goes meaningfully beyond rule based automation or classical recommender systems, have been included. Where figures come exclusively from vendor marketing, this is flagged. Pure repositioning without new functional substance has been left out. The Discovery and Transaction Split Becomes Architectural The split between agentic discovery and transaction, still treated as a hypothesis in the Q1 report, has firmed up into an architectural fact over the course of the quarter. On 24 March 2026 Walmart's EVP Daniel Danker disclosed that conversion inside ChatGP

AI Commerce Status Report Q2 2026: From Promise to Division of Labour

The preceding six articles in this series have described the structural shift in the digital commerce decision journey. The classical funnel loses explanatory power, stable KPIs increasingly mask opacity, a growing share of the purchasing decision forms outside measurable systems, classical attribution reaches its technical limits, new metrics must abandon the claim to certainty, and the system itself shifts its role from observation instrument to decision architecture. What follows is an operational question, no further diagnosis. What do these findings mean in the daily work of a retailer who must take decisions today on investments, roadmaps and organisational design? This closing piece of the series translates the strategic consequences into operational measures. It is deliberately not framed as a universal solution. The order of recommendations follows the structure of the series, because the consequences derive from the findings themselves; no abstract maturity model has been imposed. 1. From channel to plug in capability The first finding of the series concerned the decay of the funnel. The consequence is to shift the steering logic, well beyond any attempt to repair the funnel itself. As long as marketing and commerce organisations think in channels, they keep optimising something that plays an ever smaller role in how users actually arrive at decisions. The relevant question today is under what conditions a company's offers become part of a machine pre selection in the first place. The route into a particular channel's funnel matters less than it once did. In operational terms, this requires a controlled break with three deeply held assumptions. The first assumption is that visibility equals reach. In agentic systems, visibility is replaced by structural plug in capability. The second assumption is that every additional platform represents another channel. In reality, each platform defines its own set of requirements for data structure, identifiers and interfaces, which need to be served selectively. The third assumption is that campaign optimisation is the lever that produces growth. Today, growth is increasingly created in a layer that sits well upstream of the campaign. Concretely, this means marketing teams must begin to shift part of their attention away from channel budgets and towards data plug in. A first concrete step is to take stock of which platforms today already access content agentically. These include at minimum the Google Shopping Graph with over 50 billion listings as the backbone of AI Overviews and Gemini, the OpenAI Merchant Program with the Agentic Commerce Protocol specification,<sup 1</sup the Perplexity Merchant Program,<sup 2</sup and the Shopify Catalog MCP interface.<sup 3</sup A second measure is a deliberate decision on which of these platforms is genuinely decision relevant for the relevant category, and which are knowingly not served because their operational relevance does not justify the complexity. This decision can be prepared today through three reproducible methods, each offering a different angle on platform relevance. First, a systematic mapping of AI referral traffic. In GA4, a dedicated channel called "AI Assistants" can be configured, grouping hostnames such as chatgpt.com, perplexity.ai, gemini.google.com, copilot.microsoft.com and claude.ai as referrer sources.<sup 14</sup Since many AI platforms do not send standard compliant referrer headers, and mobile app traffic frequently lands in the direct category, this mapping should be supplemented by a quarterly assessment of so called dark traffic on product detail pages.<sup 10</sup A meaningful shift in the share of direct visits with clear product intent on categories that are frequently recommended in agentic systems is a robust indicator that a given platform genuinely produces volume for the assortment in question. Second, a category specific Share of Model test. This involves defining a set of 30 to 50 typical purchase i

Strategic Consequences for Commerce Leaders in the Age of Agentic Commerce

With its so called Agentic Plan, Shopify is extending its role in E Commerce in a manner that may, at first glance, read like yet another sales channel, but whose consequences reach considerably further. For the first time, it becomes possible to feed product data into the Shopify Catalogue independently of the shop system in use, making those products available to AI driven search and recommendation engines without migrating the existing E Commerce infrastructure.<sup 1</sup This development is frequently described as a pragmatic solution for merchants keen to benefit from the growing significance of AI powered product search. In reality, it represents a structural intervention in the architecture of digital commerce, one that has rather less to do with additional visibility and rather more with the question of who, going forward, controls access to the customer. The Shopify Catalogue as the New Central Layer At the heart of the Agentic Plan sits not the shop, but the Shopify Catalogue. It functions as an aggregated data space in which product information is structured, enriched and made available for access by external systems. Whilst Shopify has historically been perceived primarily as a platform for transactions and shop operations, this role now expands to include an upstream function that sits closer to the point at which purchasing decisions actually form. The logic is straightforward. AI systems do not need shops. They need structured, consistent and comparable product data. Whoever controls this data layer controls not merely the presentation of products, but also their inclusion in algorithmic decision making processes. How the Agentic Plan Works Technically The technical implementation has been kept deliberately low threshold, which explains a considerable part of its strategic impact. Merchants can transfer their product data into the Shopify Catalogue via feeds, APIs or imports, where it is normalised, structured and prepared for consumption by AI systems.<sup 2</sup This is not a classical system integration, but a decoupling of the data layer from the transaction layer. The shop system remains entirely unchanged, whilst product data is fed in parallel into a second system that functions as a distribution layer for AI driven discovery. Once integrated, this data becomes accessible via standardised interfaces to systems such as ChatGPT, Microsoft Copilot, Google AI Mode, the Gemini App and the Shop App.<sup 3</sup The merchant no longer integrates each platform individually, but delivers data to a central hub that handles the distribution. Shopify also offers a Knowledge Base App through which merchants can control how their brand and product content is surfaced in AI chats. Merchants can see what questions buyers are asking about their products and provide answers on returns, shipping and other topics.<sup 4</sup [AGENTIC PLAN FLOW] The Universal Commerce Protocol (UCP) A key building block behind the Agentic Plan is the Universal Commerce Protocol (UCP), which Shopify co developed with Google as an open standard. UCP defines how AI agents transact with merchants, thereby creating a shared infrastructure for agentic commerce.<sup 5</sup The significance of this protocol becomes apparent through its support base. Walmart, Target, Etsy, American Express, Mastercard, Stripe and Visa have all positioned themselves as partners.<sup 5</sup UCP is already powering checkout in select cases within Google AI Mode and the Gemini App, and is expected to extend to certain Meta experiences shortly, enabling buyers to complete purchases with a single tap.<sup 3</sup For Shopify, UCP provides a dual strategic safeguard. On the one hand, the open protocol strengthens its position as an infrastructure provider. On the other, it reduces dependency on individual AI platforms by establishing a neutral transaction standard. Availability and Rollout The Agentic Plan was first announced on 11 January 2026<sup 1</sup and has been globall

Shopify Agentic Plan: The Grab for Product Data

The preceding articles in this series have shown where the classical e commerce funnel loses explanatory power, why stable KPIs increasingly mask opacity, where purchasing decisions form before measurable contact, where attribution breaks technically, and which new metrics hold up under these conditions. This article takes up the question that follows from these findings, namely how an e commerce system must be designed to remain effective in shaping decisions when it can no longer fully map the decision process. The answer proceeds from a fundamental shift in perspective, namely the transition from the system as observation instrument to the system as decision architecture. The changed position of the system within the decision process An e commerce system, in most cases, no longer sits at the beginning of the purchasing decision. The growing share of selection, evaluation and pre selection that takes place outside the system, before a user interacts with the shop at all, fundamentally alters its role. The system increasingly decides not whether a product is considered, but whether an already formed decision is confirmed or discarded. This shift can be demonstrated empirically. An nShift survey from early 2026 found that 58 per cent of consumers have already replaced traditional search engines with AI powered alternatives for product discovery.<sup 1</sup Shopify's own data show that AI agent driven orders on the platform grew 15 fold year on year in 2025, whilst AI referral traffic grew only eightfold over the same period.<sup 2</sup The difference is instructive. Agents are functioning not merely as a discovery channel but increasingly as a purchase channel, in which the decision has already been structured before the shop visit. For system design, this carries a clear consequence. Additional tracking logic does not improve decision relevance when the decisive impulses originate outside the system. What is required instead is an architecture designed to operate under uncertainty and to exert targeted influence on decisions by rigorously shaping the few points at which it becomes effective. Observation system or decision architecture The decisive distinction lies in whether a system is understood primarily as an observation instrument or as a decision architecture. An observation system attempts to capture interactions as completely as possible and to derive optimisation approaches from them. Its logic rests on the assumption that complete data produce better decisions. This assumption was tenable for as long as the greater part of the decision process took place within the system. A decision architecture, by contrast, is designed to shape the conditions under which decisions are made, regardless of whether every individual factor is measurable. It focuses less on fully reconstructing the past and more on deliberately shaping probability. The concept is not new. Richard Thaler and Cass Sunstein coined the term "choice architecture" in their 2008 book "Nudge," describing the systematic design of decision environments to make certain outcomes more probable without restricting freedom of choice.<sup 3</sup What has changed is the context. In e commerce, choice architecture has hitherto been concerned primarily with the design of product pages, basket and checkout. Today, the task extends beyond one's own system, because the upstream decision phases are increasingly shaped by external systems. Product data as a prerequisite for visibility in the extended decision space The quality and structure of product data have always mattered. What is new is that they no longer represent merely an internal optimisation factor but have become a prerequisite for visibility in a decision space that extends well beyond one's own shop. A product that is clearly structured, semantically unambiguous and contextually embedded within a system has a higher probability of being correctly interpreted by upstream systems and included in decision process

Decision Architecture in E-Commerce

The previous article in this series showed where attribution breaks at the technical level. This article takes up the question that follows: if the scope of classical measurability has contracted permanently, what can an online retailer actually measure when traffic arrives from AI search engines, and what remains beyond reach? The answer is more layered than the current discourse tends to suggest. Some things are measurable. Some are interpretable. And some remain, for the foreseeable future, genuinely invisible. The task is to separate these three categories cleanly, because confusing interpretation with measurement produces decisions built on ground rather less solid than it appears. During the writing of this article, Stefan Wenzel's "Agentic Commerce" was published, which addresses several of the metrics discussed here (Share of Answer, Crawl to Referral, AI Referral Traffic). At various points in the text, Wenzel's assessments are therefore referenced directly where they allow a comparison with the independently researched primary sources. The aim is not a review of the book (a separate review can be found here), but a transparent cross section that brings perspectives together and differentiates where the assessments diverge. What GA4 actually shows today Google Analytics 4 can, in principle, capture AI referral traffic, though not automatically and not completely. In its default configuration, a visit from ChatGPT appears as a referral from "chatgpt.com," provided the user clicked a link embedded in the ChatGPT interface. Perplexity shows up as "perplexity.ai." Gemini traffic, however, is frequently absorbed into regular organic traffic because the referrer structure within Google's own ecosystem is not cleanly separated. The first limitation concerns channel grouping. GA4's default setup assigns AI referral traffic to the generic "Referral" category, where it mingles with every other referral source. Anyone wishing to see the AI share in isolation must create a custom channel group and manually assign the relevant domains (chatgpt.com, perplexity.ai, claude.ai, copilot.microsoft.com and others). The configuration takes a few minutes. In practice, most shops have not done it, which means the actual AI share remains invisible in standard reporting. The second limitation is more consequential. An analysis by Loamly covering 446,000 visits found that over 80 per cent of traffic genuinely originating from ChatGPT arrives at the shop without referrer information.<sup 1</sup These visits land in GA4 as "Direct Traffic" or "(not set)" and are indistinguishable from regular direct visits. The reason lies in how users interact with AI systems. A considerable proportion do not click the link embedded in the AI response. Instead, they copy the product or brand name, open a browser and search independently on Google. In GA4, this visit appears as organic search or branded search. The actual trigger remains invisible. The Visibility Labs study, covering 94 e commerce shops and twelve months of GA4 data, puts the measurable ChatGPT referral share at roughly 1.5 per cent of non branded organic revenue, rising to 2.2 per cent in the second half of 2025.<sup 2</sup The study itself notes that the true influence of ChatGPT on revenue is likely considerably higher, because the detour through branded search is systematically uncaptured. What GA4 shows, then, is a fragment. It captures the portion of AI traffic that arrives with a clean referrer. That portion is growing rapidly (1,079 per cent year on year growth in ChatGPT visits according to Visibility Labs) but remains a small slice of the actual influence. The conversion rate question and its complications Among the most frequently cited data points in the current discourse is the observation that traffic from AI search engines converts better than conventional organic traffic. The figures, however, vary rather more than one might wish. Visibility Labs reports a conversion rate of 1.81

New Decision Metrics

Agentic Commerce: Between Structural Contribution and Conceptual Overreach

The earlier articles in this series traced the structural shift. This article changes perspective and examines the same erosion from the technical side, where it manifests in specific, documentable fractures running through the entire attribution infrastructure. The answer, it turns out, is more technically specific and more immediately consequential than the abstract framing might suggest. Where attribution breaks, technically Attribution rests on a deceptively simple premise: that a sequence of touchpoints can be assigned to a single individual. In practice, this requires a set of technical conditions that have been quietly disintegrating for the past several years. The most consequential disruption was Apple's introduction of App Tracking Transparency in April 2021. Since iOS 14.5, apps must obtain explicit permission before tracking user behaviour across other apps and websites. Global opt in rates have settled at roughly 15 to 25 per cent, which means that between 50 and 65 per cent of all iOS conversions have become invisible to pixel based tracking systems.<sup 1</sup The impact was immediate and, for anyone relying on platform reported performance data, rather alarming. Meta reported revenue losses running into billions of dollars, not because fewer conversions were occurring but because fewer could be attributed to the campaigns that had triggered them.<sup 2</sup Reported ROAS collapsed by 30 to 50 per cent for many advertisers, even where actual commercial performance had barely shifted. The second tectonic change concerns the demise of third party cookies. Chrome, commanding roughly 65 per cent of global browser market share, has repeatedly delayed full deprecation whilst progressively restricting the technical infrastructure for cross site tracking.<sup 3</sup Safari and Firefox have blocked third party cookies for years already. The result is a landscape in which cross session and cross site attribution is possible for a diminishing share of users, and that share continues to shrink. The GA4 problem When Google Analytics 4 introduced Data Driven Attribution in 2023, the promise was straightforward: an algorithmic model that would distribute conversion credit across touchpoints more intelligently than rigid last click or rule based approaches. Reality has proven rather more interesting. A growing number of practitioners report that GA4 has inflated the revenue attributed to Google Ads by 50 to 100 per cent within a single year, quite independently of whether actual sales have grown at anything approaching that rate.<sup 4</sup The explanation lies in the model's mechanics: it tends to award disproportionate credit to channels for which it holds the richest data, and that channel is, with a certain inevitability, Google's own advertising platform. Simultaneously, the proportion of traffic classified as "(not set)" or "Unassigned" in GA4 has been growing steadily. These are not edge cases. They represent user flows that the system can no longer attribute, whether due to missing cookies, blocked referrer information or interrupted session chains. In some accounts, they now account for 15 to 30 per cent of total traffic, a figure that rather undermines the confidence one might wish to place in the remaining data. Cross device and the identity gap A further structural fracture runs along device boundaries. A user who researches a product on her smartphone in the evening and orders it from a desktop the following morning will, in most systems, appear as two entirely separate people unless she is logged in on both occasions. GA4 attempts to address this through Google Signals and User ID, but both require either a Google login or proprietary authentication. For a typical e commerce shop, where the majority of visitors browse without logging in, cross device attribution remains fragmentary at best. The consequence is that customer journeys appear shorter and simpler in the data than they actually were, because their const

The Loss of Attribution

Notes from the Digital Floor 7 Why Online Retailers Are Going to Prison Now. "Don't go to prison voluntarily!" Alexander Graf advises retailers in his recent brand eins interview<sup 1</sup , referring to Google's Universal Commerce Protocol. The image is memorable, the warning clear. And that is precisely the problem, because what initially reads like a precise diagnosis of the market unfolds, on closer inspection, into a dynamic where distinct levels are collapsed into one another in ways that generate urgency where differentiation would be required first. The interview addresses virtually every relevant trend in the industry simultaneously, from agentic commerce through Chinese platforms to the transformation of the working world in online retail. The individual observations are not unfounded, but the manner in which they are connected produces a picture that appears considerably more coherent than reality warrants. What follows is an attempt to examine the central claims individually and to sharpen the focus where generalisation obscures the actual state of affairs. UCP as prison, but who isn't already inside? The core thesis of the interview holds that UCP strips retailers of all agency and leads them into total dependence on Google. As a warning, that is understandable. As analysis, it overlooks a central fact, for the overwhelming majority of retailers already find themselves in precisely this dependence. Anyone selling on Amazon today, reliant on Google Ads or purchasing visibility through Meta, is already operating within structures they do not control. UCP shifts this asymmetry. It does not create it. The more interesting question is what UCP actually changes. The protocol is designed so that the agent (in this case Google's Gemini) handles the entire purchase process, from product search through comparison to payment. The retailer formally remains the seller (Seller of Record) but loses all contact with the customer, and with it the ability to build relationships, collect data or influence repurchase. That is a meaningful loss of control. But it primarily affects those retailers who have never built a direct customer relationship in the first place, and there are a great many of those. The real question, therefore, is not whether UCP is dangerous, but for whom it is worth resisting, and for whom the ship has already sailed. The urgency that isn't Graf himself concedes that agentic commerce will likely become relevant "in five to ten years rather than two to three."<sup 1</sup At the same time, he advises waiting "until the very last possible day."<sup 1</sup These statements sit in a remarkable tension with one another, for if the time horizon is genuinely that distant and the customer need still unclear, where precisely does the acute urgency lie? Small and medium sized retailers without clear differentiating features or customer advantages face diminishing viability in digital commerce with or without UCP. This is not a new phenomenon and not a symptom of a protocol unveiled in early 2026. It is the result of a concentration that has been intensifying for years, in which platforms control the customer interface and retailers without their own brand, exclusive assortments or demonstrable customer value systematically become interchangeable. That customers increasingly buy directly from brands and manufacturers via marketplaces is likewise not a UCP specific development. This trend has been measurable on Amazon for years and is being actively driven by direct to consumer strategies from numerous manufacturers. UCP may accelerate this tendency. It does not cause it. Equally undifferentiated is Graf's claim that customers only want "cheap, cheap, cheap" and that a single euro of markup is grounds for abandonment.<sup 1</sup This is a generalisation that, in its sweeping form, is not supported by the evidence. A study by the Qualtrics XM Institute found that 72 per cent of consumers are willing to pay more for a better

Notes from the Digital Floor 7. Why Online Retailers Are Going to Prison Now.

Most purchasing decisions have already begun long before they show up in anyone's analytics. The previous articles in this series made that case in structural terms. What they did not do, however, is spell out what that invisible phase actually looks like when you follow it through real scenarios, step by step, and hold it against the data a retailer would typically see on their dashboard. That is what this article sets out to do. Three scenarios, each increasingly common, each structurally quite different, and each producing the same curious result: a customer journey that looks perfectly legible in the data whilst concealing the most consequential part of the decision. Scenario 1: The AI assisted pre selection A woman is considering a high end espresso machine. She does not begin with a Google search. Instead, she opens ChatGPT and types something along the lines of: Which espresso machine would you recommend for home use if I care about temperature stability and quiet grinding? Within seconds, she receives a tidy comparison of three or four models, contextualised and ranked. No advertisements, no sponsored placements, no discernible commercial agenda. It reads, for all intents and purposes, like disinterested advice, which is precisely what makes it so influential. She makes a mental note of two models, opens a browser, types one of the names into Google and lands on a retailer's product page. A brief comparison, a glance at the reviews, and the order is placed. The retailer's analytics tell a reassuringly familiar story: a user arrived via brand search, visited two product pages and converted. Short journey, clear intent, healthy conversion rate. Nothing to investigate. What the analytics do not tell is that the decision was effectively settled before the first click occurred. The shortlist was assembled by a system the retailer has no visibility into and no means of influencing. Every product that failed to appear in that AI generated response was eliminated from contention before the journey even began, irrespective of how polished its product page might have been or how sharp its pricing. Scenario 2: The social imprint A man follows several interior design accounts on Instagram. In one of them, a particular kitchen appliance appears in the background of a story, not as an advertised product but simply as part of the scenery. A few weeks later, a TikTok video surfaces in his feed: someone cooking with the same appliance, mentioning in passing that it has served them well for two years. At no point has this man actively searched for a kitchen appliance. There is no conscious purchase intent. Yet something has quietly taken root: a familiarity with a specific brand and a specific product, one he would struggle to articulate if asked about it directly. Months later, when his existing appliance finally expires, he does not search for "best kitchen appliance 2026." He searches for the brand name. He visits the shop, confirms the price, and orders. The retailer's dashboard paints a flattering picture: direct traffic or brand search, rapid conversion, low bounce rate, modest acquisition cost. By every available metric, this is a textbook loyal customer with strong brand affinity. In truth, that affinity was not cultivated by the retailer's marketing department. It was built, gradually and incidentally, across platforms the retailer neither monitors nor controls. The retailer can see the outcome but has no reliable account of how it came about, which means there is no way to engineer it again deliberately. Scenario 3: The recommendation behind closed doors An entrepreneur needs a new e commerce platform for her shop's relaunch. She posts the question in a private Slack community where roughly two hundred fellow D2C founders exchange notes. Within two hours, seven replies arrive. Four of them recommend the same platform, each with personal experience and specific assessments attached. Another platform, one that would objectively

The Invisible Phase of the Purchasing Decision

Most e commerce organisations have spent recent years in determined pursuit of measurability. Tracking has been sharpened, attribution models layered with ever finer granularity, and dashboards expanded until they resemble the cockpit of a mid range airliner. The ambition was clear enough: decisions should rest on evidence, not instinct. By that standard, the mission has been accomplished. There is scarcely an area of digital business as meticulously instrumented as online retail. And yet something rather awkward is happening. The data keeps getting more precise whilst the understanding of what it actually represents keeps getting less so. Stability as a misleading signal Glance at the central metrics and there is, at first sight, little cause for alarm. Traffic moves in comprehensible patterns, conversion rates respond to the usual levers, and campaigns deliver measurable returns. The systems behave exactly as one would expect them to. It is precisely this composure that ought to give pause. Stability suggests control. It implies that the underlying system remains legible and, by extension, steerable. Yet stability is not proof of completeness. It can just as easily indicate that only a portion of what is actually happening is being captured, whilst the rest unfolds quietly beyond the field of view. The shift beyond the measurement boundary The critical development lies not within the measured interactions but before them. As the previous article described, a growing share of decision formation is migrating into territory that classical instrumentation simply cannot reach. The consequence for measurement is considerable: the user who enters the measurable part of the process is no longer, in many cases, a searcher. They arrive as someone who has already narrowed the field, discarded alternatives and formed expectations elsewhere. When measurement and causation part ways Classical analytics models rest on an implicit assumption: that cause and effect can be traced along the same observable chain. A click produces a visit, the visit produces an interaction, and the interaction, in due course, produces a conversion. Causality, in this view, is something that can be read directly from the sequence of events. That logic holds up rather well, right up to the point where substantial parts of the decision form outside the chain altogether. Consider a user who first encounters a brand through an AI generated recommendation, returns weeks later via a generic search query and eventually converts through a direct visit. In the data, this appears as a perfectly legible sequence of familiar touchpoints. What it does not reveal is that the decisive moment, the one that placed the brand on the shortlist in the first place, occurred in a space the system never observed. The measurement remains consistent. The causality, however, has quietly left the building. The return of seemingly familiar patterns This shift produces a side effect that is worth pausing over. Channels that had been written off as declining are regaining prominence in the data: direct traffic rises, brand searches increase, returning visitors appear more stable than they have in years. At first glance, one might interpret this as vindication of existing efforts. As the fourth article in this series will show in technical detail, it frequently turns out to be an artefact, a statistical echo of decisions that were formed elsewhere and merely executed within the measurable system. Why more data does not solve the problem The instinctive response to uncertainty is to measure more: to deploy additional tools, refine attribution yet further, collect another layer of data points. It is a natural impulse and, within certain limits, a reasonable one. The difficulty is that it addresses the wrong constraint. The problem is not a shortage of data within the system. It is the fact that a meaningful share of decision relevant information now originates outside the system entirely. Increa

The Illusion of Measurability

The e commerce funnel is one of those constructs so thoroughly woven into the fabric of digital business that questioning it feels faintly impolite. Awareness, consideration, conversion, retention: these terms organise dashboards, justify budgets, shape team structures and underpin entire reporting architectures. They have been around so long that most organisations treat them less as a model and more as a description of reality itself. Which is, of course, precisely the problem. A model that explains the world it built The funnel emerged in an era when e commerce decisions unfolded, for the most part, within controllable and measurable systems. A user noticed an advert, visited a website, added something to the basket and bought it. The path was linear, the touchpoints countable, the attribution pleasingly traceable. That world has not vanished. It has simply become a smaller and smaller fraction of what actually happens. Purchasing decisions now form, with increasing frequency, in spaces that classical tracking cannot see. They take shape in conversations between colleagues, in messaging groups, in the tangential observations of a podcast, in comment threads beneath content that has nothing to do with commerce, in the recommendation logic of platforms that were never designed for shopping, and in the synthesised responses of AI powered search systems that summarise, evaluate and contextualise product information without a single click ever reaching a retailer's page. By the time a user first appears on a product page, the decision is not beginning. It is, in many cases, already substantially made. The structural shift What has changed is not the funnel's existence but its explanatory reach. Within the boundaries of measurable systems, it still describes events accurately enough. It simply has less and less to say about everything that happens before those boundaries. This shift shows up in several places at once. The attention phase has splintered across dozens of channels and formats that no central media plan can govern. What once sat neatly under "awareness" now disperses through organic social dynamics, algorithmic recommendations and word of mouth in closed groups where no tracking pixel has ever set foot. The consideration phase has migrated, quietly but decisively, into environments that e commerce systems cannot observe. Users compare products in contexts that lie entirely beyond instrumentation. They form preferences before they so much as open a search engine. The conversion phase, meanwhile, grows shorter and more predetermined. A visitor arriving on a product page has, in a growing number of cases, already made their choice. The funnel, in this scenario, captures only the execution of a decision whose formation it never witnessed. Why the model survives regardless The funnel's persistence owes less to its analytical precision than to its organisational convenience. It structures teams, allocates budgets and provides a shared vocabulary between marketing, sales and the executive floor. It makes messy, nonlinear decision processes feel manageable by arranging them into tidy phases. That organisational function is genuine and valuable. It should not, however, be mistaken for analytical validity. A model can be thoroughly useful for running an organisation whilst simultaneously offering an increasingly incomplete picture of reality. It is precisely this gap that is widening. The consequence for e commerce steering If the funnel describes a shrinking portion of the actual decision process, the steering instruments built on top of it gradually lose their explanatory power. Metrics such as conversion rate, cost per acquisition and return on ad spend remain perfectly serviceable for evaluating individual channels. They simply answer, with diminishing reliability, the rather more important question of why customers buy in the first place. A retailer who optimises exclusively against funnel metrics is optimising for the vi

The Quiet Decay of the E-Commerce Funnel

AI Commerce Status Report Q1 2026 Reality, Readiness and the Gap Between Hype and Execution This piece is intended as a status report rather than a manifesto. It draws solely on publicly documented and verifiable initiatives and is meant to provide a calm, factual reading of where AI in a commerce context actually stands at the start of 2026, as opposed to where keynote slides would prefer it to be. Executive Summary By early 2026 artificial intelligence in digital commerce has comfortably outgrown the speculative phase. It now sits inside numerous initiatives, pilot programmes and a respectable number of genuinely productive applications. The accompanying public discourse, by contrast, has not aged quite as gracefully. The vocabulary of the moment, "Agentic Commerce", "AI First", "Autonomous Shopping", is deployed with admirable confidence and rather less precision, often blurring the line between what is shipping today, what is being built, and what remains a slide in someone's strategy deck. This report tries to put that line back where it belongs. It covers only publicly documented and verifiable use cases, alongside initiatives whose technical maturity is described in terms one can actually scrutinise. Hypothetical scenarios and uncosted visions are left, politely, for another occasion. The picture that emerges is consistent. Several productive building blocks now exist that tend to be filed under "Agentic Commerce", chiefly conversational interfaces, in chat checkout flows, and new commerce surfaces accessible through AI platforms. These mark a structural shift in how customers interact with commerce systems. They do not, however, amount to autonomous purchasing agents operating at scale. The agents most often invoked in conference programmes are still, for the most part, somewhere between proof of concept and press release. The applications in production today rest almost entirely on explicit human intent. AI supports search, selection and checkout within new surfaces, but does not make purchasing decisions on its own account. The genuine progress lies less in machine autonomy than in the decoupling of interface from commerce logic. API based architectures allow new modes of interaction to be plugged into existing transaction systems without rebuilding the underlying decision making fabric. Publicly disclosed initiatives from major retail and fashion businesses underline this. AI platforms are increasingly used as additional sales channels or as dialogue led experience layers. In parallel, large groups continue to invest in AI for decision support across product development, demand forecasting, inventory and marketing. Useful, scalable, occasionally transformative, but pursuing rather different objectives from the vision of fully autonomous commerce agents. What is striking is that the bulk of the current momentum is happening not at the customer touchpoint but at the level of plumbing. Across the industry, work is under way on protocols, standards and authorisation mechanisms intended to let AI systems execute transactions safely and within the law. Payment providers, platform operators and the larger marketplaces are quietly investing in identity, fraud detection and approval workflows. These are preconditions for greater autonomy rather than autonomy itself, and they tend not to feature on opening keynote slides. The contrast between rhetoric and operational reality is hard to miss. Where keynotes and op eds favour a binary in which one is either "AI First" or shortly to be irrelevant, the documented use cases tell a more textured story. Progress is incremental, scoped to clearly defined applications, and shaped by regulatory, organisational and technological constraints that have not been suspended for the occasion. The sober conclusion, for organisations actually running commerce businesses, is that the central task is not to deploy autonomous agents but to build the conditions under which new forms of interaction

AI Commerce Status Report Q1 2026: Reality, Readiness and the Gap Between Hype and Execution

Update: Strategic Pause on Agentic Commerce Experiments (March 2026) Since this article was first published in January 2026, the dynamics within the Agentic Commerce ecosystem have shifted noticeably. In early 2026, several early experiments involving direct purchasing and agent driven commerce interactions within ChatGPT were paused or scaled back. The reasons lie less in technological shortcomings than in structural maturity gaps across the broader ecosystem: missing identity layers, unresolved consent questions, and still insufficient infrastructure on the merchant side. At the same time, open protocols such as MCP have continued to evolve, and the first platform vendors are positioning themselves more clearly. The fundamental assessment in this article remains valid, though these developments lend it additional relevance. A comprehensive analysis of the current state can be found in the AI Commerce Status Report Q1 2026. For a detailed comparison of the protocols underpinning Agentic Commerce, read UCP vs ACP. Agentic Commerce in E Commerce: A Calm Reality Check for Online Retailers in 2026 Anyone currently involved in artificial intelligence in e commerce as an online retailer will inevitably come across the term "Agentic Commerce". Hardly any other buzzword is currently being discussed, predicted and interpreted as frequently. Autonomous AI agents that compare products, make purchasing decisions and trigger orders are considered by some to be the next evolutionary stage in online retail. For many retailers, however, this development seems less inspiring than unsettling. This uncertainty does not arise primarily from the technology itself, but from the speed and polarisation of the public debate. Between visionary images of the future, marketing narratives and technical detail discussions, there is often a lack of objective classification. This article aims to provide precisely that: a calm, fact based consideration of where Agentic Commerce actually stands at the beginning of 2026 and what this means in concrete terms for online retailers. Threads this article unfolds Why Agentic Commerce currently creates more unrest than orientation Where we actually stand in early 2026 and what is already reality How decision making logic is quietly shifting in online retail What role control and discovery will play in the future Who will actually use Agentic Commerce in the next 12–24 months Which retailers need to act sooner – and which later Why composure can currently be a strategic attitude The Current State of Agentic Commerce at the Beginning of 2026 Agentic Commerce is not a sudden upheaval and not a short term disruption of the classic online shop. Rather, an additional infrastructural layer is gradually developing between consumers, AI systems and retailers. This layer enables AI agents to understand structured information, evaluate purchasing options and technically execute transactions correctly. Major payment and technology providers such as PayPal are already investing in agent capable checkout and catalogue interfaces. In parallel, open standards and protocols are emerging that define how products can be made discoverable, how identities are verified and under what conditions a purchase may take place. A significant proportion of these standards are still under development, although some components are already in productive use. What is particularly important for retailers is this: the role of Merchant of Record remains intact. Prices, margins, payment processing, business rules and customer relationships remain with the retailer. Agentic Commerce changes the access routes, not the economic responsibility. What is Quietly Shifting in Online Retail The actual change brought about by Agentic Commerce is less spectacular than many headlines suggest. It is not about replacing human purchasing decisions, but about supplementing existing decision making logic. While classic e commerce optimisation focuses heavily on visibilit

Agentic Commerce in E-Commerce – A Sober Assessment for Online Retailers (2026)

Across many e commerce organisations, a new kind of uncertainty is quietly taking hold as several simultaneous shifts reshape the digital landscape. Established traffic patterns are gradually dissolving, search engines are beginning to generate direct answers rather than mere lists of results, and conversational interfaces are gaining increasing prominence in how users interact with digital systems. And yet, it remains far from clear what concrete steps retailers ought to be taking in response to these developments. Between alarmism and denial, what is frequently missing is a structural framework. This is precisely where the concept of Generative Engine Optimization comes in. This article provides that framework, describes the mechanics behind generative discovery, and outlines realistic measures that retailers can implement today without overhauling their entire architecture. What Is Generative Engine Optimization? Generative Engine Optimization describes the optimisation of digital content and commerce structures for generative answer systems that synthesise, weight, and present information as a basis for decision making. Unlike traditional search engine optimisation, the focus shifts away from ranking positions towards referenceability. Content within generative systems is found, interpreted, combined with other sources, and placed into entirely new contexts. The principle can be described in fairly straightforward terms. Whilst traditional SEO aims to appear as high as possible in an ordered list of results, Generative Engine Optimization aims to flow into synthesised answers as a trusted source. The difference between these two approaches is a fundamentally structural one. Search engines in their established form operate on an indexing model. Content is crawled, evaluated, and placed into a hierarchy. Position within that hierarchy determines visibility. Generative systems work differently. They reconstruct information from diverse sources and produce new, composite answers. The decisive question thus shifts from indexation towards whether content serves as a reference point within the synthesis. For retailers, this represents a shift in optimisation logic. It becomes less about occupying individual keywords and more about building substantive content, structure, and consistency that generative systems classify as trustworthy and citable. How Generative Discovery Actually Works Generative systems reconstruct decision spaces. They aggregate content from different sources, recognise patterns in reviews, product descriptions, and structured data, and produce answers or recommendations from this material. Several factors play a role in this process. Structured data forms the machine readable foundation upon which generative systems identify relationships between different pieces of information. The depth of content signals expertise and increases the probability of being drawn upon as a reference source. Coherence between sources, meaning the consistency of information across channels and platforms, strengthens the credibility of any sender. Brand authority, understood as the sum of mentions, reviews, and consistent positioning over time, influences how strongly a source is weighted in the synthesis. And finally, the consistency of product information across different touchpoints determines whether an offering is perceived as reliable enough to include in generated recommendations. Generative engines are, in this sense, less index machines than synthesis machines. Their logic follows interpretation, and their weighting is guided by trustworthiness and relevance within the given context. The implications are considerable. A product page that merely lists technical specifications may well be indexed, but it is rarely synthesised. A product page, by contrast, that answers decision oriented questions, enables comparisons, and is embedded within a coherent informational context has a substantially higher probability of becoming pa

Generative Engine Optimization in E-Commerce: Why Visibility Is Reorganising

Update: Strategic Pause on ChatGPT Shopping (March 2026) Since this article was published in October 2025, the landscape has shifted considerably. In early 2026, the direct purchasing experiments within ChatGPT described here were paused or scaled back. The reasons lie less in technological shortcomings than in structural maturity gaps across the broader ecosystem. A detailed assessment of this development can be found in the AI Commerce Status Report Q1 2026. Shopping via ChatGPT How Agentic Commerce is Changing the Rules of Online Retail A silent revolution is taking place: ChatGPT is becoming a shopping platform. What initially looks like a feature update marks the beginning of a new era in e commerce Agentic Commerce. What is Agentic Commerce? Agentic Commerce describes a paradigm shift: No longer does the customer search for products, but AI agents act on their behalf. ChatGPT is the first mass market player implementing this concept. How Does Instant Checkout Work? 1. Conversation : Users describe their need 2. Recommendation : ChatGPT suggests suitable products 3. Checkout : Direct purchase processing without website visit 4. Delivery : Processing through connected merchants The Impact on E Commerce End of the Click Journey? Traditional e commerce lives from the customer journey: Awareness → Consideration → Purchase. Agentic Commerce skips these phases. Consequences: SEO becomes less relevant Brand awareness loses importance Product quality and reviews become more critical Price transparency increases New Rules 1. Visibility in AI Models The question is no longer "Do I rank on Google?" but "Does ChatGPT know me?" 2. Structured Product Data AI agents need perfectly structured data to recommend products. 3. Integration in AI Ecosystems Merchants must be integrated into AI platforms not just present on their own websites. Opportunities for Merchants First Mover Advantage Merchants entering partnerships with AI platforms now secure advantages. Efficiency Increase Agentic Commerce can dramatically increase conversion rates as purchase barriers fall. New Target Groups AI affine users are often affluent and early adopters. Challenges Loss of Control Merchants have less control over product presentation. Margin Pressure Increased price transparency can put margins under pressure. Dependency New dependencies on AI platforms emerge. How to Prepare 1. Optimize Data Quality Invest in structured, high quality product data. 2. Build Partnerships Explore integration opportunities with AI platforms. 3. Prioritize Reviews Authentic customer reviews become the main factor. 4. Perfect Fulfillment Delivery reliability becomes even more critical. 5. Adapt Strategy Develop an Agentic Commerce strategy parallel to traditional e commerce. The Future of Online Retail Agentic Commerce will not replace traditional e commerce, but complement and transform it. Merchants must excel in both worlds. For a deeper analysis of the protocols behind Agentic Commerce, read our article UCP vs ACP: Which Protocol Will Shape Agentic Commerce in 2026?. A sober assessment for online retailers can be found in our Agentic Commerce reality check. Glossary ACP (Agentic Commerce Protocol) A protocol introduced by OpenAI and Stripe for standardised commerce interactions within AI interfaces, focusing on rapid integration through predefined schemas. Agentic Commerce Commerce scenarios in which AI agents research, compare, and initiate transactions on behalf of consumers or businesses, with degrees of autonomy ranging from assisted selection to independent purchasing decisions. In Chat Checkout Purchase completion within a conversational AI interface without redirection to a conventional shop website. UCP (Universal Commerce Protocol) An open standard initiated by Google for communication between AI agents and commerce systems, featuring dynamic capability negotiation between agents and merchants.

Shopping via ChatGPT - How Agentic Commerce is Changing the Rules of Online Retail

Notes from the Digital Floor 6 The Half Life of Certainty Certainties have a remarkably short shelf life on the digital floor. At the start of the year, the matter seemed settled. Anyone in e commerce who failed to embrace artificial intelligence immediately, and more specifically Agentic Commerce, had already missed the boat. At conferences, on panels and across LinkedIn, the message was delivered with a conviction that brooked no dissent. The train was leaving. If you were still on the platform, you might as well tear up your ticket. What was remarkable was not the vehemence so much as the precision, or rather, the absence of it. In a surprisingly large number of presentations, it remained unclear what Agentic Commerce actually meant. Autonomous purchase decisions? AI powered assistants? New modes of interaction between platforms and users? Or simply a fresh name for an old hope: that technology might elegantly dissolve the complexity of commerce. None of this stopped anyone from passing the message along with full conviction. New vocabulary enters the stage, and with it the certainty that this time, everything will be different. The mechanism is ancient. Then a few weeks passed. And suddenly, a rather different piece of news took the stage. ChatGPT announced that it would, for now, be stepping back from its initiative around direct purchases within the platform. A strategic decision, soberly worded. In the digital echo chamber, however, it was an event that immediately produced new certainties. Now other voices declared that Agentic Commerce had been a misunderstanding all along. What digital commerce really needed to focus on was what customers actually wanted: better prices. The formula has an unbeatable advantage. It is simple. It always works. And it is technologically far less demanding. Between these two positions, the overheated vision and the swift disillusionment, lies the real spectacle of the digital floor. Topics are rarely developed slowly. They are driven across the stage, one certainty replacing the next before the previous one has even been properly assessed. It is worth pausing, occasionally, to consider some of the more famous lines from the history of innovation. Henry Ford once observed that people, when asked what they wanted, would probably have said faster horses. Steve Jobs put the same thought differently. People often do not know what they want until they see it. The point is less the genius of individual entrepreneurs than a rather human pattern. We imagine improvements within familiar frames. Transformations of the frames themselves tend to appear unnecessary, incomprehensible, or simply too complicated. Perhaps that is precisely why the digital floor is so susceptible to dramatic changes of direction. Between euphoria and disillusionment, there are often only a few weeks, sometimes nothing more than a new presentation at an industry conference. Those who spend enough time there eventually notice a reassuring detail. The truly interesting developments rarely follow this zigzag course. They tend to emerge more quietly, more slowly and considerably less spectacularly than the certainties that accompany them. The half life of a conviction on the digital floor, then, is a matter of weeks. The changes that actually endure take considerably longer.

Notes from the Digital Floor 6: The Half-Life of Certainty

OXID eShop is one of those platforms that quietly held its ground in German speaking e commerce for years. Mid market merchants and brands chose it when they wanted tight control over their commerce architecture whilst retaining the flexibility to build bespoke extensions. Many of these installations have been running reliably in production for years. At the same time, the environment in which they operate has changed considerably. Integration requirements are rising, innovation cycles are shrinking, and commerce is increasingly becoming the centrepiece of a broader digital architecture. Against this backdrop, many OXID users face a strategic question that is less about technology and more about organisation: is the path towards OXID Version 7 worth pursuing, or does a platform change become the more realistic long term option? The Historical Role of OXID in German Speaking E Commerce OXID eShop established itself over many years as a platform for mid market merchants and brands. Particularly in Germany, Austria and Switzerland, an ecosystem of specialised agencies, extensions and integration solutions developed around it. Typical deployment scenarios included mid market brands with their own direct to consumer operations, B2B commerce with complex product structures, and merchants with heavily individualised shop processes. The decisive advantage lay in the combination of a stable platform foundation and considerable freedom for custom modifications. Many organisations used OXID not merely as shop software, but as a central building block within their commerce architecture. The Technical Reality of Many Existing Installations In practice, many OXID installations today are the product of years of incremental development. Extensions, custom modules and integrations were typically built up over extended periods. Typical characteristics of such installations include tight ERP integration, bespoke business logic within the shop, and tailored extensions for specific business models. This architecture can function reliably for many years. At the same time, it frequently makes larger structural changes considerably more difficult. The more heavily a shop has been customised, the more complex subsequent modernisation steps become. What Version 7 Changes With Version 7, OXID pursues the goal of technically modernising the platform and aligning it more closely with current development standards. This includes architectural improvements, modernised development tooling, and adaptations to current PHP and framework versions. For existing installations, however, an upgrade rarely means a straightforward version bump. In many cases, custom extensions need to be adapted or reimplemented. The actual effort therefore depends heavily on how extensively a shop has been customised over the years. Headless Suitability and Frontend Decoupling One of the most frequently raised questions in any platform evaluation today concerns the ability to decouple the frontend from the backend. OXID eShop was historically designed as a monolithic system in which template engine and business logic are tightly interwoven. With Version 7, OXID has taken steps towards a more modern API architecture, yet a fully headless capable setup in the sense of an API first approach is not part of the proposition. Organisations wishing to run their storefront via an independent frontend framework (React, Vue, or a PWA solution, for instance) encounter structural constraints within OXID that typically require considerable custom effort. For organisations whose commerce strategy relies on a strict separation of presentation and business logic, this architectural characteristic becomes a material decision factor. Platforms such as commercetools, Shopware (with its Frontends initiative), or Shopify (with Hydrogen) offer conceptually different starting positions here. Omnichannel and Point of Sale Capability Another aspect of growing importance is a platform's ability to orchestrate

OXID eShop: Upgrade to Version 7 or Strategic Platform Change?

This article was published as a guest post on shapeandshift.dev. Below is a summary of the key arguments. What It's About The term headless is frequently deployed as an innovation signal in e commerce, yet behind the architecture decision lies far more than a technical question. At its core, it concerns the deliberate separation of frontend and backend, connected through API based communication. What sounds like flexibility simultaneously introduces new dependencies, cost structures and organisational demands. The Key Findings The guest post analyses headless architectures from multiple perspectives: Architectural Reality: Headless rigorously separates frontend and backend layers. This enables technological freedom, but also creates considerable coordination overhead between teams, deployment processes and testing infrastructure. Total Cost of Ownership: Initial costs represent only a fraction of the total bill. Over three to five years, significant expenditure arises from framework migrations, security patches, performance optimisation and DevOps. In many mid market projects, ongoing costs substantially exceed the original estimates. Organisational Maturity as Key Factor: The strongest predictor of headless success is not the quality of the initial implementation but the presence of a dedicated technical product owner who understands both business requirements and architectural constraints. Decision Matrix: The article includes a practical matrix that maps organisational maturity against technical complexity. The conclusion: only when high maturity meets high complexity does headless become strategically compelling. AI Perspective: The growing importance of AI in e commerce adds a new dimension to the headless discussion. Yet many AI use cases, from product recommendations to dynamic pricing, operate at the API level and are entirely independent of frontend architecture. The Author's Conclusion Headless is neither silver bullet nor wrong decision. It is a tool whose impact depends on context, organisation and objectives. Those who choose it should know why. Those who don't should be equally able to justify their reasoning. For an analysis of the pitfalls lurking in partner selection for complex E Commerce projects, see our article on unrealistic project proposals. For a vendor independent comparison of leading platforms, explore our System Deep Dives. 👉 Read the full guest post on shapeandshift.dev

Guest Post: Headless in E-Commerce, Architecture Decision or Organisational Question?

Notes from the Digital Floor 5 The Quiet Architect Public discourse tends to belong to those who articulate most sharply, compress most swiftly and appear most frequently. But spend enough time around commerce and you encounter a different figure, one who rarely takes the stage and even more rarely generates headlines. You recognise them not by their reach but by the quality of their questions. The quiet architect cares less about trends than about what will actually hold up. They listen to buzzwords without rushing to adopt them. They examine what an organisation genuinely brings to the table. They understand that every system decision reaches into existing processes, ties up budgets and shifts cultural dynamics. Whilst others talk about scaling, they think in integration steps. Whilst panels debate disruption, they study data models, interfaces and lines of accountability. They are less interested in what is theoretically possible than in what can be run reliably under real world conditions. Their work is seldom spectacular. It consists of alignments, iterations, occasionally the deliberate choice to slow down. They know that technological innovation without organisational readiness creates friction rather than progress. They know that client trust is not built on visions but on processes that actually work. The quiet architect steers clear of grand promises. They place recommendations in context. They spell out risks with the same clarity as opportunities. And they accept that sound decisions need time to show their worth. In the age of cognitive infrastructure, this disposition matters more than ever. Systems grow more capable whilst decision making grows simultaneously more complex and more tightly interwoven. It is precisely for this reason that we need people who understand structures before they reach for the accelerator. The quiet architect does not court applause. They seek fit. Their authority rests not on volume but on reasoning you can trace. Those who work with them often notice the difference only in hindsight: projects unfold with less drama, transitions proceed with less panic, expectations arrive more clearly defined. Perhaps it is this kind of professionalism that remains too seldom visible in the discourse. It generates neither viral quotes nor existential warnings, yet it leaves behind systems that work. In a time when attention is cheaply won, reliability remains a genuinely scarce resource.

Notes from the Digital Floor 5: The Quiet Architect

Notes from the Digital Floor 4 The Consultant as Brand The consulting world has changed visibly in recent years. Expertise alone no longer opens doors. Visibility has become a second currency. LinkedIn profiles now read like miniature enterprises. Fractional titles, podcasts, keynotes, advisory boards. Consulting increasingly presents itself as a brand. On the face of it, this is neither surprising nor objectionable. In a competitive market, differentiation is necessary. Attention determines who gets invited, recommended or commissioned. It becomes interesting where visibility and substance start pulling in different directions. The market rewards clarity, speed and unambiguous messaging. Complexity sells rather less well. A nuanced "it depends" rarely wins a pitch. A pointed "this is what you must do now" frequently does. Consultants are not immune to these incentives. Winning mandates means promising orientation. Promising orientation tempts one to tidy away uncertainty before it has been properly understood. Between market logic and intellectual honesty runs a rather fine line. Then there is the proximity to technology ecosystems. Vendor partnerships form part of many business models. They provide access, resources, project volume. They also shape perspective. Every platform arrives with its own narrative. Those who are closely embedded inevitably absorb parts of that logic. The trouble arises not from any individual but from the structure itself. When visibility, urgency and ecosystem proximity converge, the discourse drifts. Solutions appear clear cut more quickly than they prove to be in practice. Recommendations land with more force than the realities of some organisations can bear. For leaders, this landscape is hard to read, and for those earlier in their careers, considerably harder still. Between panels, posts and podcasts, an atmosphere emerges in which any restraint looks like falling behind. Anyone wishing to pause must first justify themselves. Perhaps the real task of responsible consulting today lies in withstanding precisely this dynamic. Not every organisation needs the same tempo. Not every technology demands immediate scaling. Not every platform makes sense under all conditions. Consulting begins where it shares risk. Where it argues with nuance even when nuance draws less applause. Where it has the courage to advise against a project when the foundations are not yet in place. Visibility can be a sign of competence. It is not, however, proof of it. In the end, one simple question remains: does the recommendation serve the client, or the narrative in which everyone involved is currently swimming? Those who answer that question honestly change the discourse more quietly, but more lastingly, than any stage.

Notes from the Digital Floor 4: The Consultant as Brand