Amit Xerxes & Murthy Peri
“Software to manage, deliver, and optimize digital experiences consistently across every phase of the customer life cycle.”
— Forrester Research
For information technology (IT) architects and technologists who seek to design, build, and evolve digital experience platforms, a key challenge is to establish a platform architecture that integrates a variety of best-of-breed technologies and capabilities to deliver an integrated solution. That same platform must also enable a high degree of agility, scalability, and adaptability.
Based on years of experience architecting complex, omni-channel digital experience platforms for global clients—often in partnership with Adobe and deploying Adobe Experience Manager (AEM)—we have observed a range of business drivers and architectural patterns. Here, we share our perspective on the most common business drivers and our recommendations for which architecture patterns are most appropriate for different customer scenarios.
Today’s consumer does not distinguish between channels or moments of engagement, transaction, and service. A modern IT platform supporting digital experiences must allow a seamless journey from awareness to engagement and discovery, purchase and transactions, and even customer service and support. Unfortunately, this expectation is difficult for most enterprises to meet.
Many enterprises and their IT platforms have traditionally been organized by function (e.g., Marketing, eBusiness, Customer Support), leading to unwanted silos. To deliver a unified, integrated consumer experience, many organizations are bridging this gap by adopting a unified digital experience platform (DXP) that cuts across organizational silos.
A digital experience platform delivers a unified, seamless customer experience by bridging gaps between technological layers and organizational silos.
Having worked with hundreds of marquee clients over the last 20 years, we have seen a broad array of business drivers and use cases for investing in a digital experience platform like AEM. We have explored, defined, and implemented a variety of solution patterns for how a DXP can be constructed. Every DXP solution is somewhat unique—there is no single solution or silver bullet for how these platforms should be architected.
Our experience shows four key business scenarios and use cases for a modernized DXP:
There are various ways to define a digital platform and its underlying solution architecture. Consequently, there are different design patterns that can be applied. These patterns dictate how responsibilities are assigned across the architecture, how systems and technologies are mapped to these responsibilities, and how interactions between layers and systems deliver the complete solution. Rather than advocating a one-size-fits-all solution, we recommend considering a catalog of solution patterns to be evaluated and applied to each enterprise’s unique needs.
For this paper, we focus on Adobe Experience Manager as the web content and experience management solution. We highlight key architecture patterns for building digital experience platforms around AEM and show how AEM can be deployed and extended to support various needs.
The standard AEM solution is ideal for a wide spectrum of brand marketing experiences, including corporate websites, microsites, global and local brand coordination, and multi-brand/channel/regional marketing. In this pattern, IT teams separate concerns by decoupling front-end activities from back-end activities. This enables both to evolve independently and allows front- and back-end teams to focus on their strengths.
Even with a decoupled front end, this pattern retains all the authoring and marketer capabilities available out-of-the-box with AEM. The benefit is seamless integration of the front-end experience and modules with back-end AEM components, minimizing churn and rework in the integration process through a schema- or contract-driven component development model.
By adopting a front-end view library that supports a “write once, run anywhere” approach, brands can rapidly evolve their experiences from static to interactive, personalized, and dynamic. Combined with AEM’s server-side templating, this enables a multi-tenant architecture where multiple brands can reuse a shared platform while maintaining creative independence.
The standard AEM solution also uses a “progressive rendering” model, allowing much or all of the experience or page to be rendered by AEM, reusing the same front-end view library for both server and browser containers. AEM Publish serves as the primary deployment container and presentation assembly layer. Integration needs (analytics, targeting, DAM, search, consumer data) can be handled within AEM or client-side.
When integrating browsing and shopping experiences or driving content-driven commerce, an experience-driven pattern is preferred. Here, three teams collaborate: the front-end team defines templates and user-facing elements, the AEM platform team creates editable content components, and integration developers build scalable integration services and APIs. Together, they combine marketing content and editorial capabilities with data and business applications from transactional systems, enabling a unified consumer experience with integrated experience management and delivery.
As in the previous pattern, the consumer experience and back-end systems are decoupled and evolve independently. AEM retains its authoring and marketer capabilities, enabling integrated in-context authoring, editorial tasks, and experience previews. The front-end view library allows developers to build and manage user-facing elements separately, facilitating greater agility.
AEM Publish Server remains the primary presentation assembly and composition layer. The overall experience, page layout, and marketing components can be rendered by AEM Publish, while dynamic components requiring integration (e.g., product details, booking widgets) can be rendered using server-side include technologies.
The key to this pattern is an API-first mindset. A scalable service orchestration layer integrates with back-end systems (ecommerce, CRM, booking engines, etc.), especially as integrations increase and greater orchestration is required to deliver and render the customer experience.
This pattern is a game changer compared to standard AEM or other content and experience platform deployments, due to three critical factors:
Based on microservices, this pattern is ideal for heavy orchestration in omni-channel consumer experiences at web scale, such as cross-channel B2C/B2B commerce, multi-brand/multi-store product catalogs, and high-volume order processing. It is also preferred for integrating browsing, shopping, purchasing, and customer service interactions.
This pattern requires high modularity and agility, with the ability to manage, deploy, and scale each layer independently. A cloud-native, pure microservices architecture adopts an everything-as-a-service model, identifying multiple interactions (content consumption, commerce experience management, ordering, product catalogs, promotions, loyalty) as headless services. The headless approach allows for strong decoupling and flexibility, supporting evolving channels and future-proofing the platform architecture.
Like the experience-driven pattern, the microservices architecture uses an API-first model for interfacing with underlying systems and services. This enables a highly scalable, non-blocking service orchestration and integration layer, separated from the actual microservices. The service orchestration layer is optimized for each channel, combining responses from several underlying API calls into one unified customer experience.
The overall application is supported by loosely-coupled, independent modules that follow their own development lifecycles (non-monolithic deployment). It prefers a decoupled presentation assembly layer ("decoupled glass"), where a server-side MV* (Model-View-*) application enables routing, server-side presentation logic, orchestration across APIs and back-end services, dynamic experience composition, asynchronous processing, and event-driven programming models. These activities are typically done outside the underlying systems via a Universal JavaScript application pattern, allowing the same MV* application to assemble and render the dynamic experience across server and client. The application is deployed as the decoupled glass, enabling high-speed, high-performance rendering on the server, front-end rendering in the browser, and rapid updates to the customer experience layer.
This enables rapid updates with faster release cycles and more frequent deployments, isolated from the underlying mission-critical transaction system. This "bi-modal" or "two-speed IT" approach balances experimentation and agility with stability and scalability.
Experience management needs are simpler than with the experience-driven solution. Some layout, page, and template management is still required, but is often limited to specific sections or focused on component/slot placement for merchandising.
IT leaders can share, leverage, and adapt the underlying AEM solution across multiple solution patterns. For example, the experience-driven and headless patterns can be combined using the same AEM infrastructure, making sense for enterprises already invested in Adobe solutions.
Based on business needs and technological requirements, the three patterns can be analyzed as follows:
| Criteria | Pattern 1: Standard AEM | Pattern 2: Experience-led | Pattern 3: Microservices |
|---|---|---|---|
| Separation of layers/responsibilities | Low | Medium | High |
| Decoupled presentation/front-end | Loosely coupled | Moderately decoupled | Fully decoupled |
| Isomorphic rendering support | Low | Medium | High |
| Scalability/performance for content-heavy apps | High | Medium | Medium |
| Scalability for dynamic, transactional apps | Low | Medium | High |
| Scalability for dynamic rendering of non-cached pages/apps | Low | Medium | High |
| Degree of decoupling across layers/systems | Low | Medium | High |
| Modularity, flexibility, adoptability | Low | Medium | High |
| Leverage of authoring toolset/experience management | High | High | High |
| Flexibility, speed of deployment, rapid changes | Low | Medium | High |
| Complexity of deployment architecture | Low | Medium | High |
| Operations complexity (maintenance, upgrades) | Low | Medium | Medium |
| Primary benefits | Simple deployment; scales for content-heavy apps | Integration tier behind API gateway; marketer empowerment | Highly scalable/extensible for omni-channel; high agility; cloud-native |
| Primary limitations | Poor scalability for transactional/dynamic apps | Limited by Experience Manager rendering | High complexity; requires mature DevOps, automation, cloud-native design |
Organizations worldwide are betting on digital experience platforms as the new innovation frontier and primary customer engagement vehicle. Each organization’s unique business goals and drivers are vital in determining how best to architect their solution. The business scenarios and solution patterns outlined here are meant to guide IT architects and development teams in determining which architecture makes the most sense for their situation. This catalog of patterns enables architects, consultants, and IT teams to collaborate and choose the most applicable and beneficial approach.
These patterns can be further extended, modified, and tweaked as necessary. Leading companies are planning for flexible digital experience platforms and rapidly implementing them with partners to achieve business goals.
Senior Director Engineering, Publicis Sapient India
Amit Xerxes is a Senior Director Engineering at Publicis Sapient, based in Noida, India. He has over 18 years of experience in Marketing Platforms, spanning Web Experience Management, Content Management, Digital Asset Management, Portals, Digital Experience Platforms, and Global Digital Marketing Platforms. Amit has played multiple roles, from architecting marketing platform solutions for large-scale digital platforms to leading engineering teams focused on reference architectures, frameworks, and solution accelerators. He has experience with various commercial and open-source products and was the technology lead for a major engineering initiative at Publicis Sapient, evolving capabilities around Architectures and Decoupled and Headless Experience Platform solutions. Currently, he leads the Cognitive Marketing engineering program, applying Data Sciences and Machine Learning to Content and Marketing Platforms for hyper-personalized experiences.
Group Vice President Engineering, Head of Marketing Platforms
As head of marketing platforms, Murthy advises clients on digital transformation enabled by technology and marketing cloud platforms. He is passionate about realizing enterprise value through engineering excellence. With 20 years of experience deploying complex digital solutions for Fortune 500 organizations, Murthy is a subject matter expert in marketing technology solutions spanning campaign, experience, assets, and personalization. Clients include Corteva, Samsung, Invesco, SSGA, Allianz, Analog Devices, P&G, and GiantEagle. Prior to Publicis Sapient, Murthy was Vice President of Technology at Digitas, launching an enterprise mobile platform for Aflac and designing global digital platforms for P&G brands. He previously spent 10 years at Agency.com as a technical director.
To learn more about how we partner with Adobe, contact:
Publicis Sapient