The DXP Catalyst Update - Feb 18, 2025

INTRO
Welcome to This Week’s DXP Catalyst Update

With AI transforming the DXP landscape, some vendors are evolving into Intelligent DXPs, embedding AI deeper into their ecosystems. But how much of this AI innovation is actually delivering value? In this issue, we cut through the hype to examine where AI is making a tangible impact, where vendors are still catching up, and how AI agents are evolving within DXPs.

In the last newsletter, I explored the 2025 Gartner Magic Quadrant for DXPs, highlighting vendor positioning and strategic direction. This week, I’m taking a closer look at AI adoption across DXPs, analyzing real use cases, emerging AI agent functionality, and how vendors are shaping their AI strategies.

Let’s dive into how applied AI is driving value in DXPs - and where the industry is still navigating the complexities of AI integration.

LEADERSHIP GUIDANCE
Applied AI in DXPs: Real Use Cases & What’s Hype vs Reality

AI is being embedded into DXPs in varied ways across vendors and product ecosystems - but not all AI delivers the same impact. Some applications are already driving efficiency, automation, and personalization, while others are still evolving toward their full potential.

To better understand the landscape, I’ll break down what’s real and what’s hype across different areas of applied AI in DXPs. But before diving into specific use cases, let’s start with the fundamental distinction between Generative AI and Applied AI in DXPs - and how they work together.

Breaking Down GenAI and Applied AI in DXPs

There’s a lot of noise about AI in DXPs, but not all AI serves the same purpose - or works in the same way.

Generative AI (GenAI) is a subset of machine learning (ML) that uses deep learning techniques to create new content, including text, images, videos, and audio. While GenAI leverages natural language processing (NLP) for text-based applications (i.e. AI-powered chatbots, automated article writing), it also extends beyond language into image generation, video creation, and synthetic audio. In DXPs, GenAI powers tools for automated content creation, AI-driven asset generation, and real-time adaptations of marketing copy.

Applied AI in DXPs extends beyond GenAI, incorporating ML, predictive analytics, and behavioral analysis to optimize and automate digital experiences. This includes real-time personalization, predictive targeting, search optimization, workflow automation, and audience segmentation.

How They Work Together

Rather than being separate, GenAI and applied AI work in tandem to create and optimize digital experiences:

• GenAI generates copy, while applied AI determines which variant performs best for each audience segment.

• GenAI creates product descriptions, while applied AI refines merchandising strategies based on demand forecasts and user behavior.

• GenAI-powered chatbots generate human-like responses, while applied AI ensures conversations are contextually relevant based on past interactions.

• Hyper-personalization blends both technologies - GenAI generates tailored content, while applied AI leverages behavioral analysis and predictive models to deliver the right experience at the right time.

Challenges & Limitations

GenAI can generate content, but brand consistency, compliance, and emotional resonance still require human oversight.

Applied AI depends on structured, high-quality data - without it, predictive models can be inaccurate or biased.

Why This Matters

GenAI and applied AI aren’t competing technologies - they complement each other. GenAI enables scalable content creation, while applied AI optimizes and personalizes digital experiences based on data-driven insights. Together, they form the foundation of intelligent DXPs.

The Rise of AI Agents in DXPs: Who’s Leading the Charge?

One of the biggest trends in AI for DXPs is the emergence of AI agents - intelligent assistants embedded within platforms to automate workflows, optimize content, and enhance personalization. While many vendors are integrating AI across their ecosystems, only a few have rolled out fully-fledged AI agents:

• Adobe - Sensei GenAI & AI Assistant: Powers content generation, personalization at scale, and predictive campaign optimization across Adobe Experience Cloud.

• Optimizely - Opal: Embedded across Optimizely’s entire DXP ecosystem, Opal supports content generation, personalization, experimentation, and predictive analytics. Optimizely also stands out for offering BYOAI (Bring Your Own AI), allowing businesses to integrate their own AI models into the platform.

• Salesforce - AgentForce: While Salesforce wasn’t included in the Gartner Magic Quadrant for DXPs and isn’t typically considered a DXP-first vendor, its AI agent strategy is highly relevant when exploring innovations in this space. AgentForce extends AI-powered copilots across Salesforce’s broader ecosystem, including Experience Cloud, bringing intelligent automation to customer experience and marketing workflows.

These AI agents function as intelligent copilots, automating repetitive tasks, generating insights, and enabling teams to work more efficiently within their respective platforms. I expect to see other vendors follow suit and rollout this type of functionality.

While other vendors apply AI across their platforms - enhancing capabilities like search, recommendations, and automation - they have yet to introduce standalone AI agents. We’ll explore those applied AI capabilities in the next section.

How Should We Look at Applied AI in DXPs?

AI is transforming Digital Experience Platforms (DXPs), but its impact varies across different capabilities. To understand how AI is being applied effectively, I break it down capability by capability, examining the broader DXP ecosystem, which includes:

• Content Management (CMS)

• Commerce & Merchandising

• Customer Data Platforms (CDP)

• Personalization, Experimentation & Optimization

• Digital Asset Management (DAM)

• Search & Discovery

• Email Marketing & Automation

However, DXPs aren’t just a collection of standalone features - they function as connected ecosystems. That’s why it’s important to view AI not just within specific capabilities, but across the entire marketing lifecycle.

Optimizely’s Perspective: AI as an End-to-End Enhancement

Some vendors are transitioning toward Intelligent DXPs, where AI extends beyond individual features and products. Optimizely stands out to me for how clearly it showcases AI as a seamless, embedded intelligence layer within Optimizely One.

AI isn’t just an enhancement to one product - it enhances planning, creation, storage, personalization, and experimentation at every stage of the marketing journey. AI-powered tools work across multiple products and capabilities, while Opal Chat serves as an AI agent that enables automation, personalization, and optimization throughout the platform.

Rather than treating AI as disconnected capabilities, Optimizely frames it as an intelligence layer that weaves through the entire marketing journey. This full-lifecycle approach contrasts with vendors that apply AI more narrowly, reinforcing how AI can function as a core enabler of digital experiences, rather than just a bolt-on feature.

Sitecore hasn’t introduced a full AI co-pilot yet, but they’re heading in a similar direction with Sitecore Stream, which was announced at last fall’s Symposium event.

Applied AI Use Cases in DXPs: What’s Driving Real Value?

AI is being applied across multiple DXP capabilities, but not all AI delivers the same impact. Below is a breakdown of where AI is driving real value in content creation and automation - and where the hype still outweighs reality.

AI-Powered Content Creation & Automation

What’s Real:

• AI-assisted content generation (text, images, videos) within CMS and DAM platforms to streamline workflows.

• AI-driven tagging, cropping, and metadata generation for digital assets, improving content organization and retrieval.

• NLP-powered content recommendations and content gap analysis, helping optimize content strategies.

• Multilingual content generation - AI enhances content translation and adaptation for global audiences.

What’s Hype:

Fully automated, high-quality AI-generated content without human oversight. While AI can accelerate content production, human intervention remains essential for accuracy, compliance, brand consistency, and contextual relevance. AI is a powerful tool - but not a full replacement for editorial control.

AI in Personalization & Predictive Analytics

What’s Real:

• AI-powered real-time personalization and audience segmentation. AI dynamically adapts experiences based on user behavior, demographic data, and interaction history to deliver targeted content, offers, and recommendations.

• Predictive analytics for customer lifetime value (CLV) modeling and churn risk detection. AI helps identify high-value customers, detect early churn signals, and anticipate upsell opportunities, improving retention and revenue strategies.

• AI-driven customer journey mapping and next-best-action recommendations. AI pinpoints pain points, optimizes touchpoints, and predicts drop-off points, allowing for proactive engagement strategies.

• Granular segmentation, even for anonymous users. AI clusters users based on behavioral patterns, browser/system data, and engagement trends -enabling personalized experiences even without explicit user profiles.

What’s Hype:

AI that can “automatically understand” customers without structured data. AI models require clean, structured, and labeled data for accuracy - without it, predictions can be unreliable or biased.

AI in Commerce & Merchandising

What’s Real:

• AI-driven product recommendations. AI personalizes recommendations based on customer behavior, purchase history, and intent signals to increase conversions and average order value.

• Dynamic pricing optimization. AI adjusts pricing in real time based on demand, competitor pricing, and market conditions to balance profitability and customer satisfaction.

• Predictive inventory management and demand forecasting. While OMS, ERP, or IMS systems typically handle core forecasting tasks, AI enhances commerce platforms by enabling real-time adjustments to merchandising strategies, optimizing stock levels, and anticipating demand shifts based on dynamic customer behavior and market trends.

• AI-powered merchandising optimization using behavioral insights. AI dynamically adjusts product rankings, promotions, and category layouts based on real-time user interactions to maximize engagement and revenue.

• Omni-channel integration. AI ensures merchandising consistency across digital and physical storefronts, aligning pricing, inventory, and recommendations for a seamless customer experience.

What’s Hype:

AI replacing manual merchandising decisions entirely. AI enhances automation and efficiency but does not replace strategic oversight, creative direction, or brand positioning.

AI-Driven Search & Discovery

What’s Real:

• AI-enhanced semantic search and NLP-driven query understanding. AI interprets user intent and context beyond keyword matching, delivering more relevant results.

• AI-powered visual search. AI enables image-based search, improving product discovery in fashion, home decor, and e-commerce by identifying visually similar items.

• Predictive autocomplete and query optimization. AI refines search experiences by suggesting queries based on past behavior, location, and real-time trends.

• Federated search across multiple sources. AI unifies results from CMS, DAM, product catalogs, and other repositories into a single, cohesive search experience (i.e. Squiz’s Funnelback search).

• Contextual query expansion. AI broadens search relevance by automatically expanding user queries with synonyms and related terms.

What’s Hype:

AI replacing human expertise in search tuning. AI enhances but does not eliminate the need for human oversight in refining taxonomies, managing edge cases, and improving relevance tuning.

AI in Campaign & Marketing Email Optimization

What’s Real:

• AI-generated marketing email subject lines, campaign copy, and A/B test suggestions. AI generates content variations based on historical performance data, but human oversight remains essential for ensuring brand alignment, messaging nuance, and emotional resonance.

• AI-driven send-time optimization and engagement scoring. AI analyzes recipient behavior, past interactions, and engagement trends to optimize send times and improve open and click-through rates.

• Predictive analytics for campaign performance forecasting. AI predicts conversion rates, click-through rates, and ROI, helping marketers allocate budgets and optimize cross-channel strategies.

• Scalability for large campaigns. AI streamlines audience segmentation, automated follow-ups, and performance tracking, enabling marketers to manage large-scale campaigns more efficiently.

What’s Hype:

AI fully automating campaign management. AI enhances execution but does not replace human creativity, strategic planning, or contextual decision-making.

Final Thoughts

AI is rapidly reshaping how DXPs deliver content, personalization, commerce, and customer experiences, but it’s not a set-it-and-forget-it solution. The most effective AI applications are those that enhance decision-making, improve efficiency, and scale operations - not those that attempt to replace human expertise entirely.

As we’ve seen across content automation, personalization, commerce, search, and marketing campaigns, AI excels when paired with high-quality data, strategic oversight, and well-defined business objectives.

Looking ahead, I expect to see:

• More sophisticated AI agents - expanding beyond basic copilots into decision-support assistants embedded across DXPs.

• Greater emphasis on BYOAI (Bring Your Own AI) - allowing enterprises to train and integrate their own models for more control and differentiation.

• A growing focus on AI governance and trust - ensuring AI is used ethically, responsibly, and in compliance with data privacy regulations.

As AI capabilities evolve, one thing remains clear: AI is a force multiplier, but human strategy is still the real differentiator.

WHAT’S NEXT
Upcoming Topics

• Next Week: Key Considerations When Selecting a CDP