We are halfway through the year. Adobe’s Digital Trends 2025 set bold expectations for AI, data, and customer experience. What actually happened, and what should leaders do next?
At Exelement, we see two parallel realities. AI initiatives are already creating value in specific processes. At the same time, many companies still struggle to scale results beyond isolated teams. The difference often comes down to the maturity of the operating model. Clean data, shared ways of working, and a clear view of the customer determine whether AI becomes a real advantage, or just another tool everyone talks about.
Our senior Marketo consultant Fredrik Stokke highlights three takeaways from Adobe’s AI predictions and how they show up in real work.
1. AI Is Moving from Concept to Contribution
Adobe’s research shows that the most mature use cases are in customer support and chat. These are already delivering measurable ROI for a meaningful share of practitioners. Leaders expect near-term benefits to include:
- Better quality interactions
- More consistent communications
This aligns closely with what we see in the field.
Beyond support, AI-assisted personalization and content generation are growing fast. Adoption remains uneven, with many teams still in experimentation mode. And independent surveys back this up. McKinsey reports a rising share of companies that can point to revenue uplift from generative AI, with service operations among the first to show financial impact.
What This Means for You
Move your AI work away from one-off experiments and into named processes with clear owners and service level expectations. For marketing and RevOps, this usually means:
- A governed pipeline for AI-generated content tied to brand, compliance, and performance checks.
- Predictive scoring reviewed weekly with sales leadership and tuned against actual conversion and cycle time.
- Assisted service flows where AI handles classification, summarization, and suggested replies while agents own the conversation end to end.
2. Disconnected Data Systems Limit AI Performance
Customers expect consistent experiences across channels and greater control over their data. Brands cannot deliver that without a unified data strategy and real-time capabilities. Adobe’s customer engagement analysis highlights consistency as a core expectation.
The tools are already there. Customer data platforms (CDPs) unify records, improve data quality, and activate audiences. Without identity, consent, and event data in sync, every AI use case will underperform.
Salesforce’s State of Marketing report points in the same direction: high-performing teams invest in a unified data strategy and use that foundation to scale personalization.
What This Means for You
- Prioritize the data. Your personalization engine is only as good as the inputs.
- Define a golden customer profile and make it available in the tools where people work.
- Set clear SLAs for identity resolution, consent capture, and event freshness.
- Treat data quality as a product: with an owner, a roadmap, and a budget.
3. Technology Needs Human Judgment and Shared Ways of Working
AI can point to the next best action, but people still need to make the right call. The companies that turn signals into results redesign workflows and decision rights as they deploy AI. According to McKinsey & Company, senior ownership of governance and cross-functional processes is a common success factor.
Adobe’s 2025 announcements underline another shift: we are moving from AI assistants that suggest work to AI agents that execute work inside enterprise systems. This raises the bar for governance, testing, and exception handling. Teams need clear guardrails to trust what the system does on their behalf.
What This Means for You
- Codify how insights become actions.
- Create joint rituals where marketing, sales, and success review AI insights together and commit to experiments.
- Define human-in-the-loop checkpoints for anything that changes a customer record, triggers a message, or adjusts commercial terms.
- Measure the full loop, from model output to customer outcome, and retire use cases that do not move a business metric.
A Practical Midyear AI Checklist
Use this test to see if you are getting real leverage from AI right now:
- You can show before-and-after results for at least two use cases where AI improved a business metric such as conversion rate, cycle time, or cost to serve.
- Your customer profile is unified across channels and is available in the tools where marketers, sellers, and service agents work.
- You have a content operating model that governs prompts, brand standards, approvals, and performance feedback.
- Leaders meet in a standing forum to review AI insights, agree on experiments, and assign clear owners.
- You can explain how consent and privacy rules flow through your AI stack. If not, pause new use cases and fix that first.
What This Looks Like in Marketo
Here are three Marketo-centered plays we implement to turn insight into action:
Predictive Lead and Account Prioritization
Use predictive scoring to identify leads and buying groups that are likely to move this week. Pipe the list to sales with short context for a first contact. Tune the model against stage progression and closed won, not clicks or opens.
AI-Assisted Personalization at the Moment of Engagement
Combine your CDP segments with Marketo audience rules and web personalization. Serve content variations based on intent and account signals. Use human review for strategic pages, and let the system adapt lower-risk assets inside clear guardrails.
Service-to-Growth Loops
Feed support insights into lifecycle programs. When Adobe and the broader market point to chat and support as early ROI zones, treat service data as fuel for retention, expansion, and content planning.
So, Did the Predictions Come True?
Yes, in the areas that depend on well-structured processes and clean data. Not yet in organizations that expect technology to overcome fragmentation or unclear ownership. The most visible gains are in support and operations. The next wave is personalization at scale: once the data foundation is in place.
How Exelement Helps
We build the operational foundation that lets you turn signals into outcomes. As a certified Marketo partner, we connect your data, set practical guardrails for AI, and align teams around shared rituals so that insights lead to action.
If you want a focused midyear audit, we can map your current stack, quantify the gaps, and deliver a 90-day plan your teams can execute immediately.
