Financial services firms are redirecting their AI investments away from content generation and toward personalization, compliance, and customer analytics, according to recent findings from eMarketer. The research shows that marketing leaders in banking, insurance, and wealth management rank ad copy and content creation as low-priority AI use cases, while data-driven segmentation and customer analytics deliver the most measurable returns.
The finding runs counter to how most enterprise teams talk about AI. The dominant conversation still centers on generative tools drafting emails, producing social posts, and writing campaign briefs. But in financial services, every piece of client-facing copy carries disclosure requirements and compliance reviews that erode the speed advantage generative AI is supposed to provide.
Compliance friction redirects AI investment
The applications where financial services firms report meaningful AI returns operate upstream of compliance review, not downstream of it. A personalization engine that surfaces the right product offer to the right customer segment inside a mobile banking app does not require a legal sign-off on every output. A chatbot trained on a firm's own customer data to handle routine service inquiries clears compliance once, at build time, rather than per asset. That structural cost difference shifts where AI budgets deliver the most value.
Customer analytics occupies the same category. Financial institutions hold rich behavioral data sets transaction histories, account tenure, product mix, and service interactions. AI models built on that proprietary data generate segmentation precision that no purchased audience list can match. "The firms that will win the AI race in financial services are building personalization infrastructure, not prompt libraries," the eMarketer analysis said.
A trillion-dollar infrastructure backdrop
This strategic priority shift is unfolding against a wave of capital spending. The Wall Street Journal, citing Commerce Department data, reported that U.S. private investment in AI-related categories data centers, software, computers and peripheral equipment has surpassed $1 trillion at a seasonally adjusted annual rate. The numbers have climbed steeply since 2022 through the most recent data available as of mid-2026.
For enterprise operators, that spending wave means the vendors supplying AI platforms to financial services teams are absorbing large capital costs and passing new capabilities downstream fast. Features that required custom development two years ago are becoming standard. The Wall Street Journal also reported that tech companies are issuing billions in debt to fund AI infrastructure purchases, signaling the investment cycle can sustain itself for years. For procurement leaders at financial institutions, a platform chosen today may look significantly different in capability and pricing within 18 months.
Where operations teams should focus
The eMarketer analysis points to three application areas already delivering returns in the sector: next-best-offer personalization at the customer level, lifecycle marketing automation triggered by behavioral signals, and AI-assisted analytics for campaign attribution and audience refinement. All three are data-infrastructure plays, not creative plays.
Content generation is not irrelevant. It works well for internal uses like summarizing research, drafting regulatory comment letters, or producing first-draft RFP responses with a human editor in the loop. But content generation consistently underperforms compared to customer data and personalization use cases across the sector's AI deployments.
Why this matters for finance professionals
For marketing operations, CIO, and digital transformation leaders in financial services, the gap between where most teams have deployed AI and where the returns sit is the operative challenge. Teams that still treat AI primarily as a content accelerator are underinvesting in the workflows that drive measurable lift: customer retention, cross-sell conversion, and service cost reduction. Those metrics trace directly to personalization and analytics quality, not to content production speed.
The infrastructure wave means better tools are arriving. Earning that value requires first getting the underlying customer data architecture clean, unified, and accessible to the models that will act on it. The eMarketer assessment is a signal to audit current deployments against use-case ROI, benchmark personalization and analytics pilots against what comparable institutions have returned, and reassess vendors on 12-month cycles. An AI learning path for finance managers that prioritizes data infrastructure and compliance workflows will deliver faster returns than further investment in prompt libraries alone. For applied AI for finance teams, the biggest gains come from knowing customers better, not from writing faster.
Your membership also unlocks:
