Why Marketing's AI Progress Is Stalling at Scale - New Research from Screendragon

This is less a demand-shock for AI and more a budgeting signal that the next dollar of spend shifts from experimentation to integration. That should favor enterprise workflow platforms with distribution into the operating layer — Adobe, ServiceNow, Salesforce, Atlassian, Monday.com — because the ROI now depends on owning request/approval/measurement, not just generating content. By contrast, standalone point tools and small AI SaaS names risk slower net retention as buyers consolidate around suites that can prove cycle-time reduction and governance.

The second-order effect is on services and implementation revenue: if 80%+ of organizations still lack connected workflows, the near-term monetization likely accrues to systems integrators, digital agencies, and consulting partners rather than model providers. Over 1-3 months, this is more of a “show-me” setup than a hard catalyst; the trade only works if management teams start talking about workflow attach rates, module consolidation, and AI-driven seat expansion in upcoming quarters. If those metrics do not improve, the market will treat the AI narrative as productivity theater and compress multiples for vendors selling standalone copilots.

Contrarian view: consensus may be underestimating how quickly procurement pressure follows these surveys. Once buyers see that the value pool is in orchestration, they may pause new point-solution purchases and force vendors to bundle AI into existing contracts, which can slow ARR expansion for niche tools even if usage rises. The key falsifier is whether large software vendors report acceleration in cross-sell/attach or whether AI remains a low-ARPU feature that fails to move net dollar retention over the next 2 quarters.