AI is not arriving in marketing budgets at one speed. On Publicis Groupe’s October 13 earnings call, CEO Arthur Sadoun described a “two-speed” shift: clients are moving faster to adapt marketing as consumers use AI to search and compare, while larger enterprise AI transformations are moving more cautiously. For business leaders, the distinction matters: near-term customer-facing work may be funded before a full technology overhaul.

The company’s Q3 release reported 5% organic net-revenue growth. Connected Media, which includes media, commerce, CRM and influencer marketing, grew at a high-single-digit rate; Publicis said client demand for AI-powered products and services was one driver, alongside market-share gains and new addressable markets. That is a signal from one major agency group—not proof that every marketer is reallocating budgets the same way. (Publicis Groupe’s Q3 results; MediaPost’s Q3 results coverage)

Two AI agendas are competing for the same budget

Sadoun’s explanation was direct: consumers are already using AI in ways that change discovery and comparison, so brands feel pressure to adapt quickly. By contrast, AI projects that replace or connect core enterprise systems require longer implementation, process changes and substantial capital commitments. Publicis said clients were more cautious on large transformation projects amid macroeconomic uncertainty, even as they continued investing in AI-powered marketing services. (MediaPost’s earnings-call coverage)

This helps explain why AI search readiness, product data, creative testing and campaign optimization can get approval before a company-wide data-platform rebuild. The first group can be scoped as a growth experiment with a defined audience and outcome. The second often requires IT, security, legal and operations teams to change how systems work together. The two initiatives are related, but they should not be sold internally as the same business case.

Marketing leaders should fund outcomes, not AI theater

The practical opportunity is to connect AI-related marketing spend to customer behavior: better qualified discovery, higher conversion, retention or reduced campaign waste. Publicis said its Connected Media business represented 62% of Q3 net revenue and posted high-single-digit growth, while its technology business improved sequentially. These results reflect multiple factors, so they do not establish that AI alone caused the growth. (Publicis Groupe’s results release; Marketing Dive’s analysis)

For an in-house team, that means testing improvements to the information customers and AI assistants encounter: clear product facts, useful comparison pages, reliable pricing and current inventory or service details. For an agency, it means translating “we need an AI strategy” into a specific customer or revenue problem—and showing how a short pilot will be measured. Neither requires promising that a brand will appear in every AI answer.

Keep transformation moving, but stage the commitment

Slower enterprise adoption is not a reason to postpone foundational work indefinitely. It is a reason to sequence it. Identify the data, identity and workflow constraints that limit customer-facing experiments, then select one or two bottlenecks for a controlled pilot. Set a baseline, owner, security review, budget ceiling and a decision date before expanding. This makes infrastructure investment answerable to a business need rather than to a vendor’s broad AI roadmap.

Agency leaders should also prepare for a two-part client conversation. The first is urgent: how will customer discovery and purchase behavior change, and what can the brand test this quarter? The second is structural: which systems, permissions and data practices must change to scale what works? Sadoun described precisely this split between faster marketing commitments and more measured enterprise transformation. Marketing Dive’s reporting also notes that Publicis’ technology arm has faced a slower growth trajectory even as it helped secure major client work—another reminder that long-term capability building and near-term results do not always move in sync.

Three moves for the next 90 days

  • Separate the budgets. Label customer-facing AI marketing experiments apart from enterprise platform or data transformation spending.
  • Choose a measurable customer problem. Test one change to discovery, comparison, creative performance or conversion, and define success before launch.
  • Build a scale gate. Expand only when the pilot shows value, data quality is adequate and the required privacy, security and approval controls are in place.

Publicis’ results offer a useful strategic signal, not a universal forecast: customer-facing AI pressure may move faster than enterprise modernization. Leaders who plan for both clocks can act now without mistaking a quick marketing pilot for a complete transformation.

Want a practical AI marketing plan tied to measurable growth? Contact Real Internet Sales at 803-708-5514 or visit realinternetsales.com.