Technology
Dashboards
Business Intelligence & Insights
Communication Portal
Intelligent Collaboration
All Channel Activation
Activate Media on Every Channel
Consulting
Resources
Key Takeaways:
Marketing teams have moved quickly to adopt AI. But when it comes to delivering measurable returns on those investments, most organizations are still finding their footing.
For the second consecutive year, agency leaders named AI as their top investment priority, with more than three-quarters planning to increase their AI spend over the next 12 months. However, only about 29% of organizations across sectors say they can dependably measure ROI on their AI initiatives, and CEOs report that just 25% of their AI initiatives have delivered expected ROI.
Right now, the gap between investment and demonstrated impact is significant. Closing it requires a deliberate approach to how tools are selected, how a strong foundation is built before implementation, and how their impact gets translated into stories that resonate with stakeholders.
Getting real returns on AI starts well before a tool is ever deployed. Key steps include:
With a proliferation of AI solutions crowding the market, tool selection has become an increasingly consequential decision. According to Lauren Johnson, Effectiveness Lead at Basis, a deep understanding of how an AI tool works is the starting point for evaluating whether it will deliver meaningful results.
"If we are going to ask AI to help us evaluate datasets, we need to understand how it's executing that task," says Johnson. "What are its outputs based on? What historical data is it drawing from? Marketers need to understand how their tools work in order to assess whether they'll be able to drive the impact they're looking for."
Given the state of the market, deep evaluation is critical: Gartner has warned that many vendors are engaging in “agent washing,” or positioning existing products as agentic AI without adding genuine agentic functionality. Of the thousands of vendors pitching themselves as agentic AI providers, Gartner estimates only around 130 actually deliver on that promise.
A deeper understanding of how AI tools work also makes the case for seeking out specialized solutions over generic ones. Out-of-the-box AI solutions have strengths when used strategically, but specialized tools trained on large volumes of relevant industry data tend to produce more precise and differentiated outputs. An AI media planning tool trained specifically on advertising performance data, for example, can generate cross-channel recommendations grounded in real campaign outcomes—something a general-purpose model isn't equipped to do.
Data quality is a make-or-break factor for AI performance, but most marketing organizations aren't yet where they need to be. Only 21.4% of industry professionals describe first-party data as “foundational” to their organization's AI initiatives, and roughly one-third say first-party data plays little-to-no role in their current AI use at all. Half of leaders also say their businesses don’t have the technical or data stack readiness to support AI agent deployment.
Without clean, unified, accessible data, AI tools produce outputs that are generic at best and misleading at worst—neither of which supports the kind of differentiated results that justify continued investment. Leaders must treat data readiness as a strategic prerequisite. That involves:
Finally, understanding the impact of AI investments requires knowing exactly what success looks like before deployment. “Just like with anything in our world,” says Johnson, “assessing effectiveness starts with setting a goal for what you want the tool to do for you.” If saving time on a specific workflow is the goal, for instance, measure how long that workflow takes today and define a clear target for how much AI should reduce it.
Currently, only 40% of marketing professionals are using or planning to use defined KPIs specifically for their AI solutions. Crafting a phased roadmap for AI adoption, with concrete milestones and tool-specific performance targets, gives teams both a framework for evaluating impact and the foundation for communicating that impact to stakeholders.
How marketing teams use and communicate about their AI investments is just as important as laying the groundwork for them to succeed. The teams best positioned to demonstrate impact tend to:
To achieve maximum value from AI tools, humans must consistently scrutinize what they produce. The technology can provide weak or inaccurate outputs if trained on low-quality data, and of course, there’s the matter of hallucinations: One study found that close to half of marketers spot inaccuracies in AI outputs several times a week.
“AI tools are not going to tell you when they're wrong,” notes Johnson. “Teams need to continually push back against and stress-test AI outputs in order to get real value.”
Leaders should nurture a team culture where pushing back on AI outputs is both expected and encouraged. When that standard is set by leadership and integrated into how teams operate on a daily basis, it drives higher-quality AI usage across the board. Teams that operationalize this approach are better positioned to extract demonstrable value from their AI investments over time.
Proving AI’s value to stakeholders is a significant challenge for marketing teams in 2026. In fact, only 41% of marketers can confidently prove out the ROI of their AI investments, down from 49% in 2025.
This is where the goals and KPIs set during the planning phase become essential. Tracking performance against those benchmarks over time—whether that’s hours saved on a specific task, improvement in campaign performance, or lift in a business outcome—gives teams the evidence they need to make a credible case for AI’s impact.
Beyond quantifying AI’s impact, marketing leaders must strategize around using that evidence to craft compelling narratives for stakeholders. For one thing, while AI can (and should) drive meaningful improvements in speed and cost, leaders should focus more on how their AI investments contribute to business outcomes. AI efficiency gains can be significant and are worth including in these stories, but positioning AI’s value primarily around efficiency risks reinforcing the perception of marketing as a cost center rather than a strategic growth driver.
Connecting investments to business outcomes is also what tends to land with senior decision-makers. The most resonant executive narratives tie major investment asks to concrete business drivers that connect marketing’s AI investments to the revenue and business outcomes that sales and finance leaders care about.
Another key to making those narratives credible is grounding them in a deep understanding of the tools themselves. “Executives want to hear that we're using AI, but they also want to know that it's grounded in real data,” says Johnson. “They want to know what’s powering these tools and that they can trust the outputs.”
That deep understanding is part of what makes an AI narrative credible at the executive level. Leaders who build expertise in how their tools work, what data powers them, and how that functionality is advancing business goals will be poised to earn sustained buy-in.
The ability to drive and demonstrate ROI on AI tools is quickly becoming one of the clearest dividing lines between marketing teams that lead and those that fall behind. Getting there requires a deliberate approach at every stage: selecting tools with a deep understanding of how they work, fueling them with the data infrastructure they need to perform, and translating their impact into narratives that resonate with stakeholders.
—
Want more insights on how marketing teams are approaching AI? Our AI and the Future of Marketing report synthesizes findings from a proprietary survey of professionals across leading brands and agencies, covering adoption trends, workforce shifts, and the biggest barriers teams are still working through.