Marketing said yes to AI, but nobody gave them what they need to deliver it

Marketing said yes to AI before anyone decided who actually owns it. The data shows what happens next.

Leadership is demanding AI adoption and ROI accountability simultaneously, but most organizations haven't built the ownership, data infrastructure, or governance needed to make either possible.

To find out where the gap really lives, we surveyed 435 marketing leaders across five countries in late 2025 — director level and above, at brands and agencies. This analysis draws on the same survey data used in the 2026 Marketing Data Report, going further on accountability, ownership, and the specific points where AI execution breaks down. The AI readiness gap is the distance between the pressure organizations feel to adopt AI and their actual capacity to deploy it accountably.

And the pressure is hitting teams that are already stretched: 45% say they often feel rushed and don't have enough time to properly analyze the data, and 50% wait 1–3 business days just to get ad hoc data questions answered.

Key takeaways

The C-Suite wants AI. Marketing has to make it work.

Pressure to adopt AI isn't evenly distributed. It flows downhill from leadership to marketing teams, who have the least structural support to catch it.

Among the 435 leaders surveyed, 56% of SMB marketers and 67% of enterprise marketers say that pressure to adopt AI comes from senior leadership or C-suite executives. But the data strategy that determines whether AI can actually work? In 47% of organizations with fewer than 1,000 employees and 59% of those with more than 1,000 employees, that strategy is owned by a team entirely outside marketing. The mandate flows in one direction, so marketing teams absorb the accountability for AI performance while another team controls the data that determines whether it's achievable.

Only 6% of both SMB and enterprise teams describe AI as "fully embedded" in their workflows. Most are experimenting, partially adopted, or, at best, ad hoc.

In early findings from Supermetrics' ongoing AI Readiness Poll, 85% of respondents either have no formal AI strategy, have only partial clarity on use cases and ownership, or say AI is mentioned in planning but never prioritized. Only 15% report a clear strategy with defined owners and success metrics, consistent with what we found in the broader survey.

Where is AI actually being used? Mostly in content creation (52% SMB, 50% enterprise), market research (41% SMB), and reporting (37% SMB, 40% enterprise) — individual, task-level work that looks more like a chatbot than a system. Most teams are using AI to move faster, not to make better decisions, which is a significant difference.

The areas where AI could actually close the accountability loop (the chain connecting data ownership, AI execution, and provable ROI) are where adoption is lowest. These are the use cases that would let marketing teams connect cause to outcome, prove spend decisions in real time, and execute across systems without a manual handoff. They're also the hardest to deploy without a solid data foundation, which is why most teams default to the easier wins.

AI adoption is splitting into two speeds. In content and creative work, AI errors are visible and recoverable. A bad headline gets caught before it ships. In reporting and analytics, failure is quieter. A marketer without strong data skills may not recognize when an AI-generated number is wrong, and most won't catch it until it's already in a board presentation.

Andrea Linehan, CMO at Supermetrics, describes what that pattern produces: "AI accelerates the production of dashboards and assets but insights still fail to bring confidence or conviction at leadership level. Speed increases, but belief is stagnant. These teams aren't blocked by technology — what's missing is clear value creation."

Most teams are already working from fragmented, inconsistent data. When AI gets applied on top of that, it hides the gaps. A connected data foundation is what separates AI-powered reporting that can be trusted from AI-powered reporting that only looks accurate.

Marianna Imprialou, Data Scientist at Supermetrics, puts it plainly: "Their problem at the moment is not that AI is not embedded in their workflows — it is that their workflows are broken. They have no budgets, they don't know what they're doing. AI should help them, but hasn't helped them yet enough."

Leadership is pushing AI adoption without transferring the data ownership, governance, or integration work that makes it possible to deliver on. Marketing teams, on the other hand, are being held accountable for AI outcomes while operating with fragmented data, limited system access, and workflows that weren't built for the kind of analysis AI-powered decisions require. The pressure arrives before the infrastructure does, and that gap is where the accountability problem lives.

Why don't marketing teams control their own data strategy?

64% of organizations surveyed say their data strategy is primarily owned by an analytics team outside marketing or by external consultants. That means marketing teams are accountable for their own AI performance — the campaigns, the measurement, the ROI — but often can't control the data strategy that determines whether any of that is achievable.

In 22% of SMB and 28% of enterprise organizations, a CDO or Head of Data owns the strategy at the org level. In 47% of SMB organizations and 59% of enterprise organizations, the analytics team is entirely outside marketing.

As Andrea Linehan, CMO at Supermetrics, puts it: "When marketing is excluded from data strategy, data becomes operationally impressive but commercially vague and even useless. Leadership is presented with dashboards that explain what happened, but not why it happened — and never meaningfully inform what to do next."

The collaboration data makes this concrete. Only 10% of smaller organizations and 11% of enterprise teams describe their marketing and data/analytics relationship as "fully integrated." 35-39% coordinate only on specific projects and work largely independently the rest of the time.

Where marketing should own the data strategy versus where it makes sense to collaborate depends on where you are in the pipeline. At the collection stage, marketing needs a seat at the table. Without it, organizations may optimize for what's easy to capture rather than what's useful to understand. Analysis works best as shared territory: analytics brings rigor, marketing brings commercial context.

Activation is where marketing's ownership becomes non-negotiable. If marketing doesn't control how insight gets translated into action, it gets reduced to a reporting function. That's the point where the accountability loop most commonly breaks.

The insight-to-action gap — the breakdown between identifying an insight and acting on it — is a downstream symptom of fragmented data ownership. When marketing doesn't own the data strategy, it can't control data quality, access speed, or how insights are structured for action. It just receives outputs and is held accountable for results.

The data bears this out: 40% of SMB and 34% of enterprise teams surveyed say the biggest blocker between identifying a key insight and acting on it is the lack of system integration between their analytics tools and activation platforms. Another 20% of SMB and 27% of enterprise teams cite time-consuming manual data handoffs. Combined, 60% of teams are stuck at the last mile — not because they lack skills, but because the pipes don't connect.

As Richard Jonkhof, Supermetrics' Global Director of Professional Services, observed during our research: "It's not that they don't have any data. The biggest gap is not the collection of data or the availability of data — it's the availability of data for the marketer." That gap between the data that exists and the data marketers can actually use runs through every section of this report.

Why can't marketers turn data insight into action?

When a marketing team spots an underperforming creative or a high-value audience segment, 40% of SMB and 34% of enterprise teams cite a lack of system integration as the biggest blocker to acting on it.

Another 20% of SMB and 27% of enterprise teams say it's time-consuming manual data handoffs. Add those up and the majority of teams, 60% of SMB and 61% of enterprise, are stuck at the last mile for structural reasons.

The data most teams are working with isn't good enough for AI

Only 11% of organizations surveyed describe their marketing data as “extremely high quality, accurate, and accessible across systems.” 29% are moderately confident in their data, and 24% trend toward uncertainty or outright disagreement.

AI built on shaky data produces shaky outputs. As Outi Karppanen, Supermetrics' Lead Marketing Industry Strategist, puts it: "AI is only as good as the data you feed it. If your marketing data is messy or scattered, your AI won't just fail. It'll do something worse — it'll mislead you."

Training won't fix a problem that's really about access

Just over half of marketing teams (51%) rate their data competency as "intermediate" — capable of basic tasks but not complex analysis or strategic insight. 46% say the number of team members with advanced data skills hasn't changed year over year.

The industry prescribes training. The ceiling isn't skills — it's access. A team with intermediate data skills and a connected, reliable data infrastructure can act on insights that a technically advanced team with fragmented, siloed data simply can't.

The last mile between insight and action is where execution dies

Even when the data is there, and the skills exist to read it, 40% of SMB and 34% of enterprise teams hit a wall at the last mile — the final step where insight should become a live campaign change — because their reporting tools and ad platforms simply don't talk to each other. Another 20–27% are still moving data manually between systems.

The organizations using marketing data effectively are running first-party data directly into paid channels (40% SMB, 42% enterprise), segmenting audiences in near real time (40-49%), and eliminating the manual step between insight and campaign change (28-34%). Most teams want to be there. The integration gap is what's keeping them out.

AI readiness is about whether the insight that surfaces in your dashboard on Monday can change what's live in your campaign by Monday afternoon — without a ticket or a three-day wait. Most organizations aren't there yet.

Why can't marketing teams prove ROI when it matters most?

Marketing teams can't prove ROI cleanly because the data behind it is scattered across channels and too slow to tie spend to outcome in time to matter.

The scale shows in the numbers: 40% of organizations surveyed cite proving ROI across channels as their top challenge in justifying budgets, and 44% say difficulty measuring ROI is a primary barrier to optimizing performance. The problem isn't effort. Teams are stretched thin: 45% of marketers surveyed often feel rushed and don't have time to thoroughly analyze their data, and another 17% rarely have time at all.

Marianna Imprialou, Data Scientist at Supermetrics, is direct: "The typical marketer never gets this time to think and analyze. The extreme pressure of running specific trends or specific sales goals means the marketing department will always be trying to chase its targets. That's why there is this big gap — for someone or some technology to come in and facilitate, and take some of that load so they can do the strategic tasks that really matter."

Only 7% of marketing teams say data requests are answered in real-time. Half (50%) wait 1–3 business days for ad hoc data questions, and 42% say that wait time doesn't meet their needs. That lag has a compounding effect. Marketing teams are already being asked to produce more with AI than they could before. When the data needed to make a decision takes days to arrive, the speed AI is supposed to deliver is entirely absorbed by the wait.

Zach Bricker, Supermetrics' Lead Solutions Engineer, puts it plainly: "Marketers know real-time optimization drives results, but fragmented data keeps them reactive. Teams need to move from simply having data to activating it. AI and automation remove technical friction and allow experts to focus on strategy and revenue impact."

That distinction matters because most teams are drowning in data they can't act on while simultaneously missing the specific data they need. 46% of marketers say better data integration across platforms would do more to close their capability gaps than any other single investment.

Early findings from Supermetrics' ongoing AI Readiness Poll reinforce this. 66% of respondents say they either can't prove the ROI of AI at all or can only show activity without connecting it to business results. That tracks directly with the survey finding that the ROI proof problem isn't getting better.

Andrea Linehan, CMO at Supermetrics, on the core contradiction: "Boards do understand that every investment has a risk profile. What they don't appreciate is how many unknowns and assumptions we're working with in marketing despite all the data. We have so much data, so that means we have all the data we need. But that's not the case." AI hasn't resolved that gap for many teams. The expectation to move faster and prove more has arrived before the infrastructure to support it has.

59% still justify marketing budgets to finance by demonstrating ROI. But the survey data shows how they're actually doing it: 59% link marketing goals to revenue growth, 36% link to profitability, and 28% use data-driven predictions.

Most aren't proving ROI with clean attribution — they're making the case through revenue narratives and forward-looking projections. It works until it doesn't. When AI raises the output bar and leadership expects faster, more granular proof, a revenue narrative is no longer enough.

Leadership demands AI adoption and proof of ROI at the same time. Marketers are accountable for delivering both — with data that's slow, siloed, and behind the conversation that matters. Until organizations close the gap between who owns the data and who's held responsible for its outcomes, AI won't save them from the accountability problem. It will make it worse.

More tools won't close the AI gap — better data infrastructure will

The gap between 80% pressure and 6% full adoption is an ownership problem, not a technology issue. It won't close until marketing can take collaborative ownership of the data strategy AI depends on.

The organizations getting there aren't treating AI as a standalone investment. They've built the data ownership and connected infrastructure that makes AI worth deploying in the first place.

CEO Anssi Rusi puts it directly: "AI alone is not the answer. AI needs high-quality data, transparency, and trust — and it needs to be integrated in a way that marketers understand what it does and have control over it."

The fix isn't a better AI tool. If your team is still waiting days for data answers or manually moving numbers between platforms, Supermetrics is the marketing intelligence platform that connects your data, surfaces the insights that matter, and gives your team what it needs to act — without the queue.

Methodology

Supermetrics surveyed 435 marketing leaders across the United States, United Kingdom, Germany, Australia, and Singapore in November 2025. Respondents held director-level or above positions — including C-suite executives, vice presidents, department heads, senior directors, and senior managers — at brands, marketing agencies, digital agencies, and media agencies. Respondents represented organizations ranging from 100 to 10,000+ employees and spanned industries including retail, financial services, software and SaaS, CPG/FMCG, media and entertainment, and marketing and advertising. All respondents held decision-making or influencing authority over media or adtech/martech budgets. The survey was conducted by NewtonX.