AI budgets are growing faster than proof of ROI
Companies are spending heavily on AI, but most agent projects are still short of full scale and many marketing teams can’t show clear financial value.
AI is getting funded faster than marketers can prove its value.
Two-thirds of enterprise organizations now have dedicated AI budgets, and 95.3% have AI agents on the roadmap, according to The Martech Weekly’s “Enterprise Martech Outlook 2026.” Yet only 10.8% of agent initiatives are fully scaled in production, while 40.2% of martech leaders still can’t show a clear financial contribution from their technology investments.
Companies are spending. They’re experimenting, and they’re flailing. Most AI agent initiatives haven’t reached full-scale production, and martech teams are hard-pressed to prove ROI for the new technology.
AI is helping marketers more than customers
AI is making a bigger dent in marketing operations than in customer experience.
Some 69.3% of those surveyed say AI is having a reasonable, clear, or substantial impact on the martech stack. Fewer, 59.9%, say the same about customer experience.
That gap makes sense when you look at where AI works best right now. Internal uses like writing, editing, analysis, reporting, and creative variation are easier to test, govern, and fit into workflows marketers already know.
You see the same thing with the use of agents. Co-pilots, copywriting and editing, data management, data analysis, and video generation are further along in production and report better ROI. Journey optimization, decisioning, audience selection, campaign creation, attribution, loyalty optimization, and offer agents still have more ground to cover.
The agent roadmap is well ahead of production
Nearly every organization in the survey has AI agents somewhere in its plans. Very few have gotten them to full scale.
The low number of successful, scaled agent initiatives says a lot. Nearly 90% of organizations are still in planning, proof-of-concept, or limited-production stages. Autonomous marketing remains a long way off.
The hard part isn’t getting an agent to produce an answer. It’s being sure that the answer is true. Journey optimization, decisioning, attribution, and campaign creation are just a few of the jobs where AI is inheriting already tricky data streams and processes. If the workflow is shaky, the agent just makes it shaky faster.
Humans are still firmly in the loop
Enterprise brands clearly understand AI’s limits and risks. They’re willing to use it broadly, but most still want a human between the model and the customer.
Just 1.6% allow fully automated AI-generated customer-facing content. Most (43.3%) allow external use of AI-generated content only after it’s been reviewed, edited, and verified. Another 24.4% limits generative AI to internal use only.
The problem is that the high cost of AI was supposed to be justified by increased worker productivity. There’s no increase if time saved on production is spent on verifying the output.
AI still has to survive the budget meeting
The broader martech measurement problem hasn’t gone away just because AI has its own budget line.
Some 40.2% of martech leaders say they cannot demonstrate a clear, measurable contribution to financial objectives or accepted financial proxies. Teams that could prove value were much more likely to report budget growth: 56% received increases, compared with 37.5% among organizations relying on what the report calls “faith-based” value demonstration.
AI may get special treatment during the investment phase, but it still ends up in the same budget conversation.
That puts the burden on marketers to distinguish between the AI that makes work better and the AI that merely makes more work happen. The first group is easier to defend because the result is visible. The second tends to disappear into vague claims about productivity, transformation, or future potential.
Dedicated AI funding is plentiful right now. The harder currency is proof.