June 4, 2026 | 5 min read
Last Monday, a global survey landed that should make every marketing leader uncomfortable.
Comviva released its Global CMO Survey 2026, titled “The AI Efficiency Divide: Measuring AI’s Real Value Beyond the Hype.” The report surveyed more than 200 senior executives across telecom, retail, and e-commerce worldwide. The headline finding stopped me cold: 90% of organizations have increased their AI marketing investments over the past two years. Only 12% can quantify the revenue those investments actually generated.
That is not a measurement problem. That is an accountability crisis hiding inside a spending boom.
What Actually Happened
The survey was released June 2, 2026, and it paints a picture that most of us have quietly suspected but rarely said out loud. The AI marketing industry is in a phase where everyone is buying, very few are measuring, and almost nobody is willing to admit it to their board.
The specific numbers are brutal.
“86% of marketing leaders have been asked by their board or C-suite to justify AI spending in the past 12 months. Only 16% said they were confident they could.”
That is not a small confidence gap. That is the vast majority of senior marketers walking into budget meetings with estimates, partial data, and incomplete attribution, hoping the conversation stays high-level.
What is making measurement so hard? Three structural problems keep surfacing. First, cost fragmentation: 62% of respondents said they cannot consolidate their total AI spend because it is split across cloud infrastructure, software subscriptions, talent, data vendors, and integration costs. Second, revenue attribution complexity: 58% cannot draw a direct line between an AI initiative and a revenue outcome. Third, the CX-to-revenue disconnect: 55% see genuine improvements in customer experience metrics but cannot translate those into numbers the CFO will accept.
The report also flags a hidden multiplier: many organizations are underestimating total AI costs by 30 to 50% because they only track the obvious line items. If your ROI calculation is built on an incomplete cost base, every number is wrong by definition.
Why This Hits Different for Marketers
This is not an academic finding. This is the shape of your next budget conversation.
Think about what the accountability era looks like in practice. A year ago, you could show your leadership team an AI pilot with promising early signals and get funding to expand. That window is closing fast. The same survey shows 86% of boards are now explicitly demanding proof of business return on AI initiatives, not just adoption metrics or cost-per-task efficiency gains.
The marketers who cannot prove ROI are not just at risk of losing budget. They are at risk of losing credibility at the exact moment AI becomes the most important capability shift in their careers.
Here is the before-and-after that matters most. Before this accountability wave hit, a campaign team could say “we ran AI-personalized subject lines and open rates improved 15%.” That felt like proof. Now the CFO is asking: “What did that 15% open rate improvement translate to in pipeline? In revenue? How does that compare to what we spent on the AI tools, the data infrastructure, and the team hours to implement them?” Most teams do not have the answer.
The good news from the report is that AI does deliver when deployed in the right places. Customer segmentation and targeting is the top performing use case at 57%, followed by campaign automation at 43%, predictive personalization at 41%, and pricing optimization at 39%. Notice the pattern: every one of these connects directly to a transaction or a conversion. They are measurable because the output is a business action, not a content asset.
What to Do This Week
First, audit your AI cost stack. Not just your SaaS subscriptions. Pull together cloud costs, any data licensing fees, the time your team spends managing AI tools, and any integration or API costs. If your current ROI calculation does not include all of these, your numbers are optimistic by definition. Spend two hours this week building a one-page cost baseline.
Second, pick one AI initiative and build a measurement chain from output to revenue. Choose the initiative that is closest to a transaction, whether that is email personalization, lead scoring, ad targeting, or pricing. Map the chain: AI action taken, engagement metric changed, conversion metric affected, revenue outcome produced. Even if the chain has gaps, documenting it forces you to see exactly where attribution breaks down and what data you need to fix it.
Third, get ahead of your next board conversation. If 86% of leaders are being asked to justify AI spend, you will be asked too, likely sooner than you expect. Draft a one-page AI accountability summary now: total investment, top three use cases, measurable outcomes achieved, and your measurement improvement plan for the next quarter. Bring it to your next leadership meeting before they ask for it.
The organizations that get ahead of the accountability curve will not just protect their AI budgets. They will earn the credibility to increase them.
The gap between 90% and 12% is not a technology problem. It is a measurement discipline problem. And it is entirely solvable.
What is your current approach to measuring AI marketing ROI? I am curious whether teams are finding ways around the attribution problem or just sitting with the uncertainty.
Found this useful? Drop a comment below, I read every one.
I built a free tool that audits your website in 60 seconds. You get a score, an AI voice summary, and a step-by-step PDF. → edwinrogers.com/audit
Follow for daily AI marketing breakdowns, every morning → @edwinrogerscom
#AI #Marketing #DigitalMarketing #AIMarketing #MarketingROI #MarTech #CMO #MarketingLeadership #AIAdoption #AIAccountability #AISpend #CustomerSegmentation #CampaignAutomation #PredictiveMarketing #ComvivaReport #LinkedInMarketing #ThoughtLeadership #DigitalArchitect #MarketingStrategy #BuildInPublic
