Everyone's Spending on Media. Nobody Knows If It Works.

This year, less than 6 in 10 marketers believe their organization has the right media mix. Last year it was 64%. Now it's 57%. The drop seems small, but it says something important: even the professionals managing million-dollar budgets aren't sure the money is going where it should.

Kantar's Media Reactions 2026 report confirms a trend we see from our agency seat as well: confidence in how marketing budgets are distributed is declining, even as the budgets themselves grow. It's a paradox worth paying attention to. Not because the allocations are necessarily wrong, but because nobody has a clear framework to verify whether they're working.

What's even more telling: confidence in balancing brand building with performance marketing dropped from 60% to 50% in a single year. Half of all marketers no longer know how much to allocate to brand versus performance. And that uncertainty shows in every campaign.

Channels don't work in isolation. That's a problem.

Between 2020 and 2025, the combined effect of channels working together accounted for 43% of campaign performance, according to Kantar's data. Before 2014, that figure was 18%. The impact of channels working together has nearly tripled in the last decade. And yet, most measurement frameworks still treat each channel as if it operates alone.

What does that mean in practice? A channel evaluated on its own tells only part of the story. Google Ads performs differently when it runs alongside a social campaign versus when it runs solo. Email converts differently when preceded by a YouTube video. An organic Instagram post works differently if the user has already seen a Google ad.

The problem is that most companies still measure channels in silos. "How much did Facebook bring in?" is a common question, but an incomplete one. The right question would be: "How does Facebook contribute to the performance of the entire mix?" And that requires a portfolio approach, not budgetary silos.

We recently wrote about how Google Ads is trying to give more channel-level control. But control without portfolio-level visibility remains a partial tool. You know which levers you have, but you can't see how they connect.

On a recent project, we tested exactly this: we ran the same campaign with and without a YouTube pre-roll component. The difference in direct conversions was minimal. But when we analyzed assisted conversions across the full funnel, YouTube contributed to 22% of the purchase journey. Without portfolio-level visibility, that channel would have been cut from the budget.

Micro-creators are winning. But not for the obvious reason.

Kantar's data shows marketers plan to increase micro-creator budgets by 66% and nano-creator budgets by 61% in 2027. Increases for mega and macro-creators are smaller: 42% and 48%, respectively.

It's tempting to read this as a pure cost play. Micro-creators cost less, so they get more budget. But the reality is more nuanced.

What works about micro-creators isn't just the price. It's the fact that their audience trusts them. In a period where consumers are increasingly skeptical of brand messaging, a creator with 15,000 followers recommending something can have more impact than an influencer with 2 million. The relationship is closer, the recommendation is more credible, the engagement is more real. It's a form of earned trust that money alone can't buy.

This doesn't mean macro-creators are disappearing from the equation. It means the creator mix is becoming just as important as the channel mix. And just as complex to manage. You need a clear strategy: who speaks, to whom, on which platform, with what objective.

We recently discussed the measurement challenge in influencer marketing. The problem persists: budgets grow, but attribution tools don't keep up. And without proper attribution, every budget decision is partly guesswork. You increase micro-creator spending because it "feels" right, not because you have the data to confirm it.

AI promises a lot. Consumers aren't convinced.

75% of marketers plan to increase AI assistant investment next year. That's an impressive number. But it comes paired with a less comfortable one: only 23% of consumers trust AI's brand recommendations.

There's a major gap between backstage enthusiasm and front-row skepticism. Marketers see AI as an efficiency accelerator, and rightly so. You can create content variants, optimize audiences, personalize messaging at scale. But consumers don't see those processes. They just see the output. And often, they don't find it credible.

This isn't an argument against AI. It's an argument for discernment. AI works excellently as a backend tool: analysis, segmentation, A/B testing at scale, performance prediction. It works less well as a frontend voice, at least for now. Consumers sense the difference, even if they can't articulate it precisely.

Our recommendation at difrnt.? Use AI where it brings speed and precision, but keep the human voice where trust matters. Don't let automation replace authenticity. An AI-written email that's properly personalized can perform well. An AI-generated ad that sounds generic loses from the start. And a chatbot recommending products without context loses even more: it loses the trust you spent months building.

Visibility, not volume

The problem isn't that marketers aren't investing enough. The problem is they're investing without an integrated view. Kantar recommends a portfolio approach: don't evaluate each channel separately, evaluate how they create value together.

In practice, this means a few things. Move from "how much did channel X bring" to "how does channel X contribute to the marketing ecosystem." Diversify your creator mix the way you diversify your channels. Use AI strategically, for what it does well, not just because it's available. And invest in measurement tools that reflect reality, not just isolated per-channel performance.

If you're in a position to make marketing budget decisions, ask yourself one question before any allocation: do you have a clear picture of how your channels interact? If not, no budget increase will compensate for the lack of visibility.

Media mix confidence doesn't recover with more budget. It recovers with more visibility into how the budget works. And that starts with a simple question: are you measuring what matters, or just what's easy to measure?