2026年营销智能报告

New research from Funnel shows how high-performingmarketers harness analytics and AI to go full throttle. The marketingdata paradox Marketers today are surrounded by more data,tools and technology than ever before. When scrolling through LinkedIn, it looks likeeveryone has cracked the code: AI-poweredcampaigns, predictive analytics, automated When we asked marketers to grade theirperformance, they gave themselves a B- (82%)at best. And when asked to rate their agencies,marketers gave a similar, mediocre grade: 81%. For This middling self-evaluation isn’t modesty. It’s acandid admission that — despite all the investmentin martech, data platforms and now AI — something In our 2026 study, we wanted to unpack this issue:Why is it that with so much innovative tech attheir fingertips, marketers rate their impact only Despite allthe investmentin martech,data platforms andnow AI, somethingfundamental isn’t In other words, we want to give you a prescriptiveplaybook to close the gap between where you areand where you need to be. We won’t give away all •From reporting to discovery:Using dashboards as hypothesis engines —where analysis isn’t the finish line, but the starting •From caution to experimentation:Building cultures where testing new approachesis rewarded, not punished — powered by From fragmented data to AI-ready infrastructure: Creating unified, trustworthy data foundationsthat enable advanced analytics today and prepare Read on for a clear-eyed look at what’s working,what’s broken and what to do about it. Table of contents So much data, so little insightAI to the rescue?Disruptors wantedUnlocking advanced analytics About the research Funnel surveyed 238 in-house marketingprofessionals as well as agency professionalsglobally. We also interviewed six agency leadersfrom high-performing, data-savvy shops in the U.S. So much data,so little insight Funnel’s latest research shows many marketers are surrounded by ample dataand advanced tools, but lack true insight and direction. They’re piloting advanced machines, but still flyingblind. Just look at the numbers: One marketer explains, “We’re drowning ininformation, but what’s the insight? What’s the •72% of in-house marketers and 55% of marketersworking in agencies say they have mountains of The problem: Many teams spend an enormousamount of time generating dashboards that lookimpressive but reveal little about what to donext.And in our experience, too many teams are •86% of in-house marketers and 79% of agencymarketers say they don’t have a clear signalthrough the noise. In other words, they struggle to “We have plenty of data, but turning it into usefulcustomer insights is challenging.” To be fair, the sheer complexity of modernmarketing means many teams struggle to identifywhat’s working across a maze of channels,touchpoints and devices. Sixty-eight percent of Let’s repeat that: More than two in five in-housemarketers say they’re simply documenting pastperformance, with little investigation into root And Funnel’s research shows in-house marketers,far more than their agency counterparts, feeladrift: 41% of in-house marketers say that whenthey report results, they don’t analyze the “why” or Agencies are 2x more likely than in-house marketers tomake ‘robust recommendations’ in performance reports Q: Do your marketing performance reports includerecommendations for action, or just summary data? AI to the rescue? AI tools and automations are hailed as a lifeline to rescue marketers from thecomplexity of modern marketing. Will they? The first problem: AI doesn’t fix messy data; itamplifies it. When the inputs are inconsistent andfragmented, the outputs may suggest accuracy butbe totally unreliable (that’s one of the problems with A study by SurveyMonkey found 93% of marketerssay AI helps them generate content faster. Butgreat marketing isn’t just about speed, volume and So the question becomes: Is AI actually deliveringon both promises? Can it make marketers faster andbetter? The answer, it turns out, is complicated. Gen AI vs. Machine Learning When marketers talk about AI, they often referto very different things. There’s generative AI:tools like ChatGPT, Gemini and Midjourneythat create text, images and ideas thatmarketers directly use and edit. Then there’s The two are often conflated, but they work atdifferent layers. With machine learning, youconsume the AI’s output indirectly throughimproved results (e.g., better CPCs, higher Both rely on data, but without a strong datafoundation, neither delivers meaningful Nearly half (46%) of marketers admit creative rolesare most at risk of decline or replacement due to AI.Yet evidence shows the opposite should be true, thatcreativity is marketing’s biggest multiplier. Researchcovering 1,250 campaigns and $140 billion in ad AI and creativity:A relationship still taking shape Funnel’s research points to other uncomfortablecontradictions at the heart of AI-driven marketing. Marketers a