At the ANA Masters of Marketing in Orlando, EY’s director of digital marketing says a crisis of confidence in marketing has less to do with AI and more to do with accountability. His prescription? Brand discipline, better data and real AI literacy.
Lou Cohen
Lou Cohen isn’t interested in more AI hype. He’s interested in restoring something rarer – confidence. Not the keynote-stage kind, but the kind you can take into a boardroom when the CFO asks what marketing actually delivered.
“The ANA’s new ‘Confident Marketer’ research hit a nerve,” he says. “Most companies can’t see the real impact of their work. That’s a confidence killer. We’re entering an era where outcomes, not activity, define credibility.”
Cohen frames it as a lifecycle problem: early-career marketers tend to be overconfident, mid-career leaders lose it when new tech leaves them behind and the best CMOs regain it by taking smart risks. “You’ve got to experiment, fail fast and own it,” he says. “That’s how competence and confidence finally meet.”
Advertisement
EY is embedding AI on two fronts – what it builds for clients and how its own marketing runs. Internally, the firm is piloting WPP Open and its secure in-house environment, EYQ, a sort of “enterprise ChatGPT” locked behind EY’s firewall.
One recent experiment produced a content-analysis agent that scores and improves thought-leadership pieces without rewriting them. “We don’t let AI create finished content,” says Cohen. “But we do let it guide us to better quality, faster. It’s a huge productivity gain.”
It’s an operational approach to AI, as opposed to a generative one – systems that amplify human judgment rather than replace it. The next step, he says, is neuro-symbiotic AI, where generative models begin to replicate human reasoning instead of merely predicting words.
That distinction matters. ChatGPT is probabilistic – it produces the most likely answer based on patterns in data. Ask it to add one and one a thousand times and you might occasionally get something other than two. Google Search is deterministic – it retrieves a fixed answer from a verified source. One and one always equals two.
“The future of search will fuse those two worlds,” Cohen predicts. “Probabilistic tools give ideas and nuance; deterministic systems give facts. Marketing has to learn how to show up credibly in both.”
For EY, that fusion creates a fresh challenge: ensuring its expertise appears in AI-generated answers – a new frontier he calls GEO, or generative engine optimization. “If a CEO asks an LLM, ‘Who can fix my supply chain?’ and EY isn’t part of that response, we’ve already lost,” he says.
Cohen’s team now maps which sources large language models rely on, then adjusts distribution accordingly. “Right now it’s like SEO in the 90s,” he says. “AI engines don’t yet know how to judge quality, so random auto-written sites creep into results. Until algorithms mature, we’ll keep raising our own content standards and pushing it into credible ecosystems like Wikipedia and Reddit.”
At the ANA, talk of “brand” kept resurfacing, but Cohen believes that many marketers still misunderstand who their real customers are.
“B2C marketers forget their real buyer might be a retailer or distributor,” he says. “B2B marketers forget brand altogether. Both are rediscovering that trust is the multiplier.”
He uses Sherwin-Williams as the perfect example. The paint giant is widely seen as a consumer brand, but its most lucrative customers aren’t DIY homeowners buying a couple of tins every few years. They’re contractors and developers who buy paint in bulk, month after month.
“Sherwin-Williams markets with consumer polish, but sells like a B2B powerhouse,” Cohen explains. “By making its consumer brand famous, it makes the B2B buyer’s job easier. Contractors trust it. Builders specify it. The brand reduces friction all the way through the supply chain.”
Cohen says the same dynamic applies at EY. “We’re not the hero – our clients are. If they succeed, that’s our brand in action.” For a firm built on trust and independence, evidence matters more than emotion. Audit-quality improvements, transparent reporting and rigorous compliance aren’t just process details; they’re brand assets. “Confidence,” he says, “is built through proof, not promises.”
Advertisement
For Cohen, AI isn’t just a marketing trend; it’s the defining leadership test of the decade. Every CEO, he says, is wrestling with the same paradox: how to modernize at speed without breaking trust.
“EY’s business is built on managing risk, not creating it,” he says. “Our job is to make AI safe to use – to de-risk innovation so clients can move faster.” That means building systems that are explainable, secure and compliant, while helping boards move beyond fear into action.
That same philosophy shapes Cohen’s view of the wider economy. The future may look uncertain, but he’s quietly optimistic. “M&A is about to surge,” he says. “Companies have been sitting on cash through tariff uncertainty and they can’t hold it forever. Manufacturing and supply-chain investment are picking up and CEOs will need help turning that momentum into execution.”
In other words, the confidence revolution he describes within marketing might soon play out across the broader US economy. Let’s hope he is right.
The confident marketer’s checklist:
Prove outcomes, don’t just measure impressions.
Treat AI as an operating system, not a campaign gimmick.
Invest in generative discoverability – GEO is the new SEO.
Put brand back at the top of the funnel.
Build trust through evidence and process.
Advertisement