Listen To The Show

Transcript

Welcome to The Brief by Kuro House, your daily update on the stories shaping the marketing and advertising landscape. Today, we’re diving into seismic shifts in data privacy, a philosophical debate at the heart of AI advertising, the real value behind LinkedIn’s steep ad prices, the merging of marketing and communications roles, and a fascinating new tool for market research: synthetic audiences. Let’s get into the details.

First up, from Digiday, the European Commission is set to unveil its Digital Omnibus package, which could mark the most significant change to European privacy law since GDPR. Leaked drafts suggest the package will amend not just GDPR, but also the AI Act, ePrivacy, and other frameworks. The most notable changes? Narrowing the definition of “personal data” so that pseudonymous identifiers may fall outside GDPR, limiting data-subject rights, and reducing protections for sensitive data unless traits are explicitly revealed. The Omnibus would also allow companies to train AI models on personal data without explicit consent, merging ePrivacy elements into GDPR for certain device access, and raising the threshold for what counts as a “systemic risk” AI model. For marketers and publishers, this could mean more addressable inventory, smoother measurement, and a return to some pre-GDPR fluidity in data flows. However, the proposal also hints at browser- and OS-level privacy settings, which could hand even more power to big platforms, echoing Apple’s App Tracking Transparency. While some see the Omnibus as overdue simplification, privacy advocates warn it could be a rollback of hard-won protections, especially for sensitive data. The upshot: expect faster enforcement, fewer compliance headaches, and a renewed focus on pseudonymization—but also a new era of competition and risk, especially as the biggest platforms stand to benefit most.

Next, let’s talk about the “hot dog vs. sandwich” problem in AI advertising, also from Digiday. As AI agents begin managing programmatic ad buys, the industry is scrambling to put guardrails in place. Enter the Ad Context Protocol (AdCP), a kind of standardized language for AI agents to communicate across the ad supply chain. The analogy is perfect: if you ask an AI to make a sandwich, you might get a hot dog—because the AI interprets instructions differently than a human might. Now imagine that ambiguity applied to six- or seven-figure ad buys. AdCP aims to create a common dictionary for AI agents, but there’s concern that key stakeholders—advertisers, publishers, and broadcasters—aren’t yet involved in shaping these standards. There’s also skepticism about who’s backing AdCP, with big players like Index Exchange, OpenX, and Amazon notably absent. The industry wants to avoid the mistakes of the early 2010s, where infrastructure became dominated by a single ecosystem. The move to agentic workflows is inevitable, but as with any powerful new tool, inclusivity and careful planning are essential to prevent a repeat of the past’s pitfalls.

Switching gears, let’s tackle the question: why are LinkedIn’s ad prices so high? According to Digiday, LinkedIn’s CPMs average around $23.42, compared to $5–$9 for platforms like Meta, TikTok, and Pinterest. Jae Oh, LinkedIn’s head of ads measurement, argues that the platform’s premium is justified by the quality of its audience and the strength of its data and intent signals. “Clicks don’t pay the bills, pipeline quality does,” says Oh. While LinkedIn’s reach is smaller and its users visit less frequently than those on Meta or TikTok, its audience is packed with decision-makers, budget holders, and professionals with buying authority. Agencies acknowledge that while LinkedIn is expensive, it’s often worth it for precise B2B targeting, higher education, or SaaS campaigns. For B2C, however, the costs are harder to justify. Despite the price, LinkedIn’s ad revenue is on track to hit $8.2 billion this year, outpacing Snapchat, Pinterest, and Reddit. The platform’s expansion into B2C could change the calculus for some advertisers, but for now, the message is clear: you pay a premium for quality pipeline, not just impressions.

Now, to a trend reshaping the C-suite: the blurring of lines between Chief Marketing Officers (CMOs) and Chief Communications Officers (CCOs). Digiday reports that companies like Hewlett Packard, Simon & Schuster, Geisinger, and T-Mobile have merged these roles, reflecting a world where brand storytelling and message discipline are inseparable. The logic is straightforward—communications is no longer just reactive, and brand building isn’t just about creativity. Both functions must navigate a politicized environment where every message can move markets or provoke backlash. Leaders who can bridge consumer, shareholder, and media narratives are in high demand. But the merge isn’t always smooth: CMOs are wired for growth and metrics, while comms leads are focused on control and reputation. The pandemic accelerated this convergence, making it clear that brands need unified accountability and agility. The path to these hybrid roles is no longer defined by traditional marketing or comms backgrounds; instead, it favors those who can tie storytelling to revenue and retention, and who understand that every platform, every day, shapes brand perception.

Finally, let’s explore the rise of synthetic audiences, as explained in a fascinating Digiday article. Imagine testing a new product or campaign with your audience—without actually bothering real people. Synthetic audiences use AI to create virtual copies of audience behavior patterns, drawing from CRM data or survey results to build “digital twins” of real cohorts. Publishers like The Times have used platforms such as Electric Twin to simulate focus groups, test editorial initiatives, and refine campaign messaging. These synthetic panels can be queried like chatbots, allowing brands to ask anything from “Why would you buy this product?” to “What would make you unsubscribe?” The benefits are speed and cost: synthetic research can be thousands of times faster and cheaper than traditional focus groups, and there’s no risk of “survey fatigue.” However, the data is only as good as the inputs, and synthetic audiences should supplement—not replace—human validation. There are quirks: sometimes digital twins give odd or unexpected answers, mimicking human behavior like skipping questions or offering unusual justifications. The key is to use synthetic data as a first filter, then validate with real people before making final decisions. For publishers, this means faster product launches, deeper audience insights, and more agile marketing strategies.

That’s it for today’s edition of The Brief by Kuro House. We covered sweeping regulatory changes in Europe, the philosophical challenges of AI in advertising, the true value behind LinkedIn’s premium ad prices, the merging of marketing and communications roles, and the exciting potential of synthetic audiences. As always, the common thread is adaptation—whether it’s to new laws, new technology, or new ways of understanding your audience. Stay curious, keep questioning, and we’ll see you next time.