Hey! I'd love to hear your thoughts, send me a voice note.

In this episode of The Marketing Rundown, we cover the aggressive wave of forced ad automation hitting search campaigns—from Microsoft stripping Max CPC bid caps on October 1st to Google automatically upgrading Search campaigns to AI Max. We also dissect a brutal case study of an account with 11x platform ROAS that was actually losing money on every order, reveal how Meta’s new Muse AI agent shops and checks out for 1B users, and break down WebMCP, the new web standard for agentic commerce.

📋 Also Worth Noting:

• Google Ads Local Customer Optimization: A new feature optimizing campaigns specifically for "near me" local intent.
• Google Preferred Sources: Users can now select preferred websites to appear consistently in their AI Overviews and AI Mode.
• Lovable PLG Conversion Data: Free product usage treats acquisition spend with a 3-month payback; targeted partner giveaways convert at 40–60%.
• Google Meridian GeoX: Google released a free, open-source geo-incrementality testing library to calibrate MMM models.
• ChatGPT Ads $500 Matching Credit Expansion: OpenAI expanded its $500 matching credit offer across 31 countries.

---

This podcast is AI generated. It may contain mistakes. Check important info.

Hey! I'd love to hear your thoughts, send me a voice note.

In this episode of The Marketing Rundown, we dissect alarming new SEMrush data revealing that AI chatbots are actively talking 57.5% of users out of making purchases. We look at why building more blog content won’t fix inaccurate brand information inside AI models, and how to resolve these upstream data conflicts. Finally, we share highly actionable performance tactics: a "borrowed desire" copywriting framework, an aggressive competitor-poaching growth loop run directly in personal DMs, Google's automated promo extraction feature, and Netflix’s system for measuring untrackable offline media spend.

🔥 Top Picks & Actionable Takeaways:

• The AI Chatbot Conversion Killer: SEMRush data reveals AI chatbots discourage over half of AI searchers from buying. Because AI has become a primary product research channel, bad assistant answers are killing conversions. Takeaway: Shift focus from top-of-funnel reach to deep engine optimization. Ensure AI recommendations actively direct users to check out rather than sending them away.
• The Brand Data AI Search Risk: Publishing more blog content will not correct false brand facts already trained into LLMs. The real threat is conflicting upstream information. Takeaway: Systematically search ChatGPT, Gemini, and Perplexity for your brand, trace wrong answers back to old press, outdated directories, or broken schemas, and fix them at the source.
• "Borrowing Desire" Copywriting: The best ad copy doesn't create desire—it borrows it. Conversion Rate Experts recommend highlighting the real emotional purchase driver, naming the obstacle, and depicting a vivid outcome. Takeaway: For your next ad, skip basic feature lists. Write a 2–3 sentence scenario showing your customer in their exact moment of success.
• Personal DM Competitor Poaching: Developer tool Cursor generated massive organic reach by using personal DMs to offer their $200 plan free to users who sent a screenshot proving they canceled a rival platform. Takeaway: Set up an outreach campaign offering incentives for competitor cancellation screenshots. Use a personal team account instead of a corporate brand page to capture intimacy and urgency.
• Google Ads Automated Promotions: Google Ads can now automatically extract active offers (like BOGO or percentage discounts) directly from landing page copy and apply them to Search and PMax ads. Takeaway: Enable account-level automated promos and format your landing page text to be highly machine-readable to prevent broken or outdated offers.
• Netflix's Geographic Holdouts: Standard A/B testing cannot measure offline channels like TV, billboards, or sponsorships. Netflix solves this by running geographic holdout tests, splitting entire regional markets rather than individual users. Takeaway: Match baseline regional metrics to design geographic holdout tests before launching unmeasurable brand campaigns.

---

This podcast is AI generated. It may contain mistakes. Check important info.

Hey! I'd love to hear your thoughts, send me a voice note.

In this episode of The Marketing Rundown, we dissect a monumental shift in search engine architecture as OpenAI quietly tests its own crawler index to bypass third-party search results. We also dive into Google’s experiment with restrictive match types inside AI Mode, TikTok’s launch of the industry’s first agentic paid social automation suite, and a startling Google study showing that LLMs forget up to 74% of the facts they store. Finally, we break down how to stop paying a "triple tax" on your affiliate and creator programs, and share critical updates on Gen Z brand loyalty, Reddit reputation management, and automated YouTube sponsor tags.

🔥 Top Picks & Actionable Takeaways:

• OpenAI’s Private Search Index ("prefer-index-over-serp-v3"): OpenAI is testing its own crawl index against third-party SERPs, utilizing separate sub-indexes for news, shopping, PDFs, and YouTube. With ChatGPT.com hitting 1.09B monthly visits (+48% YoY), AI search is severely squeezing click-through rates. Actionable Takeaway: Because GA4 misclassifies ChatGPT and Perplexity referral traffic as "direct" (sometimes accounting for up to 78% of traffic), audit your direct traffic spikes over the last 12 months and set up specific source/medium filters to expose AI referrals.
• Exact and Phrase Match Serving in AI Mode: Google is quietly testing exact and phrase match keywords inside AI Mode. This means tightly managed, highly specific search campaigns may be bleeding into AI Mode placements with zero report segmentation. Actionable Takeaway: Monitor your Search Impression Share and track weekly CPC/CVR volatility to identify sudden, unauthorized AI Mode exposure.
• TikTok Launches Agentic Paid Social Suite: TikTok Ads MCP, Agentic Hub, and Skills are live. Unlike passive AI recommenders, these agents directly automate creative fatigue alerts, budget reallocations, QA, and reporting based on team workflows. Actionable Takeaway: Map out your team's most manual paid social tasks (such as budget pacing checks) and run a pilot using TikTok Ads MCP to automate a single workflow before scaling.
• The LLM Recall Gap: A Google study reveals advanced LLMs store 95–98% of data but successfully recall only 26–34% of facts. Recall fails most on rare facts and "backward" queries (e.g., answering "what does company X make" but failing on "who makes product X"). Actionable Takeaway: Search your brand facts on major engines; rewrite and republish missing or broken facts in both directions (what you make vs. who makes it) on your site.
• Waze Ads Expands Across 42 Countries: Google has unlocked Waze ad inventory across 42 countries, including all 27 EU states, running exclusively via PMax store goals. Actionable Takeaway: If you manage physical EU locations, launch Waze store goals this week to capitalize on low CPMs and minimal competition before other brands scale up.
• Eliminating the Multi-Touch "Affiliate Tax": When PR, affiliate, GEO, and social teams silo their budgets, brands often pay 3x for the same creator relationships. Actionable Takeaway: Filter publisher conversion data to check if converting customers had visited your site or searched your brand before hitting the review site.

---

This podcast is AI generated. It may contain mistakes. Check important info.