TLDR MARKETING 2026-11-02

Scan your homepage for free

MAJORITY OF TEENS AND TWEENS NOW USE CHAT BOTS, NEW RESEARCH FINDS

AI chatbots are becoming a normal part of life for American teens and tweens. Nearly half of 9-12-year-olds have bypassed the 13+ age restrictions. Students use chatbots for homework and emotional support. A separate study of 5,000+ students found that using AI during emotional distress is associated with worse mental health outcomes. Researchers have not established whether AI contributes to these issues or attracts students already struggling.

AI SEARCH IS GROWING FASTER THAN ITS BUSINESS MODEL

AI search is growing faster than platforms can monetize it, prompting Google and OpenAI to expand conversational ads. Google search queries grew at their fastest rate in 5 years while ad revenue growth slowed to 12%. More ads could erode trust in AI-generated answers, which 70% of regular users in the US and UK currently trust.

CHUNKING YOUR CONTENT FOR AGENT READABILITY

Chunking breaks documentation into smaller passages that AI agents retrieve independently, so each section needs to make sense without the full page. Keep sections under your retrieval system's token limit (Google's Agent Search defaults to 500 tokens) and use specific headings with proper HTML tags. Break up long tables and procedures into smaller sections. Keep code samples self-contained and describe important images in text. Test how your content is actually chunked using an agent skill or retrieval system API.

RUN YOUR TONGUE ACROSS YOUR TEETH

Claude Hopkins built Pepsodent's campaign around a problem customers could recognize, using a 3 second test to demonstrate sticky film on teeth. He offered free samples with instructions to test the toothpaste's effectiveness. Unique coupon codes helped track headline performance and revealed that leading with “free” reduced response by 75%. He also tested campaigns in one city before expanding nationally.

HOW TO TEST INCREMENTALITY BEFORE PEAK SPENDING STARTS

Run incrementality tests before peak season to identify which marketing channels actually generate new sales rather than taking credit for purchases that would happen anyway. Use geographic or audience holdouts to compare exposed customers against groups that don't see a channel. Aim for 20-30% holdouts and run tests for 4-8 weeks. Audit UTMs and tracking before launch. Use incremental ROAS and new customer acquisition to reallocate budget toward channels that drive genuine lift rather than those reporting the highest last-click revenue.

JOIN THE WEBINAR

HOW WE RUN OUR NEWSLETTERS ON BEEHIIV'S MCP

Strategy Breakdowns uses beehiiv's MCP integration with Claude Code to automate newsletter production. Claude analyzes hundreds of past A/B tests to identify winning subject line patterns. Testing 4 variations before sending to the full list increased open rates by at least 1.24 percentage points. It also segments subscribers and researches topics using trusted chart sources. A custom dashboard called SB Dash connects these workflows and automates sponsor reporting. This cut production tasks from 2-3 hours per edition to 10 minutes a day. Human editors still write every word.

THE WEBSITE GO‑LIVE CHECKLIST: 5 CRITICAL CHECKS BEYOND GOOD LOOKS

Before launching a website, review it from the customer's perspective to ensure the offer is clear and visitors can easily find information and take action. Check that the design works on mobile without unnecessary scrolling or distracting elements. Score clarity, branding, and usability against specific criteria rather than relying on subjective feedback. Use tools like Loom or Tella to record and track revisions.

NOBODY YET KNOWS HOW TO MEASURE AN AD INSIDE A CHATBOT ANSWER

Ads in ChatGPT and Google AI Mode reach consumers while they're expressing purchase intent, but neither platform can prove they generate incremental sales. Conversion tracking shows which purchases follow an ad but not whether they would have happened anyway. Facebook research found observational measurement can misstate results by 3X or more. Treat AI ads as an experimental channel and use geographic holdouts to measure their actual impact until platforms offer randomized lift studies.

WHY AI FAILS IN MARKETING: THE ORGANIZATIONAL PROBLEM LEADERS OVERLOOK

AI in marketing mostly fails for organizational reasons, not technical ones. Marketing owns revenue and retention targets but rarely controls pricing or product, so AI insights go unused. The highest-value uses also depend on sales, finance, and product sharing goals and data. Judged on cost savings alone, marketing AI tools just makes ineffective activities cheaper, not more effective.

CLAY'S $1-100M ARR MARKETING + GROWTH PLAYBOOK

Clay grew enterprise ARR 4.7X by attracting new customers through a self-reinforcing UGC loop.

19 LINKEDIN CONTENT DISTRIBUTION TIPS BEHIND MY 70M VIEWS

Plan distribution upfront, repost strategically, activate employee profiles, and leverage collaborations and ads.