AI drains search data as Zara's death pants go viral

The EU’s Digital Markets Act now forces Google to hand over anonymized search query, click, view, and ranking data to eligible rival engines and search‑enabled AI chatbots. The data, stripped of personal identifiers and Google’s proprietary algorithms, is meant to let competitors build their own retrieval systems and improve AI grounding.

Branded search traffic dropped 11.1% last month even though overall search volume stayed flat. The dip lines up with AI Overviews now handling 95.9% of commercial queries, answering users before they click a brand. Marketers can’t treat branded search as a clean demand proxy any longer.

Vendor tests show AI tools (ChatGPT, Gemini, Perplexity) give wrong facts for 64% of UK high street retailers, often misassigning postcodes or listing closed businesses. These inaccurate AI overviews appear in many local queries, meaning customers might get false info before you can correct it. Auditing across five platforms is essential.

Ann Handley debunks the “end of reading” narrative with data showing print‑book sales are higher than a decade ago and nearly 400 independent bookstores opened in 2025. Readers now spend more time on books daily and long‑form subscriptions are surging, meaning audiences still crave deep content.

Netflix is bundling live sports, third‑party TV channels, and short‑form video into its platform to shift from occasional binge‑watching to a daily habit. The move aims to boost ad revenue, keep subscribers engaged, and counter competition from YouTube, free services, and traditional broadcasters.

Short‑form videos of shoppers face‑planting in Zara’s $45 flow‑y wide‑leg trousers have racked up millions of views under hashtags like #ZaraDeathTrousers. Injuries range from bruises to broken wrists, prompting calls to pull the style. The backlash shows how a design flaw can become a viral earned‑media catalyst for a brand.

Google will move Local Services Ads into the Google Ads UI as Performance Max pay‑per‑lead campaigns, starting August 2026. The shift centralizes budgeting, bidding and reporting but deprecates manual CPA bids, separate dashboards, and some callouts, forcing advertisers to adapt to a daily budget and unified lead inbox.

Most CTV incrementality tests break because they compare viewers to non‑viewers, conflating targeting with lift. The guide shows how to create a pre‑defined control group, pick the right randomization unit, and report uplift that survives CFO scrutiny. Follow the checklist to prove CTV’s true contribution to revenue.

AI has driven creative costs to near zero, flooding B2B ad channels with endless variants and sending CPMs sky‑high. The only sustainable edge now is a tightly‑curated, CRM‑enriched audience that you sync across Meta, Google, LinkedIn and Reddit, feeding real conversion data back into platform algorithms.

Marketers are pouring money into AI tools, but the cost is catching up to them faster than measurable gains. Agencies like PMG are capping daily token spend and wrestle with governance, while the industry still lacks solid proof of ROI.

Isolated AI chat tools force marketers to copy data between apps, wiping out productivity gains. Embedding generative models directly into core systems lets data flow natively, automates personalized messaging, cross‑platform campaign updates, and compliance checks, turning AI hype into measurable enterprise ROI.

A top document‑editing service saw a 39% lift in quote requests after scrapping its prominent ‘Get Quote’ button on the homepage. Without the button, first‑time visitors were steered through service pages, gaining context before encountering the form, cutting friction and raising conversion rates. The test shows that visible CTAs can backfire for new users.