The feed
Martech news and analysis. Every story says what changed, why it matters and what marketers should do next.
On October 14, 2026 attorneys general from 39 states and DC filed a 28-page notice in Character.ai's Delaware bankruptcy case, setting nine conditions for any buyer of the app or user data. Eleven states did not join.
Why it matters
User chats can contain intimate information, and a bankruptcy sale is a rare moment when it could change hands.
noyb's Annual Report 2025, published October 13, 2026, reports 21 new complaints, 460 pending cases and a surplus of about €280,630. CEO Max Schrems says work on the EU Digital Omnibus limited resources for other matters.
Why it matters
noyb's complaints have driven major rulings in ad tech and consent, so its capacity affects enforcement risk for marketers.
On October 13, 2026 the European Data Protection Board adopted an eight-page preliminary statement on smart glasses. It says makers and AI service providers may be controllers and a terms clause asking wearers to inform bystanders is not enough.
Why it matters
Camera glasses record bystanders who cannot consent, and any product that sends such data to AI services will face these questions.
Norway's data protection authority fined Xandr NOK 92 million (about €8.5 million) in a decision dated October 2, 2026 and made public October 12. It found Xandr handled Grindr users' special category data without valid consent.
Why it matters
It shows that platforms in the bidding chain can be held responsible as joint controllers for consent they did not verify.
A regulation published October 10, 2026 takes effect November 1. It bars under-15s from social media and requires a state token confirming only whether a user is 15+ or 18+; existing accounts must be verified by March 31, 2027.
Why it matters
State-issued tokens are a different model from platform self-checks, and other countries may copy it.
Salesforce announced on October 13, 2026 that it will buy Loops, an AI analytics startup that explains why business metrics changed. The price was not disclosed and closing is expected in early November.
Why it matters
It fits Salesforce's push for proactive, agent-driven analytics, and it makes driver analysis a likely feature in Tableau.
A Lumen Research study with Amazon Ads and Cint of 270 campaigns finds a median 47 seconds of cumulative attention for purchase intent. The full white paper is gated, and the method and markets are undisclosed.
Why it matters
If attention thresholds hold, frequency and reach plans could be judged by cumulative seconds looked at, not impressions served.
A marketing mix study announced October 7, 2026 by Circana and Best Buy Ads says Best Buy Ads' ROAS was 79% higher than other media, and nearly 16% of associated sales occurred at competing retailers. The method was not published.
Why it matters
Retail media measurement often stops at the retailer's own transactions, so a cross-retailer figure changes how halo effects are valued.
Google's October Demand Gen update claims over 20% average conversion uplift for advertisers with tag gateway, and adds a Data Strength Uplift metric. The footnote for the figure was not visible, and The Zebra reports 16% more conversion volume.
Why it matters
Better signal can raise reported conversions without raising sales, so teams need to separate measurement gains from real demand.
DoubleVerify's "Scroll, Stop and Shop" report, emailed October 14, 2026, surveyed 22,000 consumers in 21 countries and 2,020 marketers in 25. Social purchasers fell to 25% from 30% and influencer influence to 48% from 54%.
Why it matters
Social commerce budgets often assume rising purchase behaviour, and a survey-based decline is a reason to check your own data.
