Paid Media News (Week 43)
Paid Media News: Week 43
This week: 81 updates, observations, references and events. Start with PMax video settings, customer-value bidding, and ChatGPT creative enhancements, then work through the channels that matter to your accounts. This archive presents what was reported in that edition; it does not establish current product availability. Original visuals remain at the source.
A digest of the original edition. Updates and availability reflect what was reported at the time.
Performance Max sourced videos become an opt-out setting
Default change · opt out by 8 November Google’s sourced-video setting is enabled by default for new Performance Max campaigns and may have been activated on some existing campaigns. It can add up to ten videos per asset group from approved landing-page, social and YouTube sources. Anthony Higman shared an account notice giving an 8 November opt-out date. Google’s help page also requires source approval and describes a staged rollout. Check both the toggle and approved URLs, especially where organic clips use music, talent or material without paid-media rights. Google Ads Help · Reported by: Anthony Higman · 20 Oct 26
Read the original edition →Channel prioritisation and budget caps are live in accounts outside the US
Rollout note · follow-up to Week 42 Thomas Eccel reports the channel prioritisation and budget allocation controls from Week 42 showing in many accounts outside the US, at account level under settings and overridable per campaign for Performance Max, Shopping and Search, with Demand Gen to follow; PPC News Feed confirms the sighting. Google has not published a geographic rollout schedule. The cap on Search Partners or Display may already be available to you, so check Account settings before building workarounds. PPC News Feed · 🥇 First contributor: Thomas Eccel · 21 Oct 26
Read the original edition →Merchant Center notice points to passkeys for sensitive actions from 29 October
Advertiser notice · from 29 October Emmanuel Flossie reports a Merchant Center notice introducing passkey checks for sensitive account actions from 29 October, following similar controls in Google Ads. The report does not establish a complete list of affected actions or that every account will change on the same day. Check the notice in your own account and ensure authorised users can authenticate before they need to make an urgent change. Avoid relying on one shared login or assuming every feed edit will require a passkey. LinkedIn · 🥇 First contributor: Emmanuel Flossie · 22 Oct 26
Read the original edition →Data Manager will stop requiring hashed customer data from 16 November
Advertiser notice · effective 16 November · catch-up A Google notice says that from 16 November Google Ads Data Manager will accept encrypted rather than SHA-256-hashed customer data by default for CRM connections, cloud imports, file uploads and the API, with decryption, normalisation and matching done inside a hardware-isolated trusted execution environment; it applies to Customer Match and enhanced conversions. The confidential-matching help page still tells you to hash, and the notice does not say how existing hashed uploads, key custody or opt-outs work. The hashing step in your upload pipeline becomes optional rather than wrong, so wait for the help page before changing it. LinkedIn · 🥇 First contributor: Anthony Higman · 18 Oct 26
Read the original edition →Image extraction switches on for every new Performance Max campaign from 4 December
Announcement · API · effective 4 December From 4 December 2026, newly created Performance Max campaigns will have image extraction set to opted in by default across the Google Ads API, the interface and Google Ads Editor, even where Dynamic Image Extensions are off at account level; existing campaigns keep their setting. To opt out at creation, API users set the asset automation setting to OPTED_OUT, and because the field is repeated, any other automation values must be resent or they clear. Google reminds advertisers they must hold rights to landing-page images. Campaign templates built before December will start pulling site images into ads, so update your creation scripts and briefs now. Google Ads Developer Blog · Reported by: @Dora Sun · 21 Oct 26
Read the original edition →New-customer value can now be a percentage, and retention modes reach Search and Shopping
Announcement · rolling out globally this month Google is expanding customer lifecycle goals with percentage-based new-customer values, High Value mode for Shopping, and retention modes across Performance Max, Search and Shopping. The new multiplier can range from 1.1x to 5x, allowing the value assigned to a new customer to scale with the order rather than use a fixed bonus. The changes are rolling out this month. Review the resulting conversion values against your customer economics before replacing a fixed adjustment; a multiplier should reflect the additional value of acquiring that customer. Google · Official clarification: Ginny Marvin · 22 Oct 26
Read the original edition →Smart Bidding Exploration is documented for Performance Max and Shopping
Documentation · rollout of a May announcement A new help page explains how Smart Bidding Exploration works beyond Search: it needs a target ROAS, lets the system bid 5–30% below target to find new converting queries, covers Search and Shopping traffic with one toggle on feed-based Performance Max, and Search traffic only on campaigns without a feed. Campaigns should not be budget-limited, and moving Shopping or Performance Max into a portfolio strategy turns exploration off. Google announced the Shopping expansion at Marketing Live in May. The ROAS tolerance now applies to Shopping queries too, so set the target with that headroom in mind. Google Ads Help · Reported by: Hana Kobzová · 22 Oct 26
Read the original edition →Google rewrites its Smart Bidding learning page: the status "never lasts more than 7 days"
Documentation · follow-up to Week 42 The help page formerly titled "Duration of the learning period" is now "About Smart Bidding learning" and states that Learning status "never lasts more than 7 days". It lists what triggers it (a new or reactivated strategy, campaigns, ad groups or keywords added to or removed from a strategy, a significant conversion-goal change) and what does not (budgets, targets, creatives and targeting settings). This puts in writing what Tal Akabas told Ginny Marvin on Ads Decoded last week. The status label is not a guarantee of stable results after a week. Record major changes and allow for conversion delay when evaluating performance, even if the interface no longer shows Learning. Google Ads Help · Reported by: Barry Schwartz · 21 Oct 26
Read the original edition →Seasonality adjustments now list Demand Gen
Documentation Google's seasonality adjustments page now says the feature is available for Search, Shopping, Demand Gen and Display campaigns using Target ROAS or Target CPA, and for Performance Max and App campaigns in beta with any strategy; Travel campaigns remain excluded. Use the adjustment for events of one to seven days rather than long sales. Google Ads Help · Reported by: Hana Kobzová · 21 Oct 26
Read the original edition →Demand Gen cost adjustments can last up to 60 days
Documentation follow-up Google’s current help page says the performance-based cost adjustment for eligible Demand Gen Target CPA and Target ROAS campaigns can remain in effect for up to 60 days, subject to duration and adjustment limits. This is an eligibility-limited beta, not a guaranteed refund whenever a campaign misses its target. Check the applicable conditions before including an adjustment in a campaign forecast. Google Ads Help · Reported by: Anu Adegbola , via PPC News Feed · 25 Oct 26
Read the original edition →Display Expansion will move to channel prioritisation in February
Documentation update Google’s updated documentation says Search campaigns using Display Expansion will automatically move to channel prioritisation in February 2027. Advertisers can already set High or Low priority and optional daily allocation caps. Google says a Low priority setting with a 50% daily budget allocation preserves the previous Display configuration. Check account defaults and campaign overrides before the transition, because the former automatic allocation will disappear. Google Ads Help · Reported by: Hana Kobzová · 23 Oct 26
Read the original edition →Performance Max asset groups gain a "Save changes as an A/B test" toggle
Spotted in the Wild · account observation Joey Bidner spotted a “Save asset group changes as an A/B test” toggle in the Performance Max asset editor. The screen says the asset group cannot be edited while the experiment runs. The control offers a way to compare changes to headlines, descriptions, images and videos within the asset workflow. Prepare a complete alternative creative set and a clear test question before starting; this remains an account observation rather than confirmation of access for every advertiser. LinkedIn · 🥇 First contributor: Joey Bidner · 19 Oct 26
Read the original edition →AI Max appears auto-enabled in some Google Ad Grants accounts
Spotted in the Wild · catch-up PPC News Feed reports, from Dan Kabakov, Ad Grants accounts in which AI Max shows as enabled without the manager switching it on; Search Engine Land followed on 20 October. Google has not commented, and Ad Grants accounts have their own policy limits on bidding and quality. Grant accounts need the same AI Max audit as paid ones, so check each campaign's setting and its search terms before the next policy review. PPC News Feed · 🥇 First contributor: Dan Kabakov · 18 Oct 26
Read the original edition →Google Ads Translation appears in the Tools menu
Account observation · availability follow-up Jyll Saskin Gales reports that Google Ads Translation is appearing in the Tools menu, following Dario Zannoni’s earlier coverage of the documented beta. The workflow can translate campaign text and keywords, replace text in images and prepare localised landing pages. Translated campaigns are created paused, while copied budgets need a separate currency check. This is a useful account-availability update rather than proof that every advertiser has access. Have a fluent reviewer check the offer, keywords and destination before launching. Google Ads Help · Reported by: Jyll Saskin Gales · 21 Oct 26
Read the original edition →Fast delivery annotations now have per-country day thresholds
Documentation The "About fast delivery" page now lists the maximum delivery days that qualify in each country: one day in Japan and New Zealand, two in India, three in Australia, Canada, the UK and the US, and four in Brazil. Annotations can be enabled for Australia, Canada, India, Japan, the UK and the US, and the "Get it fast" filter remains US-only. Compare your delivery promises and shipping data with the rules for each target market. Meeting a day threshold alone does not guarantee that an annotation will appear. Google Merchant Center Help · Reported by: Hana Kobzová · 20 Oct 26
Read the original edition →Demand Gen accepts HTML5 ads, but serves them only on the Display Network
Documentation The HTML5 help page now says HTML5 ads can be connected to Demand Gen campaigns, where they serve exclusively on the Google Display Network and not on YouTube, Discover, Maps or Gmail. The upload eligibility rules are unchanged: an account older than 90 days with more than $9,000 lifetime spend and a good policy history, or an application with more than $1,000 spend under extra criteria. Rich-media creative can join Demand Gen without reaching its video surfaces, so weigh that against the extra production cost. Google Ads Help · Reported by: Hana Kobzová · 22 Oct 26
Read the original edition →Google is testing ads in AI Mode for people who connect their apps
Reported test · catch-up Google told Search Engine Land on 16 October that it is "starting to test ads in AI Mode in Search for users who connect their apps through Personal Intelligence". Its help page says ads "will not be based directly on your Gmail inbox or your Google Photos library", but summaries and inferences from connected apps may be used, sensitive categories are excluded and advertisers get no access; the privacy policy effective 16 October adds a matching qualifier. In March Google had said there were no ads for connected-app users. Nothing changes in targeting settings, so note it for client questions about AI Mode placements. Google Help · Reported by: Anu Adegbola (Search Engine Land) · 16 Oct 26
Read the original edition →Google publishes a Commerce Media partner directory
Documentation A help page now lists ten commerce media partners whose inventory can be bought through Google Ads, with onboarding links for each: Altex (Romania), Blinkit, Instamart, Myntra and Zepto (India), Drogasil and Droga Raia (Brazil), Noon (UAE), PCHome (Taiwan) and Shopee (Southeast Asia). The page says the list may change and does not describe formats. Retail media in those markets can be planned from Google Ads, so check the onboarding terms with each partner before committing budget. Google Ads Help · Reported by: Hana Kobzová · 22 Oct 26
Read the original edition →Hotel advertisers get a formula for length-of-booking bid adjustments
Documentation · niche A new help page explains that legacy Hotel campaigns multiply the base bid by length of stay automatically, while Search campaigns for travel do not, so advertisers migrating need a length-of-booking bid adjustment; it gives the conversion formula and worked examples. The change ties to the API v26 renaming of hotel fields to a vertical-agnostic length_of_booking. A migrated travel campaign can under-bid on long stays, so recalculate adjustments before moving budgets. Google Ads Help · Reported by: Barry Schwartz · 20 Oct 26
Read the original edition →Financial Services Verification will extend to the US and Canada
Policy · rollout early 2027 Google will require advertisers running certain financial services ads in the United States and Canada to show they are authorised by a relevant regulator or recognised registry, extending a programme it says already covers 42 countries. It is working with the American Bankers Association registry in the US and engaging the Canadian Bankers Association, with rollout "planned for early 2027"; accepted registries, credentials and onboarding steps will be published before enforcement, and affected advertisers will be notified. For you, this means North American finance accounts need a verification owner before Q1, so check which entity holds your licences now. Google Ads & Commerce Blog · Official author: Keerat Sharma · 20 Oct 26
Read the original edition →Local Services Ads drop "featured professionals" and make headshots optional
Advertiser notice An in-account notice says that "starting in October, featured professionals will no longer show on your ads" and that a headshot "is optional and won't affect ad serving"; photos and reviews are unchanged. Anthony Higman posted the notice in the first hour of Monday 19 October (UK time), after an earlier post on a headshot test, and Search Engine Roundtable and PPC News Feed followed. Google's LSA help page had not been updated. The person-led ad unit is going, so review which trust signals remain in your profile and keep review volume up. Search Engine Roundtable · 🥇 First contributor: Anthony Higman · 19 Oct 26
Read the original edition →An LSA-style result appears beside the local map pack for injury searches
Account observation Anthony Higman shared a mobile personal-injury search result showing sponsored lawyer listings beside the local map area. He identifies the placement as an LSA, but one screenshot cannot establish that Google has replaced map-pack ads nationally. The observation is worth checking for legal advertisers: record the query, location, device and visible ad labels when comparing results. Keep the placement sighting separate from claims about lead quality or campaign performance. LinkedIn · 🥇 First contributor: Anthony Higman · 22 Oct 26
Read the original edition →Campaign creation can start from a text prompt in some accounts
Follow-up · account observation Some accounts see a new first step in campaign creation: enter a website URL, describe the campaign goal in plain language and let Google AI suggest goals and assets, with "Create campaign manually" still available. Dima (@skills_web) posted the test on 9 October and Ginny Marvin confirmed it as an experiment the same day; Vivek Gupta saw it in his account this week and notes it is still rolling out. Junior team members may be guided into AI-suggested goals, so agree a review step before any prompt-built campaign goes live. Search Engine Roundtable · 🥇 First contributor: Dima (@skills_web, X) · 22 Oct 26 · This week’s account example: Vivek Gupta
Read the original edition →Google documents AI review summaries in Search ads for four markets
Documentation · follow-up to a September test A new help page describes AI-generated review summaries that can appear with Search ads in the US, the UK, Australia and India in English, on all devices and positions, across verticals except sensitive ones such as pharmaceuticals. The summaries are positive-only, cannot be edited by the advertiser, show only when Google predicts they will improve performance, and the page describes no opt-out. Sachin Patel reported the test in September. Review language you did not write may sit beside your ad, so keep your rated reviews current in Merchant Center and third-party sources. Google Ads Help · Reported by: Hana Kobzová · 24 Oct 26
Read the original edition →Shopping ads test product features alongside the offer
Account observation Sachin Patel spotted Shopping ads displaying key product features on a grey background alongside the product image, name, price and rating. His example came from a search for home ramps. The additional details could change which information shoppers see before clicking, but this is a layout observation, not a new advertiser-controlled format or a confirmed rollout. Search Engine Watch · 🥇 First contributor: Sachin Patel · 21 Oct 26
Read the original edition →Google tests scrolling reviews with a “Visit Site” button
Account observation Alex Greenland’s example shows ratings and reviews moving in a scrolling strip beside a “Visit Site” call to action in a sponsored Search result. PPC News Feed credits Greenland and Barry Schwartz for the report. It is a format sighting, with no confirmed advertiser control or general availability. PPC News Feed · 🥇 First contributor: Alex Greenland · 24 Oct 26
Read the original edition →A "Policy summary" page appears under Admin > Policy
Spotted in the Wild · account observation PPC News Feed reports a new Admin > Policy > Summary page that gathers account-level policy issues and ad or asset-group issues with eligibility status and fix or dismiss actions, sortable by cost, conversions and ROAS; Policy Manager remains under Tools > Troubleshooting. We found no help page for it, and the certification applications that Week 42 reported moving into Admin > Policy sit in the same area. Policy triage can be ordered by spend, so start the weekly check there if your account has it. PPC News Feed · 🥇 First contributor: Hana Kobzová · 24 Oct 26
Read the original edition →Conversion action statuses get icons
Spotted in the Wild · account observation Google Ads conversion actions are showing icons alongside their status, including an hourglass for Awaiting conversions. Adriaan Dekker credited Arpan Banerjee for the discovery, and Jhony Ruiz shared a later example. The underlying status names were introduced earlier; this observation concerns their presentation. The icons can make the table easier to scan, but an Awaiting conversions label alone does not diagnose a faulty tag. Check the action’s recent activity and diagnostics before changing the implementation. PPC News Feed · 🥇 First contributor: Arpan Banerjee (credited by Adriaan Dekker) · 19 Oct 26
Read the original edition →Google Business Home brings Business Profile, Merchant Center, YouTube and Ads into one place
Closed beta · US and UK Google's help page describes Business Home as "a closed beta experience, available only to select business categories in the US and the UK" for verified profiles with fewer than ten locations, linking one YouTube channel, one Google Ads account and one Merchant Center account, with AI content tools, a "360° business view with prioritized growth steps" and RCS messaging. Barry Schwartz received an invitation email on 22 October; a sighting on X dates to 25 September. Small-business clients may soon see a Google dashboard that recommends ad actions, so ask who will own those prompts. Google Help · Reported by: Barry Schwartz · 22 Oct 26
Read the original edition →UAE cryptocurrency and Colombian lottery policies took effect on 21 October
Policy · effective 21 October Two policy entries took effect this week: in the UAE, cryptocurrency exchanges and wallets need licences from the CMA, DFSA, FSRA or VARA, and CFD or rolling spot forex providers from the CMA, DFSA or FSRA, with a seven-day warning before suspension; in Colombia, online lottery ads are allowed from operators authorised by CNJSA. Both were posted earlier in October. Existing UAE crypto and Colombian lottery campaigns now need their certification on file, so submit it before the warning period expires. Google Ads Policy · 21 Oct 26
Read the original edition →Microsoft rewrites its surveillance policy as "Surveillance Products and Services"
Policy · rewritten Microsoft Advertising has renamed its "Surveillance Equipment" policy "Surveillance Products and Services" and broadened it. Prohibited: ads or landing pages suggesting illegal use such as stalking, keywords like "secret tracking app" or "spy cam", hidden recording devices for unauthorised surveillance, keyloggers and tools for intercepting communications. Permitted: home, property and business security systems, trackers for valuables, vehicles, luggage or pets, and child- and family-safety products. No effective date or market list is given. Security and tracker advertisers should re-read their keyword lists against the new examples, so remove anything that implies covert use. Microsoft Advertising Policy · 🥇 First contributor: Hana Kobzová · 20 Oct 26
Read the original edition →Three Microsoft-network ad tests: a CTA button in Copilot ads, reviews and popular links on Yahoo, and sponsored results at the very top of Bing
Spotted in the Wild · tests Three format sightings on Microsoft-served search this week and last. Ad units on Bing, Copilot and Yahoo are diverging from Google's, so check how your sitelinks and reviews render there before the holiday period. "Learn more" button on the left of sponsored results inside Bing Copilot answers : Sachin Patel, Search Engine Watch, 13 October; PPC News Feed, 21 October. A review line under a shopping ad and a "popular links" unit on Yahoo search ads : Sachin Patel, Search Engine Watch, 21 October; follows the Bing "Popular Links" test in Week 42. Sponsored ads placed at the very top of the Bing results page : Khushal Bherwani on X, 12 October; PPC News Feed, 20 October. Not covered in Week 42. Search Engine Watch · 🥇 First contributor: Sachin Patel · Khushal Bherwani · 21 Oct 26
Read the original edition →OpenAI will switch on creative enhancements for every ChatGPT Ads campaign unless you opt out
Default-setting change · conflicting start dates OpenAI plans to enable creative enhancements for new and existing ChatGPT Ads campaigns while preserving existing opt-outs. The feature can rewrite text, translate copy, resize images and adjust destination URLs without approval of every variation. Its current help page gives 31 October, but pairs it with an incorrect weekday; earlier reporting cited 30 October. Check the setting now rather than relying on the conflicting deadline. Review landing-page claims and decide whether campaign-level overrides should differ from the account default. OpenAI Help · 🥇 First contributor: Thomas Eccel · 20 Oct 26
Read the original edition →ChatGPT Ads expands to 26 more countries and passes 100,000 advertisers
Announcement · 89 countries OpenAI announced ChatGPT Ads availability in 26 additional countries, bringing the total to 89, and says more than 100,000 advertisers have run campaigns since launch. The expansion includes new Latin American, African, European and Asian markets. For teams considering a test, confirm advertiser eligibility, target-country access and budget requirements in Ads Manager. Platform-wide advertiser adoption is useful context, but it does not establish performance in a particular market. OpenAI · 21 Oct 26
Read the original edition →ChatGPT Ads offers location groups such as "European Economic Area + Switzerland"
Spotted in the Wild · account observation Thomas Eccel's screenshot shows a location-group option in ChatGPT Ads targeting that adds the EEA and Switzerland in one selection. OpenAI's targeting documentation describes country codes and location IDs but no preset groups, so this is an interface sighting without documentation. Pan-European campaigns can be set up faster, so check the resulting country list before launch rather than assuming it matches your legal footprint. LinkedIn · 🥇 First contributor: Thomas Eccel · 22 Oct 26
Read the original edition →ChatGPT ad observations show a borderless unit and a new Learn more treatment
Spotted in the Wild · format Juozas Kaziukėnas shared a ChatGPT ad layout without the previous surrounding border. SERP Alert separately documented a Learn more link with an arrow and dotted underline. These are presentation observations rather than an announced format rollout. Keep paid destinations consistently tagged so you can distinguish ad traffic from organic ChatGPT referrals; the visual change alone does not alter how an analytics platform classifies a visit. LinkedIn · 🥇 First contributor: Juozas Kaziukėnas · 19 Oct 26
Read the original edition →OpenAI confirmed the ChatGPT UTM drop was a bug and rolled out a fix on 15 October
Official clarification · follow-up to Week 42 Search Engine Land reports that after Adobe's Cedric Huesler contacted OpenAI on 14 October, OpenAI confirmed the missing utm_source=chatgpt.com parameter was "a known bug, with a fix expected that week"; the fix began on 15 October and tagged clicks returned to roughly prior-week levels by that morning. OpenAI has published nothing itself. Week 42 covered the gap, which ran from 8 October. Annotate the affected reporting window and check source/medium patterns. Missing UTMs can obscure attribution, but do not assume that every affected visit was classified as Direct. Search Engine Land · Reported by: Anu Adegbola, mMBA (Search Engine Land) · 20 Oct 26
Read the original edition →Analytics Debugger adds DataLayerDB to save and replay dataLayer pushes
Resource David Vallejo's Analytics Debugger browser extension can now save any dataLayer state or push and replay it on demand, so a purchase event can be tested repeatedly without completing another checkout. The feature is demonstrated in a short video; pricing tiers for the extension are not stated in the post. Saved events can speed up tag debugging, but replaying a dataLayer push does not replace an end-to-end checkout test. LinkedIn · Official author: David Vallejo · 19 Oct 26
Read the original edition →Ginny Marvin publishes her SMX Advanced Europe measurement framework
Resource Google's Ads Liaison has written up her SMX Advanced Europe talk as a roadmap built on three questions: impact (what the media budget achieved), optimisation (where the next pound should go) and profitability (whether spend matches business goals), with her slides attached. It is a framework rather than a product announcement. A ready structure for a measurement review, so map your current tools to the three questions before the next planning cycle. LinkedIn · Official author: Ginny Marvin · 23 Oct 26
Read the original edition →AI referrals remain small in Marie Haynes’s 100-site analysis
Research study Marie Haynes reviewed 12 months of GA4 data across 100 sites and more than 145 million sessions. In that sample, ChatGPT sent 0.62% as many sessions as Google organic search, while the recorded conversion rates were 13.8% for ChatGPT and 3.02% for Google organic. The figures describe this dataset, not a universal benchmark. Compare AI referrals with your own acquisition mix, conversion definitions and sample sizes before moving budget on the strength of a headline. Marie Haynes · Research by: Marie Haynes · 21 Oct 26
Read the original edition →Meta introduces on-device conversion matching with a staged November rollout
Advertiser notice · default on from 12 November · not in the EU or UK An Events Manager notice shared by Dario Zannoni describes a gradual rollout of “Measure conversions on customer devices” from 12 November, with the setting enabled by default for eligible use. The notice describes matching on the customer’s device and sending attribution results to Meta. Treat the availability and exclusions in your own account as the implementation reference. A change in attributed conversions need not mean a change in actual sales, so preserve a reporting baseline and check consent and pixel settings. LinkedIn · 🥇 First contributor: Dario Zannoni · 19 Oct 26
Read the original edition →Choosing an audience per ad creative is rolling out with documentation
Rollout · follow-up to Week 34 Jon Loomer reports that the ad-level audience option that Bram Van der Hallen spotted as a test in August now appears across every performance goal he tried, with a Meta help page: a toggle at ad-set level further limits reach, after which each creative gets its own age and gender setting. It is on by default for Sales once reach is limited and off for other objectives, and Loomer calls it a rollout rather than a test. One ad set can carry age-specific creative without splitting, so test it on a campaign with clearly segmented offers first. Jon Loomer · 20 Oct 26
Read the original edition →Ads Manager shows the top products in a catalogue campaign
Spotted in the Wild · account observation Bram Van der Hallen reports that product catalogue campaigns now open a "Top products in your campaign" view inside the performance charts, listing product amount spent, clicks, attributed orders and their value, views, purchases and purchase value, with a click-through to a "Product insights" dashboard and the same view per ad set. Availability beyond the observed account is unconfirmed. Catalogue winners are visible without a product-level breakdown export, so check which SKUs absorb spend before the peak season. LinkedIn · 🥇 First contributor: Bram Van der Hallen · 22 Oct 26
Read the original edition →Hovering over Reach at ad-set level opens audience insights, including interests you never targeted
Spotted in the Wild · account observation Bram Van der Hallen also reports an audience insights panel on the Reach metric at ad-set level, showing reach against estimated audience size, over-represented demographics and the reached audience's top interests relative to all Meta accounts over the past seven days, even when the ad set uses no interest targeting. A linked dashboard adds a bubble chart that groups age and gender segments by cost per result and CTR, with Meta's creative suggestion for each quadrant. Use the panel to develop creative hypotheses, then test them. An audience characteristic reported after delivery is not evidence that targeting that interest would improve results. LinkedIn · 🥇 First contributor: Bram Van der Hallen · 23 Oct 26
Read the original edition →Two Ads Manager interface changes: comparison with similar businesses, and clearer creative-enhancement controls
Spotted in the Wild · account observation Two interface observations arrived from practitioners this week. For you, this means Ads Manager is surfacing more benchmarks and defaults, so check what is switched on before you read the numbers. A comparison with similar businesses inside cost per result, showing all metrics in one panel rather than metric by metric : Matias Axel Ramayo, 19 October. Account observation. A green dot beside each enabled creative enhancement, a "Review details" button for default settings, a revised media upload interface and fewer enhancements on by default : Bertrand Vanel, 21 October; follows the status dots Bram Van der Hallen reported in Week 41. Account observation. LinkedIn · 🥇 First contributor: Matias Axel Ramayo · Bertrand Vanel · 21 Oct 26
Read the original edition →LinkedIn CTV Ads can now be bought through Display & Video 360
Announcement · US targeting only LinkedIn's connected-TV inventory is now available in Google's Display & Video 360, letting DV360 buyers apply LinkedIn's first-party professional targeting such as job title, industry and seniority to streaming inventory inside existing workflows. It is "currently available for US targeting only", accessed through a LinkedIn or Google representative, and delivery runs through Microsoft Monetize; LinkedIn cites its own Kantar study claiming 2.6x higher B2B awareness effectiveness. For you, this means B2B CTV can sit in the same DSP as your other video, so ask your Google representative about pricing before the next upfront. LinkedIn Pressroom · Reported by: Victoria McNally (AdExchanger) · 22 Oct 26
Read the original edition →Reddit adds a Campaign Planner and a campaign-level Optimization Score
Announcement · rolling out through October Reddit Ads Manager gains a Campaign Planner that predicts reach, impressions and cost before launch and shows audience activity and dominant subreddit themes, plus an Optimization Score at campaign level with tailored recommendations; both are "rolling out to all advertisers through October". Campaign budget optimisation is "now available", in beta for Sales, Lead generation and App promotion. Reddit's claim that advertisers scoring 80% or higher saw 19.6% lower blended CPA is a correlation with the methodology in footnotes. A forecast before spending, so treat the score as a checklist rather than a performance promise. Reddit for Business · Reported by: Andrew Hutchinson (Social Media Today) · 20 Oct 26
Read the original edition →Amazon seller notice introduces prepaid return-label requirements
Seller notice · effective 18 December A seller notice describes prepaid labels for US seller-fulfilled returns from 18 December, with a temporary exception for heavy and bulky items. Sellers would have seven calendar days to upload a label through Manage Returns or issue a refund without requiring the item back; an address-only label would no longer satisfy the requirement. Review the notice in Seller Central and the return workflow. Supplying a prepaid label does not, by itself, settle who ultimately bears its cost under every return circumstance. Amazon Central · 19 Oct 26
Read the original edition →Seller Fulfilled Prime in the UK will be judged by size group from 30 October
Seller notice · UK Oleksandr Kovalov reports an Amazon UK notice splitting Seller Fulfilled Prime products into standard, oversize and extra-large groups from 30 October, with Prime performance checked per group rather than per account. Two rules follow on 6 January 2027: standard-size products must show a one-day delivery date on at least 25% of page views, and extra-large products may ship with any tracked carrier if on-time delivery stays at 97% or above, with warning emails only until January. We could not open the Amazon notice. One slow size band can cost the Prime badge, so check your one-day delivery share per group now. LinkedIn · 🥇 First contributor: Oleksandr Kovalov · 19 Oct 26
Read the original edition →Amazon confirms its extended return window for 2026 holiday purchases
Policy · catch-up Amazon’s holiday notice covers purchases from 1 November through 31 December 2026, with returns generally accepted until 31 January 2027 and a shorter 15 January deadline for Apple products. The extended window applies to eligible FBA and seller-fulfilled orders, with other return conditions still relevant. Build the longer return period into stock planning and January reporting. Treat this as confirmation of the 2026 window rather than assume every future season is fixed permanently. Amazon Seller Central · 16 Oct 26
Read the original edition →A Delta Vacations ad in ChatGPT was bought through Amazon DSP
Spotted in the Wild · pilot Juozas Kaziukėnas spotted a Delta Vacations ad in ChatGPT whose click URL routes through Amazon's click-tracking server, the first live example he has seen of the Amazon Ads–OpenAI pilot announced on 10 September, when Delta Vacations was named as a first tester. The ad format is native ChatGPT; Amazon handles targeting, and neither company has described measurement. For you, this means ChatGPT inventory may reach your DSP plan before ChatGPT Ads self-serve does in your market, so ask your Amazon Ads contact whether the pilot is open. LinkedIn · 🥇 First contributor: Juozas Kaziukėnas · 23 Oct 26
Read the original edition →Global Warehousing and Distribution now accepts expiration-dated products
Seller announcement Amazon's seller news says products with expiration dates in categories such as Health and Personal Care, Beauty and Baby can now be stored in Global Warehousing and Distribution centres in Shenzhen, Shanghai or Ningbo and replenished to US FBA. Items need at least 365 days of shelf life when the inbound shipment is created and 180 days when they leave GWD; meltable and temperature-sensitive goods remain excluded. Amazon's post is undated but predates the week. Shelf life keeps running while stock sits in China, so plan replenishment so nothing ages out of FBA eligibility. Amazon Seller Central · 🥇 First contributor: Ivan Marynych · 19 Oct 26
Read the original edition →Walmart opens its catalogue to third-party shopping agents
Announcement · beta Walmart's Agent Gateway gives AI agents a Model Context Protocol route to shop on its platform, starting with a beta for companies and developers, and Agent Profiles, "launching in the next month", let customers authorise an agent with purchase approval or spending limits, see its purchases separately and delete it at any time. Walmart names Muse, ChatGPT and Gemini integrations alongside its own Sparky, says agents never see card details, and applies its normal returns process. The post does not mention advertising or Walmart Connect. Agent-placed orders will appear in Walmart data without a clear ad touchpoint, so ask Walmart Connect how it will attribute them. Walmart · Reported by: Sanskriti Shekhar (Reuters) · 21 Oct 26
Read the original edition →Walmart Connect is pitching on-site brand lift studies directly to brands
Industry insight Digiday reports Walmart's sales team contacting brand clients directly ahead of the holidays to offer on-site brand lift studies allocated by Walmart Connect spend, with results that can take up to nine weeks; advertisers still cannot run such studies independently, and Walmart did not comment. One agency source says Q4 is when advertisers are least willing to fund spend they cannot defend. Lift studies remain a negotiating chip tied to commitment, so ask for one in writing before agreeing holiday budgets. Modern Retail · Reported by: Kimeko McCoy (Digiday) · 23 Oct 26
Read the original edition →Shopify Rollouts schedule catalogue changes for a set date and time
Release note Shopify's changelog adds Rollouts, which schedule the activation, deactivation or switching of catalogues, so market-specific pricing and availability can change at a set moment, and let merchants test offers between catalogues and launch them together with theme and checkout changes. Plan and market availability are not stated. For you, this means Black Friday pricing can be staged in advance rather than edited live, so align the rollout time with your ad schedule. Shopify changelog · 23 Oct 26
Read the original edition →Researchers report regional differences in ChatGPT shopping retrieval
Research · analysis Metehan Yeşilyurt’s Peec AI research describes different product-retrieval routes in sampled US and European ChatGPT shopping responses. Juozas Kaziukėnas also observed a Shop at button favouring one retailer even where a cheaper option appeared elsewhere. These findings offer leads for inspecting product visibility and retailer selection; they do not establish an official or exhaustive ranking formula. Check feed accuracy and the offers shown for your products, without assuming feed submission guarantees inclusion or a preferred retailer position. LinkedIn · Research by: Metehan Yeşilyurt (Peec AI) · 20 Oct 26
Read the original edition →Apple explains the Safari 27 domain blocks: identity resolution infrastructure is the target
Official explanation · follow-up to Weeks 40 and 42 Apple's WebKit team has published the rationale for the Safari 27 network blocks that Weeks 40 and 42 covered: WebKit blocks "identity resolution infrastructure" at the domain level, across iOS, iPadOS, macOS and visionOS 27, aimed at third parties that collect email addresses at login or newsletter sign-up and turn them into cross-site tracking IDs. First-party personalisation is "not tracking", but a central resolver remains a third party even for one company's graph. AdExchanger reports The Trade Desk created TTDCDN.org as a backup domain. Email-based identity partners may keep dropping off Safari traffic, so ask each vendor for its Safari 27 match rates. WebKit · Official authors: John Wilander, Charlie Wolfe, Matthew Finkel (Apple) · 19 Oct 26
Read the original edition →Meta postpones revised commercial terms to 20 November
Follow-up to Week 42 · date moved Reuters reports that Meta has moved the effective date of its revised Commercial Terms and Self-Serve Ad Terms from 30 October to 20 November following publisher and advertiser objections. The dispute includes how the terms describe crawling business websites and using their content. PPC Land subsequently reported changes to the UK English preview wording. Review the version applicable to your business and the corresponding account notice; the postponement gives teams more time to assess the terms. Reuters · Reported by: Foo Yun Chee (Reuters) · 23 Oct 26
Read the original edition →FTC proposes conditions for Publicis’s LiveRamp acquisition
Regulatory · proposed order The FTC has proposed conditions for Publicis’s acquisition of LiveRamp, including five years of non-discriminatory access to standalone products, safeguards around customer data and independent oversight. The proposed consent order is subject to a public-comment process. Agencies using LiveRamp should follow the final order and its scope rather than assume a proposal is already a completed compliance regime. The practical question is whether access, data separation and service terms remain suitable for their existing arrangements. FTC · 19 Oct 26
Read the original edition →The European Commission is examining whether Booking.com labels paid rankings as ads
Regulatory · proceedings opened The Commission opened formal DSA proceedings against Booking.com on 21 October over fraudulent listings and phishing risks, and over Article 26(1): it "suspects that Booking.com did not clearly label as ads the listings whose providers paid for a higher ranking in search results". It will examine whether users can tell who is behind each ad, who paid and why they see it. Booking.com says it is cooperating fully; the opening does not prejudge the outcome. The case puts paid marketplace visibility and its labelling under scrutiny. It is an investigation, not a finding that Booking.com has broken the rules or a confirmed change to listings. European Commission · Reported by: Greta Ruffino (Euronews) · 21 Oct 26
Read the original edition →Meta's ad revenue grew 26% as prices per ad rose 14%
Earnings · Q3 2026 Meta reported Q3 advertising revenue of $63.061 billion, up 26% year on year, alongside 10% growth in ad impressions and a 14% rise in average price per ad. Bertrand Seguin shared analysis of the results. These are company-wide measures combining markets, objectives and inventory; they are not a forecast for an individual account’s CPM or acquisition cost. When reviewing holiday budgets, compare your own placement mix and conversion performance before attributing a cost increase to Meta’s headline average. Meta press release (SEC) · Reported by: Bertrand Seguin (App Economy Insights) · 21 Oct 26
Read the original edition →Muse has 10 million users in the US and Canada, and shopping is its main use
Earnings · AI assistant Mark Zuckerberg told investors that more than 10 million people in the US and Canada have set up Muse personal agents within six weeks of launch, about 25% use it on any given day, shopping including travel is the most common use case, with more than 25% of weekly actives shopping and an average purchase above $200, and that Meta will "get paid a fee based on the transactions that we help facilitate" sooner than its usual playbook. A commerce agent with a fee model is forming inside Meta, so watch for how product feeds and ads connect to it. Earnings call transcript · Analysis by: Juozas Kaziukėnas · 21 Oct 26
Read the original edition →A New Mexico judge ordered Meta to pay $25 billion in civil penalties
Court order · appeal pending A New Mexico court ordered Meta to pay $25 billion in civil penalties following a jury finding concerning representations about user data and content moderation. AP and the Santa Fe New Mexican reported the judgment; Thomas Höppner added legal commentary. Meta has said it will appeal. This is a significant legal development to follow, but the judgment does not itself announce a new advertiser setting, campaign restriction or immediate operational change in Ads Manager. AP via ABC News · Reported by: Mikella Schuettler (AP) · 21 Oct 26
Read the original edition →DV360's November changes: Brand Control Center retires and Floodlight uses signed-in signals for cross-device attribution
Documentation · November changes · reminder for 26 October DV360's announcements page, updated on 16 October, lists November changes: the advertiser-level Brand Control Center is deprecated and its settings move to targeting rules; the "Amerilink" provider is renamed "Open Audiences"; and Floodlight will use Google signed-in data signals to recover unattributed cross-device and cross-environment web conversions across DV360, SA360 and Campaign Manager 360 reporting, with no action required. Separately, legacy mCPV and Video reach 1.0 line items are paused and archived from 26 October. For you, this means Floodlight conversion counts may rise in November without a tracking change, so annotate reports before comparing periods. DV360 Help · 16 Oct 26
Read the original edition →Netflix adds measurement tools and plans Trade Desk audience access
Announcement · Trade Desk audiences from November Netflix announced a pixel and expanded conversion API support for more than twenty lower-funnel events, alongside conversion lift, attribution and A/B testing. The Trade Desk audience capabilities are scheduled for November, with in-DSP measurement to follow. These give buyers more ways to connect streaming exposure with outcomes, subject to access and implementation. Define the test and measurement plan before committing spend; the feature announcement is separate from Netflix’s advertising-revenue forecasts. Netflix · Official author: Nicolle Pangis (Netflix) · 20 Oct 26
Read the original edition →Google Marketing Platform Drop introduces Gemini-led planning workflows
Announcement · availability not stated Google’s first Marketing Platform Drop outlines a Gemini-powered Home experience for Search Ads 360 and curated media bundles for live sports and major events in the DV360 Marketplace. It also includes LinkedIn CTV access, covered separately in Paid Social. The announcement does not provide a full availability schedule for every capability. Check which parts your account can use, and keep proposed actions separate from approved changes to campaigns or media plans. Google Marketing Platform Blog · Official author: Glenn Thrope (Google) · 22 Oct 26
Read the original edition →Microsoft Monetize adds Copilot deal troubleshooting and retires the AppNexus Prebid adapter in early 2027
Publisher update · action required early 2027 Microsoft Monetize's holiday update adds a Copilot agent that troubleshoots a deal by ID or name and turns reports into charts, passes OpenRTB 2.6 ad-pod signals to DSPs starting with The Trade Desk, and applies Amazon Dynamic Traffic Engine recommendations to direct Amazon DSP capacity to Monetize inventory. The action item: the AppNexus adapter shuts down in early 2027, so publishers must move to the Microsoft Prebid Adapter on Prebid.js 11.36.0, which carries consent passthrough that DSPs such as DV360 require. A publisher partner on the old adapter can lose consented bids, so ask your supply partners when they migrate. Microsoft Advertising · Official author: Nick Bolt (Microsoft) · 21 Oct 26
Read the original edition →Google removes two placement requirements for video inventory
Policy update Google’s policy change log removes the requirement for accompanying video content to load within the page body and for a sticky player to start in the main content before moving as the user scrolls. The change took effect on 22 October. Other video-inventory restrictions remain. Publishers can review eligible implementations, while buyers should continue assessing placement quality and attention separately from whether a layout meets Google’s rules. Google AdSense Help · 22 Oct 26
Read the original edition →Display & Video 360 API v5 enters public beta and drops Customer Match management
Announcement · API public beta The v5 API is "available in public beta starting today", adding advertiser-level YouTube video and channel exclusions, digital out-of-home insertion orders and refactored audience targeting for Demand Gen and YouTube ad groups. Customer Match audience management is removed from the API; Google points to the Data Manager API instead. Breaking changes remain possible during beta, and a v4 sunset date is not given. Any tool that uploads Customer Match lists through the DV360 API needs a Data Manager API path, so ask vendors for their migration plan. Google Ads Developer Blog · Reported by: Trevor Mulchay · 22 Oct 26
Read the original edition →Apps that use "Sign in with ChatGPT" draw on your plan's limits, and overflow billing is off by default
Help page · practical note OpenAI says Go, Plus and Pro subscribers can let participating apps draw on their ChatGPT plan. Each app can have a weekly cap, but that cap does not create an additional pool of usage. Credit overflow is off by default; enabling it can let an app continue using paid credits after the plan allowance runs out. Frederick Vallaeys highlights the practical question: what stops the tool while nobody is watching? Set the app limit and review paid-usage settings before leaving it processing client work. OpenAI Help · Analysis by: Frederick Vallaeys · 22 Oct 26
Read the original edition →Gemini Notebook adds review steps before generating slides and videos
Product update Google’s Gemini Notebook update lets users review outline or producer notes before generating a slide deck or video overview. Users can guide the topics, format and style, give feedback on the plan, then proceed to production; the faster route without review remains available. Google announced a global rollout over the following week. For campaign reports, this is a useful point to check the story, figures and distinction between findings and recommendations before generating the finished presentation. Google · Official authors: Viraj Doshi and Shan Wang · 19 Oct 26
Read the original edition →Claude brings unified chat and Cowork to Enterprise
Enterprise availability Claude Enterprise users can now move from a question to delegated project work in the same conversation. The unified experience extends to web, desktop and mobile, with agentic tasks running in Anthropic’s managed cloud and administrator controls for access and usage. It follows the earlier release for Pro and Max. For reporting teams, the useful change is continuity between discussing a brief and producing its working files, with permissions and review still part of the process. Anthropic · 23 Oct 26
Read the original edition →Anthropic opens a research preview of Claude Duet, a proactive agent for small businesses
Research preview · US only Claude Duet is a separate app on iOS, Android and the web in which Claude works proactively: it brings items that need attention, carries out tasks and keeps working when the user steps away, built on Claude Opus 5.5 and sharing the Claude account's memory, skills and connected apps, including Shopify, HubSpot and Stripe. It asks before sending anything sensitive, making a payment or deleting data. The preview is by waitlist for US Pro and Max subscribers. Another always-on agent that can touch commerce and CRM data, so keep the approval step on for anything that spends. Anthropic · Analysis by: Dan Shipper (Every) · 21 Oct 26
Read the original edition →OpenAI's models and Codex are coming to Google Cloud
Partnership · GA in the coming weeks OpenAI and Google Cloud will bring OpenAI's frontier models and Codex agentic workflows into Google Cloud's Model Garden, purchasable through Google Cloud Marketplace against existing spend commitments, with general availability "planned in the coming weeks"; Microsoft remains OpenAI's primary cloud partner. Google Cloud's Alex Rutter names GPT-6 Astra, GPT-6.1 Sol and GPT-6 Luna alongside Gemini 4 Argon. Ads data already in BigQuery can be processed by OpenAI models under your Google Cloud contract, so compare cost and governance with your current API set-up. OpenAI · 🥇 First contributor: Alex Rutter (Google Cloud) · 21 Oct 26
Read the original edition →Claude Platform becomes available through an Anthropic-managed Google Cloud service
General availability Anthropic’s new Google Cloud offering provides Claude Platform features with Google Cloud authentication, billing and spend commitments. It is generally available with US and Global endpoints. Anthropic manages this route and acts as data processor; the existing Google-managed option remains separate. Teams running analysis or automation on Google Cloud can compare the two routes against their feature, procurement and data-processing requirements. Anthropic · 22 Oct 26
Read the original edition →A brand-search review before judging account performance
Useful reference Ameet Khabra examines whether apparent Google Ads deterioration reflects changes in how brand and non-brand activity contribute to the account. Use the analysis as a prompt to separate demand capture from acquisition when reviewing results, rather than assuming an account-level trend describes every campaign. LinkedIn article · Analysis: Ameet Khabra · 20 Oct 26
Read the original edition →BackgroundR: batch image editing from Google Sheets
Useful reference BackgroundR is an open-source Google Marketing Solutions project for running batches of image-editing prompts from a spreadsheet and saving the results to Drive. It can help teams prepare creative variants, with a Google Cloud setup and associated usage costs. Review product accuracy and brand consistency in the output. The repository explicitly says this is not an officially supported Google product. Google Marketing Solutions · GitHub · Current reference
Read the original edition →Seller Assistant’s plugin connects Amazon selling work with Quick and Claude
Useful reference Amazon’s September announcement describes a Seller Assistant plugin in beta for US stores, connecting seller workflows with Amazon Quick and Claude. This is a useful reference resurfacing this week, not a fresh launch. Check store eligibility, data access and the approval steps before letting an assistant prepare operational changes. About Amazon · 23 Sept 26
Read the original edition →PSA webinar: does Ad Strength predict performance?
Upcoming event Sarah Stemen and Brad Geddes will discuss responsive search ad pinning, Ad Strength and performance evidence on 28 October, 12–1pm EDT (4–5pm London). The free online session promises analysis of tens of thousands of RSAs. This is a forthcoming discussion; its findings have not yet been reviewed. Paid Search Association · 28 Oct 26
Read the original edition →Inside Google Ads Live : Toronto
Upcoming event Jyll Saskin Gales’s one-day event takes place at the CN Tower on 29 October, from 9am to 5pm local time. The programme brings experienced Google Ads practitioners together for tactical sessions and discussion. Check the organiser’s page for tickets and the latest programme. Inside Google Ads Live · Organiser: Jyll Saskin Gales · 29 Oct 26
Read the original edition →SERP Conf. Vienna : conference and masterclasses
Upcoming event SERP Conf. Vienna runs on 3–4 November: a conference day followed by masterclasses. The English-language programme covers search, AI and ecommerce, including paid-search topics. It is a useful option for readers planning their next learning or networking event; confirm the session and ticket details with the organiser. SERP Conf. · 3–4 Nov 26
Read the original edition →The advertiser takeaway
Read historical guidance alongside its original dates and source limits. Check current platform documentation before acting on an older announcement or account observation.
Explore the original edition
Every update from this edition is linked above. Explore the original newsletter for its full context, contributor credits and reporting links.
Read Paid Media News (Week 43) on LinkedIn