Manual bid vs max delivery
INSIGHTS FROM EXPERTS ON LINKEDIN
Jyll Saskin Gales spotted ChatGPT appearing in Google Ads Auction Insights for non-brand search terms, suggesting OpenAI may be running Search ads on category keywords. After switching the campaign to brand and competitor keywords, ChatGPT disappeared from the report. She highlights that Auction Insights can uncover valuable competitive activity beyond the usual PPC competitors.
David Pourquery believes large language models like Claude are useful for analyzing Google Ads accounts, but not for making changes automatically. He warns that giving AI full control can break tracking, reset learning phases, or even put accounts at risk. AI works best as an assistant that supports decision-making, not as the one making the decisions.
Liam Collins says many marketers know their measurement isn’t perfect, but still avoid incrementality testing because they’re afraid of what the results might reveal. He shares practical tips for running reliable holdout tests, including choosing the right markets, measuring real business outcomes, and agreeing on success metrics before the test begins. Testing assumptions is the only way to know whether a channel is truly driving growth.
Liam Moroney argues that B2B SaaS companies have gradually changed the meaning of established marketing terms like GTM, ICP, and demand, often to fit sales-led or vendor-driven narratives. He believes this creates confusion, weakens marketing strategy, and makes it harder to learn from the wider marketing industry. Returning to shared marketing principles would help teams make better long-term decisions.
WHAT'S NEW IN THE INDUSTRY
LinkedIn has introduced a new option that lets users flag posts that appear to be low-quality AI-generated content, and those posts may receive less distribution across the platform. The company says it’s targeting content that lacks original perspective rather than AI itself, encouraging creators to use AI to improve their writing instead of replacing their own ideas. Authentic, original content is more likely to stand out and reach new audiences.
Software companies are rapidly changing how they price AI products, moving away from traditional per-seat subscriptions toward models based on usage, outcomes, credits, or completed tasks. While these new pricing structures better reflect the value AI delivers, they also make costs harder to predict and turn pricing into a key part of a company’s positioning. As AI continues to evolve, businesses are still experimenting to find pricing models that customers trust and understand.
Google is rolling out a new feature that lets eligible advertisers apply for certain policy certifications, including Social Casino Games, directly inside Google Ads instead of through the Help Center. The update makes it easier to manage certifications alongside campaign settings and could simplify compliance for affected advertisers. Existing certifications and pending applications won’t be affected, and the feature will become available to more accounts over time.
Microsoft has expanded Ad Preview Hub to Performance Max campaigns, making it easier for advertisers to see how their ads will look across placements before they go live. Teams can also generate secure shareable links, allowing clients and stakeholders to review creatives without needing access to the Microsoft Advertising account. The update streamlines approvals and helps advertisers catch creative issues before launch.
Google appears to be redesigning its bidding strategy menu by listing Target CPA and Target ROAS as standalone options during campaign setup. While the bidding strategies themselves haven’t changed, the new layout makes it easier for advertisers to understand and select the right approach. The update may also prepare advertisers for Google’s upcoming Smart Bidding changes later this month.
Google has released Structured Data Files v10.1 for Display & Video 360, adding support for AI transparency labels, digital out-of-home (DOOH) campaigns, and new bulk management features. The update also introduces better creative reporting and ends support for older SDF versions. Teams managing campaigns at scale will need to update their workflows to take advantage of the new features.
Google Analytics has introduced a new diagnostic tool that alerts advertisers when important campaign URL parameters are missing. These missing identifiers can reduce attribution accuracy, making it harder to measure campaign performance. The update helps marketers catch tracking issues early and improve the reliability of their reporting.
New data from Funnel shows that paid search budgets continue to grow across the US, EMEA, and APAC, despite concerns that AI search tools could reduce demand. In the US alone, monthly Search ad spend increased by 122% to $100 million, while Black Friday 2025 spending also outpaced the previous year. The findings suggest advertisers are still investing heavily in Search, even as social media advertising continues to gain budget share in some regions.
Microsoft is retiring its Predictive Matching feature and folding its capabilities into AI Max for Search campaigns. The change is part of Microsoft’s broader push toward AI-powered query matching, asset personalization, and intent-based targeting. No action is required from advertisers yet, with the transition expected to roll out over time.
Reddit reported another strong quarter, with advertising revenue up 64% year over year to $762 million and advertiser numbers increasing by more than 70%. However, the company’s stock fell after management highlighted slower US user growth and more volatile traffic from search engines. Even so, Reddit says demand from advertisers and adoption of its AI-powered ad products continue to accelerate.
That’s the scoop for this week! If you found this valuable and any useful insights caught your eye, feel free to share them with your network.
Until next week!
