GoogleGoogle

How we’re using AI in new ways to fight invalid traffic

By Per Bjorke·2025年8月12日·2 分钟阅读

摘要

无效流量(IVT)指非真实用户产生的广告活动,不仅浪费广告预算、损害媒体收入,还破坏行业信任。为解决这一问题,Google广告流量质量团队联合Google Research和DeepMind,近期推出了基于大规模语言模型的新一代防御体系,能够更精准地识别产生异常行为的广告位。

新应用通过分析应用和网页内容、广告位及用户交互行为,提供了更快速、更强大的保护。具体而言,内容审核能力大幅提升,使得由欺骗性或破坏性广告投放导致的IVT减少了40%。这帮助广告主更准确地触达目标受众,同时有效遏制政策违规者。

此外,Google持续运行全自动和人工审核流程,确保即使在广告展示后,广告主也不会被收取IVT产生的费用。这些流量质量技术创新代表了Google过去二十年应对不断演变的威胁、维护数字广告生态系统完整性的最新努力。

猜你喜欢

AppsFlyerAppsFlyer

Getting started with user acquisition for apps: The complete guide

User acquisition (UA) remains critical in the maturing app market, with non-organic installs growing annually. Key challenges include rising media costs, churn, fraud, and fragmentation. Attribution data and multi-touch modeling help optimize UA by identifying high-performing channels and audiences. Strategic budgeting, A/B testing, and app store optimization (ASO) are essential for maximizing ROI. Ad ops decision-makers should prioritize fraud protection, explore diverse media channels (paid, owned, earned), and leverage cohort analysis to drive cost-effective growth.

Shani Rosenfelder·2025年11月5日·16 分钟阅读阅读文章 →
AppsFlyerAppsFlyer

Winning mobile banking users – your ultimate guide to trust and growth in the age of privacy

Banking apps are vital digital channels requiring granular measurement to optimize user acquisition, engagement, and retention amid strict privacy regulations. Key challenges include measuring sensitive conversions, preventing fraud, and personalizing experiences without compromising compliance. Granular event tracking, deep linking, and anti-fraud solutions are essential. Banks must measure early-funnel milestones, re-activate dormant users, and leverage owned media for cost-effective re-engagement. Advanced attribution methods like SKAdNetwork, probabilistic modeling, and data clean rooms help navigate privacy changes. Effective measurement drives long-term customer value and validates mobile's impact on business outcomes.

Lesia Kupriienko·2025年10月13日·22 分钟阅读阅读文章 →
MolocoMoloco

Inside Moloco's CARA: the AI that compounds

The open internet presents unique challenges for performance advertising: fragmented identity, closed first-price auctions, and non-stationary supply. Moloco's CARA compound architecture tackles this with six integrated technical domains—Campaign Automation, Supply, Ad Recommendations, Bidding, Creative, and Signals—running on a unified ML infrastructure. Key insights for ad ops: the system continuously learns from every interaction, uses knowledge distillation to serve real-time predictions under 10ms latency, and validates improvements through rigorous live experiments. In 2025, 65 validated model updates reduced CPA by 17% and improved ROAS by 27%. The key takeaway: compound AI architectures that connect prediction, bidding, creative, and data can unlock measurable performance gains beyond walled gardens.

Tal Shaked·2026年8月7日·14 分钟阅读阅读文章 →
AppsFlyerAppsFlyer

Solving retail media measurement – your ultimate guide to earning advertiser trust

Retail media networks (RMNs) must prioritize accurate measurement to build advertiser trust and prove ROI. With 68% of advertisers ranking ROI as top priority, RMNs need user-level data, SKU-level attribution, and lift analysis to demonstrate campaign impact. The article outlines a checklist for effective measurement, including omnichannel coverage, deduplication, and easy-to-access reports. It emphasizes the importance of data collaboration platforms for bridging walled gardens and achieving precision. A case study of Wolt Ads shows a 32% revenue uplift using AppsFlyer's data collaboration platform. Key takeaways: measurement drives ad revenue, user-level data is essential, flexibility matters, and simplifying reporting is critical for brand adoption.

Zara Sultanbekova·2025年11月5日·10 分钟阅读阅读文章 →
MolocoMoloco

Beyond Last-Click: How We Prove Incremental ROAS in Retail Media

Retail media success now hinges on measuring true incremental sales, not just ROAS. 71% of advertisers prioritize incrementality, yet many rely on legacy attribution. Traditional testing methods are manual, revenue-limiting, and incompatible with AI-driven optimization. AI-native ghost bidding—using randomized controlled trials during live campaigns—cleanly separates ad-driven impact from organic behavior. Early results show iROAS from 253% to 1,609%, with up to 29% of conversions being truly incremental. Best practices include statistical rigor, clean population definitions, comprehensive conversion tracking, and regular testing. Future-proof incrementality solutions will be privacy-compliant and integrated into ad serving.

Akhil Gupta·2025年9月9日·5 分钟阅读阅读文章 →
AppsFlyerAppsFlyer

Cross-channel marketing analytics: how it works in 2026

Cross-channel marketing analytics isn't about putting Meta, Google, and TikTok numbers side by side—they often double-count the same customer journey. Fragmented identity is the real culprit; without a first-party Customer User ID, attribution measures platform credit, not customer value. The article explains that deduplicating conversions across mobile, web, and CTV can lift attributed revenue by 30–60% and improve ROAS by 20%. It walks through attribution models, warns against platform-native analytics, and advises using an independent MMP for true cross-channel measurement. Ad ops takeaway: fix identity resolution first, because AI-driven optimization and budget allocation depend on trustworthy, deduplicated data.

Rachel Siegman·2026年9月3日·14 分钟阅读阅读文章 →
MintegralMintegral

Mobile App Monetization Strategies for New Apps

New app developers must integrate monetization from day one, not after building a user base. Rewarded ads offer a value-exchange model that boosts retention. A hybrid of IAA and IAP creates sustainable growth, but requires careful design to balance user experience. Early revenue, even modest, should be reinvested into user acquisition. Continuous testing of ad formats and placements is essential. Partnerships with mediation platforms like Mintegral can maximize ad revenue without harming UX.

Mingyue Zhu·2025年10月17日·6 分钟阅读阅读文章 →
MolocoMoloco

6 Myths About AI Personalization Limiting Your Retail Media Growth

Retail media ad spend has reached $140B globally, but performance hinges on personalized customer experience, where a gap exists between retailer perception (92% believe they deliver personalization) and shopper reality (48% agree). AI-powered personalization can bridge this gap, debunking myths about data readiness, resource intensity, privacy risks, and growth ceilings. Modern AI handles imperfect data, automates campaign management, requires less PII, and enables real-time inference for higher engagement and revenue. Retailers can achieve 3-5x growth even in mature networks.

Jon Flugstad·2025年9月4日·6 分钟阅读阅读文章 →

更多来自 Google

Build campaigns that drive high-converting, sales-ready leads.

销售周期持续拉长、线索转化路径难以追踪,使得 lead gen 投放的 ROI 衡量正变得空前困难——除非用户主动填写表单,否则很难判断到底是什么在驱动转化。在纽约 Rethink ROI 现场录制的 Ads Decoded 播客中,Goo...

2026年9月18日·1 分钟阅读

Rethink 2026

文章是对 Rethink 2026 活动的预告,指出即将到来的假日购物季将由 AI 全面驱动,营销人员需要新的产品能力与实战方法才能抓住增长窗口。主办方将集中释出一批产品更新与可落地的客户触达打法,帮助广告主在旺季提升投放效率与 ROI。对...

2026年9月16日·1 分钟阅读

Rethink your strategy to drive sales this holiday season.

Google 在纽约举办 Rethink Retail 活动后,其 Ads Product Liaison Ginny Marvin 与负责零售广告体验的 Group Product Manager Enric Durany 在 Ads D...

2026年9月16日·1 分钟阅读

Drive profitable growth with new data and measurement tools

Google 正将效果衡量从被动报告升级为主动增长引擎,围绕数据基础、多重信号与因果证明三要素推出新能力。Data Manager 与 GA、DV360 打通,并推出 universal Data Manager API 与 enhance...

2026年9月10日·2 分钟阅读

Build a measurement stack you can rely on to steer your campaigns.

本文探讨了如何构建一个可靠的测量体系,以应对归因分析、增量测试和MMM(媒体组合模型)等工具之间的冲突。Google Ads产品管理总监强调,广告主应结合多种测量方法,并重视长期价值而非短期ROAS,同时介绍了支持更长购买路径的Qualif...

2026年9月2日·1 分钟阅读

Reach your audience in new ways with August’s Demand Gen Drop.

Google在2025年下半年对Demand Gen广告进行了数百项改进,平均为广告主带来30%的转化或转化价值提升。最新更新包括在YouTube上测试从广告直接跳转至消息应用以促进高意向用户转化,并为旅游行业推出本地活动、实时优惠及个性化...

2026年8月27日·1 分钟阅读