Share on LinkedIn
Share on XThis post was originally posted here by Csaba Szabo, MD EMEA, IAS
It was great to return to the Barbican with the @IAS team for the London Beyond 2026 event. We brought together almost 200 colleagues, clients, and industry partners to discuss the trends shaping advertising today. The goal was simple: tackle key challenges, share innovation, and spark meaningful conversations about what comes next. A few weeks on, I am still reflecting on the energy in the room and the shared ambition to raise the bar on media quality and performance – not to mention Claudia Winkleman’s hilarious inputs.
It’s hard to distil the depth of conversations into one LinkedIn post, but here’s my top three takeaways:
1. Media quality as a growth driver It’s clear that media quality is no longer a simple safeguard or cost of business. It should be evaluated for what it truly is- a growth driver. Prioritising quality from the start can improve efficiency and unlock stronger returns. High performance does not happen by chance. It is built on the right foundations and quality data inputs. Our clients see that when they used IAS’ technology to focus on media quality, they achieved a 35% increase in econometric profit ROI lift. Compelling numbers indeed.
2. AI must be applied with intention Of course AI came up in every conversation, and why not? The potential is huge and it’s embedded at every step of the marketing life cycle. It’s supporting ideation, helping us to spot and identify early trends. It’s present in production with generative tools helping to tweak creative to appeal to micro audiences in seconds. And it’s executing on complicated transactions in the programmatic ecosystem. But I think we must all agree, AI without structure does not solve problems, it amplifies them. The focus should be on using AI to bring clarity and building systems that teams can genuinely trust. A standout moment for me was the impact of IAS’ “multimodal powerhouse” – our Multimedia Technology. It analyses text, audio, imagery, and video down to a frame-by-frame level of scrutiny, making it 130% more accurate than text-only analysis and three times more effective at identifying high-risk environments when combining signals, protecting brands from a rapidly growing volume of low-quality AI-generated content.
AI can support scale, but it cannot replace human judgment or intuition. You still need experienced teams to interpret signals, challenge outputs, and make the right calls. I personally think that it’s the balance between technology and human thinking that will drive meaningful results. Simply putting a layer of AI on top of chaos is not going to deliver performance; it is only going to amplify the chaos. It was enlightening to hear the real-life applications of superior technology and human expertise driving real-world impact.
3. Foresight over reaction in fragmented environments The ‘fragmentation paradox’ is the realisation that multiple channels and signals mean greater scale and new opportunities to reach audiences, but often lead to less clarity, making it harder to know what really drives performance. That is why foresight matters more than ever. I believe the brands that win will be the ones that can anticipate, not just react. Clear, actionable data is what enables that shift. It was Charlotte Skornik, Head of Client Measurement, UK, Belgium & The Netherlands, TikTok, who said, “Transparency is the foundation of performance.” By leveraging tools like Quality CPM (QCPM), we move beyond reactive fixes and toward data-driven foresight, smarter supply paths, and responsible approaches. This shift ensures media quality actively drives growth, allowing customers to maximise efficiency while scaling reach with confidence.
Ultimately, high performance needs to be engineered. It all comes back to stronger ROI, building true transparency, and putting quality at the centre of every decision. When these work hand-in-hand that is when performance becomes scalable and sustainable. As an ever-evolving industry, we operate in high-pressure environments with very little room for error. The opportunity ahead is clear. If we can combine trusted data, intelligent technology and human expertise, we can turn complexity into clarity and potential into performance.
Once again, thank you to those who went Beyond with us.
