Alex Vasilyev, Head of Machine Learning at Nefta chats about AI and ML developments in today’s adtech space and how advertisers and marketers alike need to pay more attention to reduce biases when deploying these technologies for digital campaigns:

____________

Welcome to this MarTech Series chat, Alex, tell us about yourself and your journey in the tech realm through the years?

Thanks for having me! I’ve always been drawn to the intersection of technology, maths, and design.

I started my career in web design and development. After completing my PhD in applied maths I continued as a data scientist and started working for a creative management platform, where I had the opportunity to work on cutting-edge projects like automatic ad generation and content augmentation. It was incredibly rewarding to collaborate with early adopters like Uber and Nike.

Since then, I’ve transitioned to the world of Demand-Side Platforms (DSPs), which allow advertisers to buy advertising space through automation, where I’ve tackled a range of exciting challenges.

Tell us about your observations surrounding the state of adtech today and how advances in AI and ML are enhancing the overall game for end users of adtech?

The state of ad tech feels like a high-stakes game right now. We’re sitting on a treasure trove of data, with the potential to build models that deliver the perfect ad – a win for both advertisers and consumers. But like any good game, there are challenges. Increased regulations and some shady practices in ad mediation make it trickier than ever to achieve that ideal scenario.

Fortunately, advances in AI and machine learning (ML) are emerging as powerful tools to navigate these challenges. Early adopters who leverage deep learning and embedding technologies are demonstrably taking the lead in performance marketing. By intelligently selecting the most relevant ads to serve, AI and ML can significantly enhance the user experience while driving better results for advertisers.

Marketing Technology News: MarTech Interview with Gurdeep Dhillon, Chief Marketing Officer @ Contentstack

What are some of the cautionary tips or challenges today’s adtech users and advertisers should pay attention to as AI and ML create more intuitive campaign cycles and experiences?

AI models are only as good as the data they’re trained on. Biased data can lead to discriminatory ad serving, so every effort needs to be to eliminate bias, and ensure models that are accurate.

Tell us more about Nefta’s core platform and how it is unique?

Nefta’s core platform stands out for its focus on privacy-centric user understanding. We leverage innovative techniques to glean valuable insights from user behaviour, all while adhering to the strictest privacy regulations. This allows us to build a rich user profile without compromising user data.

What in today’s state of digital advertising needs to change drastically in your view?  

Focus on user experience. Right now, too many ads feel intrusive and irrelevant, disrupting the user experience. We need a shift towards contextual advertising that seamlessly integrates with content and provides genuine value to users.

The current reliance on individual platforms’ data silos is a major bottleneck. In my view, a drastic change is needed towards a system where advertisers and publishers can securely exchange contextual data features. For instance publishers sharing user interest labels, without compromising user identity. This enriched data would allow for the creation of far more sophisticated models, leading to hyper-targeted advertising that benefits everyone.  While changing protocols and standards is a challenge, the potential rewards – a more relevant user experience and improved campaign performance – are well worth it.

Can you talk about the future of AI, ML and adtech: what will the market look like down the line?

Here are a few of my predictions for the future of AI, ML and adtech:

• Increased focus on first-party aata – with regulations tightening around data privacy, the importance of first-party data will continue to grow.
• Personalisation and user experience – the future of adtech will prioritise user experience through personalised, non-intrusive advertising.
• Content integration and native advertising – as users increasingly demand seamless integration of ads into content, AI will facilitate the creation of native advertising experiences that blend seamlessly with the surrounding content.
• Advancements in AI techniques – ML techniques such as embeddings and neural networks will continue to advance, enabling more complex and accurate modelling of user behaviour and ad performance.

Marketing Technology News: Bend Rules-Based Marketing with Self-Optimizing Journeys