What effect will AI have on the (cookieless) future of media strategy?

The end of third-party cookies is a big challenge. But it’s an even bigger opportunity for the media industry to reinvent itself, says Vincent Spruyt (global chief AI officer at Kinesso). Here he explores how cookie deprecation signal loss could spark a new era of innovation.

For around 30 million Chrome users, the much-hyped cookieless future has become an actual cookieless present. On January 4th Google disabled third-party cookies for one percent of its users as it started live testing of its Privacy Sandbox solution.

The ramp up to 100% should start in Q3 of this year. A seismic disruption to the media and marketing industries? Maybe. But it could also be an outstanding opportunity for reinvention – to shake off three decades of cookie dependency and embrace the possibilities presented by artificial intelligence (AI).

A perfect (positive) storm

A perfect positive storm is brewing in media and advertising, a storm that will be a catalyst for innovation. The convergence of three powerful trends will necessitate and facilitate a period of significant change.

First, for years organizations have focused their efforts on digital transformation. According to Gartner, for example, 87% of business leaders are prioritizing company digitalization. 

Second, the explosion in capability and availability of generative AI solutions continues to redefine what businesses can achieve. 

The third element of this perfect storm is cookie deprecation itself. This provides the industrial-strength impetus for rethinking how we can serve our customers most effectively. Customers will still expect personalized experiences when the modeling and attribution data from cookies is switched off. So a solution is needed.

But it’s extremely limiting to define this next phase as a quest for cookie replacements. Activities like developing first-party data and identity solutions or using UIDS will remain valuable and vital. But no combination of software or strategies will produce a like-for-like replacement. Google has outright stated that its Privacy Sandbox APIs “are not intended to be direct, one-to-one replacements for all third-party cookie based use cases”. 

Clinging to the past has never been the path to innovation and differentiation, so how can we adapt to our new reality? 

A new approach

Though effective in its time, a traditional media strategy followed a linear path: target audience to media plan to creative design to campaign activation and measurement. Each step punctuated with stakeholder approvals. 

With third-party cookie data available, traditional processes work well: user-level tracking accuracy delivers precise audience definitions, targeting, attribution measurement and reporting. But remove cookie data and you remove some of that precision, which means the long, waterfall-like planning and activation approach is no longer efficient. 

A new approach is needed, so let’s take a look at an AI-informed future:

• AI-driven algorithms enable marketers to work in short, fast cycles of budget reallocation, planning and data analysis.
• Audience segments are assigned and reassigned by AI according to real-time campaign performance.
• Creative variations based upon shifting audiences could be generated in real time using AI.
• Using AI, multiple hypotheses, strategies or creative treatments for campaigns could be simultaneously tested and optimized across all channels in real time.

AI and automation could enable us to do a better job for our clients: faster, cheaper, much more precisely. But what are the key steps to getting there?

Pillars of the new paradigm

While there’s no cookie-cutter approach for embracing the new paradigm, it’s clear that successful transformation will rely upon innovation in two major areas: technology and change management. 

Technology & automation

How we deploy technology and our relationship with it is the first pillar of an AI-forward model. Within it there are four key success prerequisites:

• Integrated media framework / platform:
  This is the key that unlocks it all. Working in fast, short cycles requires an end-to-end, AI-driven system for campaign creation and optimization. In this integrated framework, the workflow begins with sophisticated AI algorithms that analyze audience data, shaping the planning phase with insights to build effective media strategies. Product data flows seamlessly into the platform where AI applies it – along with audience insights, product categories, and creative elements – to generate optimized campaigns.
  
  This would enable iterative testing through built-in experimentation capabilities. With the entire workflow connected through an integrated AI system, marketers could go from product feeds to live campaigns much faster, with less manual work required. 
• AI-based attribution modeling:
  Breaking away from cookie-centric techniques like multi-touch attribution, AI-based attribution modeling is more precise. It uses machine learning algorithms to analyze customer journey data, marketing channel interactions and conversions, and then assigns outcome-based attribution credit to each interaction.
  
• Intelligence systems with world knowledge:
  Large language model AIs (like ChatGPT) can ingest vast amounts of information from myriad sources and apply this ‘world knowledge’ to ground any analysis in a human-like intuition. These inferences and common sense reasoning enable AIs to aid in planning, optimizing and activating campaigns, even in the absence of specific observable measurements or historical data from clients or campaigns after cookie signal loss.
• Unified data layer / global data warehouse:
  To feed all the modules in the media cycle – from audience to planning, to activation, to reporting and measurement – there needs to be a single source of truth for all customers and markets.

Change management 

Technical capabilities are only half the story. Managing the change and adopting new approaches is also critical for a successful business evolution. Three areas in particular are key:

• Global adoption of the integrated media framework:
  This is key. An integrated end-to-end solution requires integrated teams working together in the platform, rather than separate teams leveraging a multiverse of individual martech tools. This is an opportunity to focus on integration and consolidation, to adopt a common interface and common processes in order to work most effectively for clients and get the most value from the AI-driven approach.
  
• Finance:
  Effective AI-driven campaign automation hinges on the ability to dynamically reallocate budgets based on performance data. This requires a mindset shift towards budget fluidity in which clients set strategic parameters and trust AI systems to aid in making minor, real-time budgetary adjustments within those boundaries.
  
• Operations:
  Marketing teams tend to work in silos (such as paid search, paid social or programmatic) that need to be replaced with cross-channel teams. In the absence of cookie data it’s going to be crucial to share insights equally across all channels, particularly from authentication-based data like walled gardens or retail media networks.
  
  It’s also going to be vital for teams to embrace automation. An integrated pipeline requires absolute operational precision to work. There’s no room for common manual mistakes (such as taxonomy issues across product meta data, media plans or DSPs). 

An opportunity in disguise

The media industries are at an inflection point. Cookie deprecation means the basis of many traditional operational models will evaporate. Fortunately this major disruption has coincided with advanced AI technologies that offer a new way forward, not of finding new ways to fill the silence left by signal loss, but a whole new way of thinking and working.

Embracing an integrated, AI-informed approach to media strategy might feel daunting. It will undoubtedly require a lot of change and hard work, but it also presents an extraordinary opportunity to develop fresh, innovative ways of connecting with audiences.