AI is changing how marketers work, from automating campaign production to analysing customer behaviour and making media investment decisions. But as AI tools become more accessible, a bigger question emerges: how can businesses turn experimentation into meaningful, scalable results?
Two sessions explored this challenge from different perspectives. The first examined how marketers can become builders, developing AI-powered solutions to solve real-world problems. The second explored how organisations can scale AI beyond individual projects by connecting technology to business objectives, customer needs, and reliable data.
Here are the key takeaways from both sessions, along with how these ideas can be applied to Google Analytics 4 (GA4).
Traditionally, marketers have depended on technical teams to turn ideas into working products and tools. Generative AI is changing that dynamic, making it easier for marketers to experiment, develop prototypes, and solve problems without needing extensive coding expertise.
The first presentation by Ravi Shankar Mallavarapu CMO, AirAsia MOVE demonstrated how this shift is already happening in practice. Marketing teams are building AI-powered solutions to automate repetitive tasks, accelerate campaign execution, and improve decision-making.
Examples included reducing certain media-related workflows from four days to four hours, generating hundreds of thousands of landing pages within weeks, and automating the adaptation of campaign visuals across different formats, languages, and currencies.
These are not simply productivity improvements. They demonstrate how marketers can move from identifying operational bottlenecks to building solutions that address them directly.
One particularly interesting development is the use of specialised AI agents that work together to support marketing decisions.
Rather than relying on a single AI tool, teams can coordinate multiple agents with distinct responsibilities, such as analysing market conditions, evaluating media performance, reviewing campaign recommendations, and checking outputs before deployment.
This approach can help marketers process information at a scale that would be difficult to manage manually. For example, AI agents can continuously assess market and pricing changes, identify opportunities across destinations, and recommend where media budgets should be allocated.
Creative production is another area where this approach is making a difference. Instead of manually adapting a campaign for every market, marketers can provide a core visual and generate multiple variations, freeing creative teams to focus on developing ideas rather than repetitive execution.
However, automation does not mean removing human oversight entirely. Teams still need to validate recommendations, manage risks, and determine which activities require approval before going live. Smaller experiments can provide a controlled way to test solutions before expanding their reach.
The presentation also challenged the assumption that building technology belongs exclusively to technical teams.
Marketers often have a close understanding of customer pain points, campaign performance, and the practical challenges faced by their teams. This puts them in a strong position to identify problems worth solving and develop solutions grounded in real needs.
The recommendation was to start small: identify a specific problem, build a simple solution, test it, and improve it through feedback. Collaboration with technical teams remains important, particularly when moving prototypes into production.
The takeaway: AI is giving marketers the opportunity to move beyond using tools to actively building solutions—and those who combine marketing expertise with a willingness to experiment can unlock new ways of working.
While building AI solutions is becoming easier, scaling them across an organisation requires a different mindset.
The second session is a panel discussion that includes panelists Khoo Xing Jun, Head of Digital, Media and Insights at Dutch Lady Milk Industries (DLMI); Hemanth Jayaraman, Former Head of Loyalty at Enrich (MAG); Ila Mathi, Head of APAC Digital Business and Brand Experience (B2B2C) at Kimberly-Clark; and Mihira Kulatunga, Head of Marketing (MYSG) at Unilever Food Solutions, who highlighted a common risk: businesses can feel pressured to adopt AI because competitors are doing so or because leadership expects visible progress. But investing in technology without a clear business objective can lead to disconnected experiments that fail to deliver meaningful results.
A more effective approach is to begin with the problem. What needs to improve? Is the priority reducing costs, increasing operational efficiency, improving customer relevance, or identifying new growth opportunities?
The technology should follow from that decision, not the other way around.
For marketing teams, this means looking beyond the question of how AI can generate more content. The more important question may be whether AI can help reduce campaign delays, identify ineffective spending, improve customer service, or deliver more relevant offers.
One example from the discussion illustrated how a business can have extensive customer data yet struggle to use it effectively.
In the travel industry, a customer might receive a promotional offer for a destination they have never visited or shown interest in, despite the business having years of booking and transaction history.
That information could reveal much more useful patterns: how frequently someone travels, which destinations they prefer, whether they typically travel alone or with others, and when they tend to book.
For instance, a customer who regularly visits London around the same time each year may be more receptive to a timely, relevant offer than a generic promotion sent to the entire customer base.
The opportunity is to move beyond broad personalisation towards more individualised experiences, using available data to understand each customer's needs and behaviour more accurately.
The same principle applies to operational challenges. The discussion also described how analysing approximately 17,000 unresolved customer cases helped identify recurring issues responsible for a substantial proportion of the workload. By focusing on the underlying problems rather than treating every case separately, the team could prioritise improvements more effectively.
In both situations, AI is valuable because it helps businesses make better use of information they already possess—not simply because it introduces a new technology.
Having large volumes of data does not automatically produce useful insights. Data must be sufficiently reliable, organised, and relevant to the business question being addressed.
The panel emphasised the importance of data engineering, appropriate context, and collaboration across business and technical teams. Feeding information into an AI system without explaining the problem or providing the right context can lead to misleading or irrelevant outputs.
Privacy and governance also matter. Businesses must understand what data they can legitimately use, how it should be handled, and what safeguards are needed as AI becomes more integrated into business processes.
To scale AI successfully, organisations should focus on a few fundamentals:
The takeaway: scaling AI is not about deploying more tools. It is about connecting the right technology, data, and people to solve problems that matter to the business.
GA4 can help marketers understand digital behaviour through events such as product views, internal searches, form submissions, and purchases. These signals can reveal where users encounter friction, which interactions precede conversion, and how behaviour differs across traffic sources and campaigns.
Connecting these insights with CRM records, transaction data, and other relevant business information can provide a more complete view of performance. However, GA4 alone does not capture every customer interaction or explain every motivation behind a behaviour.
The goal is to use measurement as part of a broader decision-making process: identify the problem, interpret the available signals, and connect the findings to meaningful business actions.
As marketers move from experimenting with AI tools to deploying agents that can analyse information and take action, the reliability of their underlying data becomes increasingly important.
Jasper Poon, Global Marketing Lead at Ayudante APAC pointed out that AI agents can only make useful decisions when they have access to trustworthy signals and sufficient context. Gaps in measurement, inconsistent data, or unclear attribution can undermine their recommendations before a campaign even begins.
This is where data signals and GA4 become relevant. Understanding user interactions is one part of the equation; ensuring those interactions are measured accurately and interpreted in the right context is equally important.
What this means for GA4: Reliable event tracking, appropriate consent management, and regular measurement checks help provide a stronger foundation for reporting and AI-supported decisions.
But what does an AI-ready data foundation actually look like, and what should businesses address before deploying AI agents at scale?
Read the full article, “Is Your Data Foundation Ready for AI Agents?”, from Ayudante APAC to explore the key considerations in greater detail.
Click HERE to read the full article.
Across both sessions, one theme stood out: the value of AI depends on how effectively businesses connect technology to real marketing needs.
For marketers, this means developing the confidence to build and experiment, while keeping customer needs and business outcomes at the centre of every initiative. For organisations, it means establishing clear objectives, improving the use of data, and creating the conditions for successful implementation at scale.
For GA4 users, the opportunity is to go beyond reporting surface-level metrics and use behavioural signals to understand customer journeys, identify friction, and evaluate meaningful outcomes. Combining these insights with other relevant data sources can further strengthen decision-making, provided the information is reliable and used appropriately.
As AI continues to reshape marketing, success will not simply come from adopting the newest tools. It will come from asking better questions, building practical solutions, and ensuring that the data and processes behind them are ready to support the next stage of marketing transformation.
If you're ready to tackle these challenges—from understanding customer intent and connecting fragmented data signals to improving measurement accuracy and preparing your data foundation for AI—our team of data measurement experts can help. We work with businesses to strengthen their measurement strategies, make better use of GA4 and other data sources, and build a more reliable foundation for data-driven marketing. Get in touch with our team to explore how we can help you turn better data into clearer insights, smarter decisions, and more effective marketing outcomes.
%201.jpg)
