On 15 May 2025, Ayudante APAC hosted afocused, expert-led webinar to help marketers across the region rethink theirdata strategies in response to tightening privacy regulations.
The session spotlighted GA4 Consent Mode,offering both legal context and hands-on technical guidance to help ensure thatdata collection remains robust, compliant, and effective.
I had the pleasure of hosting the event,joined by two industry experts whose perspectives are helping shapeprivacy-first digital analytics across Asia:
Shakky opened the session with acomprehensive overview of the data protection landscape across Southeast Asiaand the wider APAC region. His background in privacy-conscious measurementoffered a much-needed bridge between regulatory demands and practical marketingneeds.
“Even hashed email addresses can be consideredpersonal data. Uploading them to platforms such as Google or Meta may betreated as a third-party provision of personal data. In many countries,explicit user consent is legally required.”
He guided attendees through key legaldefinitions in Singapore, Japan, Thailand, India, and Vietnam, making it clearthat even well-intentioned tools like Enhanced Conversions must be deployedwith great care.
One especially timely point was the growingchallenge of browser-level restrictions:
“Safari limits cookie lifespan to seven days,and just 24 hours for traffic from adverts. In iOS-dominant markets like Japan,that can mean losing up to 12% of conversions from measurement.”
Shakky emphasised that Consent ManagementPlatforms (CMPs) and properly worded privacy policies are no longer optional,as they’re the foundation for lawful and sustainable analytics.
Finally, he concluded with this criticaltakeaway:
“Consent management isn’t about whether you wantto do it — it’s about whether your country requires you to.”
The second half of the webinar featuredTaku, who delivered a practical, detailed walkthrough of how to implementConsent Mode using Google Tag Manager and a third-party CMP such as Cookiebot.
“Consent Mode enables users to decide what datais sent and when. But to make it work, marketers need to link a CMP, configuredefault states, and respect what users opt into—or out of.”
His live demo covered:
Taku also explained how denied consentdirectly affects GA4 data modelling, stressing that unless minimum thresholdsare met, non-consented data won't be included in reporting.
“You need 1,000 denied events per day and 1,000users with granted consent. If you don’t meet those thresholds, GA4’s modellingsimply won’t activate and you’ll be missing data without even realising it.”
Finally, he concluded with this reminder:
"ConsentMode isn’t just a legal checkbox.
It directly affects your ability to measure, optimise, and retarget. It’s both arisk factor and a strategic asset."
The Q&A session reflected the stronginterest and maturity of our audience. Some of the most discussed pointsincluded:
Q:Are GA cookie IDs considered personal data?
A (byShakky): Yes, especially in countries like Japan. They should be treated aspersonally identifiable information (PII) under most interpretations.
Q:What’s the typical uplift from Enhanced Conversions?
A (byShakky): It varies by use case, but most advertisers observe match rates in the30–40% range. Legal compliance remains crucial.
Q:Should you capture membership IDs if consent is denied?
A (by Taku): Ideally not, especially ifthose IDs can be tied back to user profiles. Masking or redacting suchparameters is essential.
Q:What’s the difference between device-based and blended identity in GA4?
A (by Taku): Blended mode allows GA4 to usemodelled data. The gap can be significant, even up to 2× the user count in somecases.
This webinar underscored the growingimportance of user consent in today’s data strategies. As AI takes on a largerrole in analysis and decision-making, our attention must shift towardshigh-quality data, precise measurement, and well-structured data governance.
A key takeaway was the increasingsignificance of first-party data. Organisations must take ownership of theirdata infrastructure, implement compliant and user-friendly consent flows, andestablish strong foundations for privacy and trust.
Amongall available sources, user, behaviour and conversion data from owned mediastand out as the most critical.
AI will undoubtedly support analysis,particularly through modelled data. However, the quality of those insightsdepends entirely on the data we feed into the system. In the AI era, ownedmedia data (such as behavioural signals, user identifiers, and conversionevents), is not just valuable, but increasingly essential.
At the same time, measurement remains a human responsibility. Only people can decidewhat should be measured, where the value lies, and how those signals contributeto business outcomes. Measurement is not simply a technical process, but also astrategic decision.
Everyone wants high-quality results fromAI, but without sound measurement, that’s impossible. Garbage in, garbage out.
At Ayudante APAC, we are committed tosupporting teams across the region as they strengthen their measurementstrategies. Whether you are just beginning with Consent Mode or refining yourGA4 setup, we are here to help.
Let’s continue moving towards smarter, moreaccurate, and more respectful data practices — together.
