Only 27% of consumers feel understood by brands. Here's how Southeast Asian marketers can close the first-party data trust gap before it becomes irreversible.
Only 27% of consumers say brands actually understand their interests. That figure, from Site Impact’s 2026 research published via CustomerThink, should bother every marketer who has spent the last three years building a first-party data programme on the promise that it would make targeting smarter and relationships stronger.
It hasn’t. Not yet. And in Southeast Asia — where brand switching costs are low, messaging volumes are high, and consumers receive marketing across LINE, WhatsApp, Shopee, and TikTok simultaneously — the cost of getting personalisation wrong is compounding faster than most teams realise.
The Data You Have Is Older Than You Think
The Site Impact research pinpoints two specific failure modes: outdated data and repetitive targeting. These aren’t technology problems. They’re programme design problems.
Most first-party data programmes were architected to collect — loyalty sign-ups, preference centres, post-purchase surveys — without equal investment in how that data ages. A Shopee purchase signal from eight months ago tells you almost nothing about intent today. A preference recorded at onboarding reflects who someone thought they were, not who they’ve become as a customer.
The fix isn’t more collection points. It’s building decay logic into your data model — treating signals as perishable and flagging profiles when behavioural data hasn’t been refreshed within a defined window. For most mid-market brands in Southeast Asia, 90 days is a reasonable staleness threshold for purchase intent signals. Anything older should be downweighted in segmentation, not discarded, but treated with appropriate uncertainty.
Repetitive Targeting Is a Consent Problem in Disguise
When consumers describe feeling followed by the same ad after they’ve already bought the product, or receiving a promotion for a category they’ve never engaged with, they’re not just describing a bad experience. They’re describing a broken consent relationship — one where their implicit signal (a purchase, a click, a page view) was misread or over-extended.
This matters more in markets like Thailand, Indonesia, and the Philippines, where data privacy frameworks are tightening. Thailand’s PDPA has been enforced since 2022; Indonesia’s Personal Data Protection Law came into full effect in 2024. Consumers in these markets are increasingly aware that their data is being used — and the trust erosion that repetitive targeting creates maps directly onto regulatory risk.
The practical fix is to build suppression logic that’s as sophisticated as your targeting logic. If a customer converts, suppress them from acquisition flows immediately — not on the next campaign cycle. If a customer has received three emails in a category without opening any of them, that’s a preference signal your system should act on, even if they haven’t formally opted down.
Data Quality Is Now an Infrastructure Question
One reason personalisation programmes stall is that data quality is treated as a data team problem, not a marketing infrastructure problem. Monte Carlo’s recent integration updates — connecting their data observability platform across Databricks, Salesforce, Microsoft Teams, and AI agent debugging environments — illustrate where enterprise thinking is heading: toward continuous monitoring of data health across the entire stack, not just at ingestion.
For Southeast Asian brands running heterogeneous stacks — common when you’re operating across markets with different platform norms — this matters practically. A segment built in one system and activated through another can degrade silently. Audience definitions drift. Field mappings break after platform updates. Without observability at the pipeline level, your marketing team is personalising against data it can’t verify.
The minimum viable version of this for a mid-market brand isn’t a full data observability platform. It’s a defined SLA between your data and marketing teams: a shared dashboard that flags when key audience segments haven’t refreshed on schedule, when record counts fall outside expected ranges, and when consent flags haven’t propagated correctly from your CMP to your activation layer. Simple, but most teams don’t have it.
What Actually Rebuilds Consumer Trust
The Site Impact research confirms what most consent-forward practitioners already suspect: consumers don’t object to personalisation. They object to personalisation that feels presumptuous, repetitive, or out of step with where they are in their relationship with a brand. The distinction is important because it reframes the problem.
This isn’t about collecting less data. It’s about using data in ways that demonstrably serve the person whose data it is. A loyalty programme that remembers your preferred store location and surfaces relevant promotions before you have to search for them — that’s personalisation that earns trust. A retargeting sequence that follows you across platforms for a product you bought three weeks ago — that’s personalisation that burns it.
Three implementation principles that hold up across Southeast Asian markets: First, build value exchange into every data collection moment — the consumer should leave with something immediately useful, not just a vague promise of better experiences later. Second, make preference management genuinely accessible, not buried in account settings — LINE and Grab both offer notification preference controls at the channel level; your own CRM should match that standard. Third, audit your activation logic quarterly against your consent records — not annually, quarterly.
Key Takeaways
- Treat behavioural data as perishable: build decay thresholds into your segmentation model and downweight signals older than 90 days for intent-based targeting.
- Build suppression logic that matches the sophistication of your targeting — conversion events, category disengagement, and frequency caps should all trigger automatic exclusions.
- Close the gap between your consent management platform and your activation layer with a shared data health SLA, so your marketing team always knows what it’s personalising against.
The harder question for 2027 is whether first-party data programmes can outpace the trust they’ve already spent. Consumers in Southeast Asia are not naive about how their data is used — and the brands that treat consent as a competitive asset rather than a compliance checkbox will earn loyalty that’s genuinely difficult to replicate. The ones that don’t will keep wondering why their personalisation investment isn’t showing up in retention numbers.
At grzzly, we help brands across Southeast Asia build first-party data programmes that are designed to earn trust from day one — not patch it back together after the damage is done. From consent architecture to activation quality, we work at the intersection of strategy and implementation. Let’s talk
Sources
Written by
Lavender GrizzlyTurning privacy constraints into competitive advantage. Builds first-party data programmes that are compliant by design, valuable by intent, and trusted by the people whose data they hold.