Feature Discovery UX: Stop Hoping Users Will Read

Your onboarding copy isn’t being read. Not most of it, anyway — and designing as if it will be is one of the more quietly expensive assumptions in digital product work.

The Non-Reader Is Your Default User

Rita Kind-Envy’s piece on UX Collective puts a sharp point on something product and UX teams in Southeast Asia know but rarely act on: feature discovery copy is written for an audience that largely doesn’t exist. Users — especially on mobile, especially in high-context markets like Thailand, Indonesia, and the Philippines where thumb-scroll behaviour is near-reflexive — navigate by pattern recognition, not instruction.

The implication isn’t that copy doesn’t matter. It’s that copy alone can’t carry the discovery job. When Shopee or Grab roll out a new feature, the ones that stick are almost always the ones surfaced through contextual nudges, visual anomalies in familiar flows, or progressive disclosure — not through a modal wall of text that appears at login. The design has to do the heavy lifting before a single word is read.

For teams building on platforms like LINE or embedding experiences in super-apps, this is doubly true: you’re a guest in someone else’s UI conventions, and your onboarding window is measured in seconds.

Contextual Triggers Beat Announcement Modals

The tactical shift here is moving from broadcast-style announcements to moment-of-need discovery. Instead of a full-screen takeover at session start — which users have been conditioned to dismiss since approximately 2014 — effective feature discovery plants cues at the precise moment a new capability is relevant to what the user is already trying to do.

A concrete example: if your e-commerce app introduces a price-drop alert feature, the most effective discovery touchpoint isn’t a splash screen. It’s a subtle, animated indicator that appears on a product the user just bookmarked. The context creates comprehension faster than any headline can.

From a data activation standpoint, this is also a segmentation opportunity that’s chronically underused. Behavioural signals — session depth, feature interaction history, recency of specific actions — can inform which users see which discovery prompts, and when. Teams that have that segment logic in place can run controlled exposure and actually measure lift. Teams without it are essentially running a single-treatment experiment with no control group.

---

Allocating Research Resources Where Discovery Actually Breaks

Nielsen Norman Group’s framework on Research Accountability Score (RAS) offers a useful lens here. The core argument from Brian Utesch and Tammi Fitzwater: UX research teams tend to allocate effort based on stakeholder visibility and project momentum, not on where insight will actually change outcomes. The result is over-investment in concept validation and under-investment in post-launch behaviour analysis — exactly the phase where feature discovery failures become measurable.

For design and product managers in Southeast Asian organisations, where UX research capacity is often thin relative to the pace of shipping, RAS-style thinking is a forcing function. It asks: if we have budget for two research sprints this quarter, which gaps in our understanding will most change what we build? Applied to feature adoption specifically, that usually points toward usability testing with genuine non-readers — not the design-literate participants who tend to self-select into research panels.

The failure mode to watch for: teams that do the research, surface the insight that users aren’t engaging with onboarding copy, and then respond by rewriting the copy. Better copy is easier to ship than a redesigned discovery mechanism. It’s also usually insufficient.

Building Discovery That Scales Across Channels

One dimension that doesn’t get enough attention in feature discovery discussions is consistency across surfaces. A user who encounters a new capability inside your mobile app will have a completely different cognitive frame when they see it referenced in a CRM email, a push notification, or a web dashboard. If the visual language and interaction pattern don’t carry across those touchpoints, you’re creating discovery friction — the feature exists, but it feels unfamiliar each time.

This is where design system investment pays a specific, measurable dividend. Brands with mature component libraries can propagate new feature UI patterns across channels in days, not weeks. For markets like Singapore and Malaysia where users routinely switch between app and web depending on device context, that consistency isn’t aesthetic — it’s conversion-relevant. Tokopedia’s cross-surface coherence during feature rollouts is a reasonable regional benchmark: new mechanics are introduced in-app first, then reflected in web and push with enough visual consistency that the feature feels already-known rather than new.

The implementation caution: multi-language interfaces add meaningful complexity here. A tooltip that works elegantly in English at 12 words may run to 20 in Bahasa Indonesia, breaking the layout and forcing a copy truncation that removes the key action cue. Build discovery components with text expansion built into the spec, not retrofitted after QA flags it.

---

Key Takeaways

• Design feature discovery for the user who will never read your copy — contextual triggers at the moment of relevance outperform announcement modals by a significant margin.
• Apply RAS-style resource logic to UX research: prioritise post-launch behaviour analysis over concept validation when adoption metrics are the actual success criterion.
• Scale discovery patterns through your design system to maintain cross-surface consistency — especially critical in Southeast Asian markets where users move fluidly between app and web contexts.

The deeper question for product and marketing teams is whether feature adoption is even being tracked as a distinct metric, separate from acquisition and retention. Most analytics stacks can surface it — the behaviour signals are there. Whether anyone has built the segment logic to act on them is a different matter. What would your roadmap look like if feature discovery failure rates were as visible as churn?

---

At grzzly, we work with growth and product teams across Southeast Asia to connect UX decisions to the data infrastructure that makes them measurable — from segment logic that powers contextual discovery to cross-channel design systems built for multilingual scale. If feature adoption is a gap in your current stack, we’d like to think through it with you. Let’s talk