Why Human Context Is Your Most Underused Design Asset

There’s a quiet assumption baked into most design briefs: that if the interface is clean and the hierarchy is clear, the work is done. Function shapes form, and form serves the user. Tidy. But two things keep breaking that model — audiences who carry complex cultural identities, and AI tools that generate form without any sense of what the function is actually for.

The gap between those two failure modes is where the most interesting design work is happening right now.

Identity Is Not a Brand Mood Board

Designer Samar Maakaroun’s recent visual identity for The Mosaic Rooms — a London arts space focused on Arab culture — is worth studying carefully, not because it’s beautiful (it is), but because of the brief it answered. The challenge wasn’t to make Arab identity look palatable. It was to represent what It’s Nice That describes as an identity that is “entangled, messy” and filled with “often fraught history” — without flattening it into a single aesthetic shorthand.

Maakaroun’s solution used abstracted Arabic scripts and non-linear compositional logic to hold that tension visually. The result doesn’t resolve into a single reading. It asks you to sit with ambiguity.

For brand teams in Southeast Asia, that’s a directly applicable lesson. Multilingual markets — where a campaign might run in Thai, Bahasa, Tagalog, and English simultaneously — carry similar complexity. A visual system that tries to be neutral across all of them usually ends up being meaningless to all of them. The sharper move is to design from a specific cultural anchor and build deliberate flexibility around it, rather than starting from the middle and expanding outwards.

Practically: audit whether your design system has a cultural point of view, or just a colour palette. They’re not the same thing.

When Function Refuses to Disappear

Smashing Magazine’s Kyrylo Levashov makes an argument that sounds obvious until you think about how rarely it’s applied: if a system’s function can’t be made completely invisible, the interaction itself becomes part of the user’s experience — whether you designed it that way or not.

He identifies four common design assumptions that produce friction precisely because they treat the tool as neutral. The underlying error is treating UX as a layer applied on top of a functional system, rather than as something that begins the moment a user encounters any part of that system — including error states, loading sequences, and onboarding flows that product teams consider “temporary.”

In Southeast Asian markets, this has a sharper edge. Mobile-first usage means users are frequently operating on slower connections, smaller screens, and with apps competing for RAM. A checkout flow that feels seamless on a desktop in a broadband environment becomes a friction-laden obstacle on a mid-range Android on a 4G connection in a tier-2 city. The function hasn’t changed. The experience is completely different.

The implementation implication: design reviews should include device and connectivity simulation as standard, not as an accessibility afterthought. If the interaction has to exist, own it.

---

What AI Cannot Carry Into the Room

UX Design’s Fabricio Teixeira recently curated a sharp observation about AI-generated content that applies directly to design: “Real, true communication between people, especially through art, requires emotion, connection, context, and meaning. Those are all things that artificial intelligence lacks within the content it generates.”

From a data perspective, I’d frame this differently but reach the same conclusion. AI tools are excellent at pattern completion — they can generate a layout that statistically resembles successful layouts, or suggest a colour scheme that correlates with high engagement in training data. What they can’t do is know why a specific audience segment responds to a visual cue, or what cultural weight a particular image carries for a Grab user in Kuala Lumpur versus a Shopee seller in Surabaya.

That contextual layer has to come from human research — and it has to be built into the design brief before AI tools enter the workflow. Teams that hand a vague brief to an AI generator and then try to add cultural nuance in the revision cycle are working backwards. The emotional and contextual logic needs to exist upstream, so it can actually constrain the tool’s output rather than being pasted over it afterward.

Budget note: investing in a structured discovery phase — even two focused weeks of audience research with in-market users — consistently produces better downstream efficiency in design iteration than skipping straight to generation. The rework cost of culturally misaligned assets is rarely accounted for in project timelines, but it shows up.

The Human Touch as a Measurable Variable

Here’s where the data framing earns its place: human context isn’t just a creative principle, it’s a segmentation variable. Audiences who recognise their own cultural identity in a brand’s visual language behave differently — longer session durations, higher return visit rates, stronger word-of-mouth signals. These are measurable outcomes, not soft brand equity abstractions.

The practical architecture for this: build cultural context signals into your audience taxonomy early. If your segments are purely demographic or behavioural, you’re missing a layer that directly influences how design decisions should be made. A segment defined by language preference, platform behaviour (LINE vs WhatsApp, Lazada vs Shopee), and cultural occasion patterns gives your design team something specific to work against — rather than a persona that could apply to anyone, anywhere.

Design that encodes genuine human context doesn’t just feel better. It performs better. The question most teams haven’t answered yet is whether they’ve built the research infrastructure to know what that context actually is.

---

Key Takeaways

• Build your design system around a specific cultural point of view, then engineer flexibility — starting from neutrality produces assets that resonate with no one in particular.
• Treat every user interaction — including error states, load times, and onboarding — as designed experience, not infrastructure; in mobile-first SEA markets, the gap between desktop and real-world conditions is a conversion issue.
• Use AI generation tools downstream of human research, not as a substitute for it; the contextual brief is where competitive differentiation is actually built.

The question worth sitting with: if your current design process were handed entirely to an AI system tomorrow, what would it lose that your audience would actually notice? The answer to that is probably where your real brand equity lives — and it’s worth knowing whether you’ve documented it anywhere.

---

At grzzly, we work with Southeast Asian brand teams to close the gap between audience data and design decision-making — building the contextual briefs and segmentation frameworks that make creative work land with the right people. If your design process is generating assets faster than it’s generating insight, that’s a conversation worth having. Let’s talk