Kevin Semanda is a Brand Consultant, Brand Identity Designer, and UI/UX Designer with more than a decade of experience helping businesses build distinctive brands and digital experiences. In this interview, Kevin Semanda shares his thoughts on brand identity, design philosophy, AI, visual hierarchy, and why the strongest brands begin with values rather than visuals. Drawing on his experience across branding, strategy, and user experience, he explains why meaningful design is built on clear thinking long before it becomes visible.
“I got into it for the surface, I stayed for the argument underneath it.”
What drew you into design as a discipline, and what keeps you engaged in it today?
What drew me in was the discovery that design isn’t decoration, it’s argument. Early on I think most people are pulled toward design because they like making things look good. What kept me wasn’t that — it was realizing that every visual choice is actually a claim about what something is. A typeface isn’t neutral; it’s a position. Spacing isn’t filler; it’s confidence or its absence. That reframing turned design from a craft of taste into a craft of thinking, and that was the hook.
What keeps me engaged today is the gap between what a brand looks like and what it is. Most briefs still start with “what should this look like?” — and chasing the harder, earlier question of “what does this refuse to be?” is where the real work lives. That question doesn’t get easier with experience; every brand, every founder, every category resists it differently. The aesthetics are the easy 20%. Finding the identity underneath is the part that still feels unsolved every time, and that’s exactly why it hasn’t gotten boring.
So: I got into it for the surface, I stayed for the argument underneath it.
When working on a project, what signals tell you that a design is doing its job well?
The signals I trust most are rarely about whether something looks good — that’s the easy, unreliable signal. The ones that actually matter:
It survives the strip test. If you remove the logo, the palette, the typographic system, and something still feels distinctly like this brand and not a generic competitor, the design is doing real work. If it could belong to anyone, it’s decoration, not identity.
It makes refusal visible. A design that’s doing its job shows you what it chose not to do — the restraint reads as confidence, not absence. If I can point to white space, an unused color, an omitted feature, and say “that’s a decision, not a gap,” that’s a strong signal.
People describe the brand in character terms, not visual terms. When a client or user says “it feels confident” or “it feels like it respects me” rather than “I like the blue,” the personality layer has actually landed. Visual-only feedback (“looks clean,” “looks modern”) means the identity hasn’t fully transferred yet.
It holds up under translation. Good identity work survives moving from a homepage to a packaging label to a tone-of-voice guideline without losing its shape. If the system only works in one format, it was never really systemic — it was a one-off look.
It produces an obvious “no.” When new requests come in — a new feature, a new campaign, a partnership — a working identity makes some directions feel instantly wrong without a debate. That instinct, shared across a team without a style guide open, means the values layer actually took.
The weakest signal, by contrast, is immediate visual approval. Polish is easy to manufacture and easy to fake; distinctiveness under pressure is not.
Which part of your workflow tends to slow you down the most, and why?
The visual execution is almost never the slowest part, it’s spending enough time in the values and personality layers before going for form.
There is a strong pull, from clients and from my own instincts, to leap ahead to something visible. A moodboard or logo direction feels like progress. Sitting with “what does this brand refuse to be?” feels like stalling, even when it’s the actual work. That mismatch between what feels productive and what is productive creates real friction — clients want to see something, and the temptation to produce a placeholder visual just to show momentum is constant.
The second drag is translating an abstract conviction into language precise enough to brief against. Knowing a brand’s personality intuitively is one thing; writing it down in a way that a typographer, a copywriter, and a motion designer can independently apply without flattening it into platitudes (“bold,” “authentic,” “human”) is much harder. Vague language at that stage poisons everything downstream — you end up reverse-engineering the values from a visual system that was built on fog instead of fixing the fog first.
So the bottleneck isn’t drawing or designing. It’s the discipline of refusing to let the visual language arrive early, and the precision required to make the upstream thinking specific enough to actually constrain the downstream choices. Once that’s solid, execution moves fast — almost suspiciously fast, which is usually a sign the foundation was right.
When designing for attention, how do you decide what deserves the most visual
priority?
Priority follows the same logic as the rest of the philosophy: it’s not decided by what looks most important; it’s decided by what the brand’s values say matters most. A few filters I actually run:
What’s the one thing this brand refuses to let you miss? Every project has a single non-negotiable—sometimes it’s a price, sometimes it’s a single line of conviction, sometimes it’s just the name. Everything else is the supporting cast. If three things are fighting for primary visual weight, that’s a sign the values layer wasn’t resolved before the form layer started.
What earns weight versus what claims it. Size, color, and position are loud, but they’re also the cheapest tools—anything can be made big and red. Real priority comes from contrast against restraint: if most of the composition is quiet, the one loud element actually reads as important. If everything competes for attention, nothing has it. So I’m less interested in “make this bigger” and more interested in “what can I take away so this is the only thing left standing.”
Sequence over hierarchy. Rather than asking what’s biggest, I ask what someone needs to understand first, second, third—attention as a path, not a pile. That usually means resisting the instinct to put the logo, the headline, and three supporting claims all in the top third competing for the same glance.
What the audience already trusts the brand to lead with. Personality dictates pacing—a brand that’s quiet and confident earns trust by saying less with more authority; a brand that’s energetic and immediate can afford density and motion. Visual priority that doesn’t match the personality reads as noise even if the hierarchy is technically correct.
The test, again, is restraint: if I can’t justify why something has the loudest visual weight in terms of the brand’s actual priorities—not just what’s easiest to make stand out—it doesn’t deserve that priority yet.
Can you share an example of a design decision that didn’t perform as expected? What did you learn from it?
One concrete case: on Tuuko, a Uganda-focused trivia and card game, I built the feature and game-mode sections around a custom icon set—small symbolic marks meant to carry meaning at a glance (multiplayer, tournaments, leaderboards, knowledge categories) without leaning on stock iconography that would’ve felt generic against a brand built on Ugandan cultural specificity.
It looked right in the design file. The icons were distinctive, on-brand, tonally consistent with the dark gold palette. But once it shipped, the choice didn’t hold up: the custom glyphs failed to render reliably across environments, and even where they did render, several of them weren’t legible enough at small sizes to communicate their meaning without a label doing all the actual work. A “Tournaments” icon and a “Leaderboards” icon ended up reading as near-identical shapes unless you already knew which was which.
What that exposed wasn’t really an icon problem—it was a sequencing problem. I’d let a visual-language decision (custom marks for distinctiveness) outrun the harder upstream question: Did the brand actually need icons here, or did it need clarity? Distinctiveness was treated as the goal when legibility should have gated it. The icons were doing identity work—signaling “this isn’t a generic trivia app”—at the expense of doing their actual job, which was helping someone scan a page in two seconds and understand what each card meant.
What I learned and now run as an earlier check: distinctiveness and legibility aren’t on the same axis, and a visual system has to pass both independently. A mark can be completely on-brand and still be the wrong choice if it fails at the literal task it exists to perform. Now I test custom iconography the way I’d test a typeface—at the smallest size it’ll actually appear, in the actual rendering environment, before it’s locked into the system. If a symbol can’t do its functional job without a crutch (a label, a tooltip, or prior knowledge), it gets simplified or cut, even if it’s the more “interesting” mark. The brand’s distinctiveness has to live in things sturdier than a glyph that might not survive shipping—tone of voice, spacing, and the overall character of the page—not in details that are this easy to lose.
What changes in tools, technology, or user behaviour do you think will redefine design in the near future?
anyone can produce a competent logo, palette, and layout system in minutes, “looks professional” stops being a meaningful signal of quality—it’s the baseline, not the bar. That doesn’t kill the need for visual designers; it raises the stakes on the part AI can’t generate well, which is the upstream thinking: what does this brand actually believe? What does it refuse to be? What’s the one thing it needs people to feel? The craft shifts from “can you make something good-looking” to “can you make something true.” I expect a real split between brands that have a strong, specific point of view and brands that have generically competent AI output—and the gap between them will be more visible than it’s ever been, precisely because the floor for visual quality got so much higher.
Generative tools also make iteration nearly instantaneous, which changes the workflow risk. It used to be expensive to explore ten directions, so teams converged early out of necessity. Now you can generate fifty. The danger is mistaking volume of options for clarity of thinking—more directions doesn’t mean better direction if the values layer underneath was never resolved. The designers who get faster without getting better are the ones who let the tool generate the identity instead of the visual language. The discipline of deciding what the brand refuses to be becomes more important, not less, when production is cheap.
On the user-behavior side, attention is fragmenting across more surfaces with less context per surface—a brand gets seen in a three-second social clip, a notification, a voice assistant response, or a screenshot shared without the original page. That punishes systems that only work in their “home” format and rewards identity that’s genuinely portable: tone of voice that survives being quoted out of context, a visual signature recognizable even cropped badly. The strip test gets more literal—your brand will actually get stripped down by the platform, by the user, by compression. Designing only for the polished full-context version is designing for a context that increasingly doesn’t exist.
And the interface itself is shifting from things you look at to things you talk to or that act on your behalf—conversational, agentic, voice-driven. That’s a real challenge to identity work that’s lived almost entirely in static visual form. A brand’s personality has to be expressible in how it responds, not just how it looks, which means tone of voice and behavioral character move from being one input among several to being load-bearing in a way they weren’t before. Visual designers who only think in static compositions will be working in a shrinking fraction of where the brand actually shows up.
The constant across all of it: tools and surfaces keep changing, but the brands that hold up are still the ones with a real position, not just a competent look. That part doesn’t get automated—it just gets more exposed.
What mindset or habit has been the most valuable in your growth as a designer?
he habit that’s mattered most is treating every visual instinct as a hypothesis to interrogate, not a conclusion to defend.
When something feels right—a color, a layout, a tone—the easy move is to trust the feeling and move forward. The harder, more valuable habit is pausing on that instinct and asking what it’s actually claiming about the brand’s identity, and whether that claim is true or just familiar. A lot of design taste is really just exposure—you reach for what you’ve seen work before—and that’s not the same as reaching for what’s right for this specific brand. Separating those two has probably done more for the quality of my work than any single skill upgrade.
Concretely, that habit shows up as a refusal to let visual confidence substitute for visual justification. If I can’t explain why a decision serves the brand’s values and personality—not just why it looks good—I treat that as unfinished thinking, not a stylistic choice I’m entitled to. That’s slower at the moment. It’s also the difference between a system that holds up under scrutiny (a client pushback, a new format, or an edge case) and one that only works because nobody asked it to defend itself yet.
The second habit, downstream of the first: actually running the strip test on my own work, not just preaching it. It’s uncomfortable to remove the logo and the palette from something you’re proud of and ask honestly whether anything distinctive survives. Most early work doesn’t pass. Doing that check repeatedly, especially when it’s disappointing, is what gradually moved the bar from “this looks good” to “this is actually someone.”
Neither of those is really a design skill in the traditional sense—they’re more like intellectual discipline applied to a visual craft. But that’s been the actual lever. The technical execution caught up on its own; the harder growth was learning not to trust my own first instinct without making it prove itself.
Key Facts
- Name: Kevin Semanda
- Location: Kampala, Uganda
- Role: Brand Consultant & Designer, Founder of Grafrika
- Specialties: Brand Identity, UI Design, Brand Strategy, Graphic Design
- Focus Areas: Brand Identity, Visual Systems, User Experience, Editorial Design
- Experience: 10+ years in branding, graphic design, UI design, and creative strategy
- Tools: Adobe Illustrator, Photoshop, InDesign, After Effects, Sketch, Figma
- Education: B.A. (Hons) Graphic Design for Digital Media, Glasgow Caledonian University
Support the designer and follow:
