There is something interesting about the way we speak in Kenya.
You can travel a relatively short distance and suddenly notice that the words have changed.
The rhythm has changed.
The expressions have changed.
Sometimes even the meaning of a familiar word has shifted.
And yet, somehow, you still understand.
I have been thinking about this a lot lately because I grew up in Eastleigh, Nairobi, in an environment where language was never simply language.
It was identity.
It was geography.
It was age.
It was friendship.
It was humour.
It was belonging.
And it was constantly evolving.
When I was growing up, a casual greeting could be:
“Sasa?”
How are you?
Or:
“Niaje?”
And the response might be:
“Fit.”
Or:
“Poa.”
Sometimes the conversation barely needed formal English at all.
It was Sheng.
And it wasn’t something we sat down and learned.
It evolved around us.
We absorbed it.
We modified it.
We played with it.
We carried it with us.
And eventually, what began as contemporary urban language became part of the cultural fabric of our generation.
That experience has made me look differently at something I now spend much of my professional life designing:
the digital interface.
Because if people naturally change the way they speak depending on where they are, who they are speaking to and what environment they are in…
shouldn’t digital experiences understand some of that context too?
Language is not just words
In conventional web design, language is often treated as content.
The designer creates the interface.
The developer builds it.
Then someone writes the words that go inside it.
But language does much more than fill empty spaces between buttons.
Language creates tone.
It establishes trust.
It communicates belonging.
It signals identity.
It tells people whether something feels familiar or foreign.
Consider the difference between:
“How may we assist you?”
and
“How can we help?”
And then imagine a context where the brand could naturally say:
“Unahitaji msaada?”
Or perhaps, depending on the audience and brand:
“Tuko hapa kukusaidia.”
Or something even more culturally specific.
The important question isn’t which one sounds more sophisticated.
The question is:
Which one sounds like it belongs here?
That is a design question.
I grew up watching language move
Eastleigh taught me something about language that I didn’t have the vocabulary to articulate when I was younger.
Language moves.
It travels through communities.
It borrows.
It mutates.
It adapts.
It creates new meanings.
Sheng is a fascinating example.
It emerged from the interaction of languages and cultures in urban Kenya, particularly Nairobi, and has continually evolved through generations, neighbourhoods, music, media and popular culture.
But there isn’t necessarily one fixed Sheng.
That is what makes it interesting.
The Sheng you encounter in one part of Nairobi may carry different vocabulary, rhythm or references from what you encounter somewhere else.
Eastleigh has its own social and linguistic environment.
Move toward other parts of Nairobi and you begin to notice differences.
Leave Nairobi and the landscape changes again.
Come to Mombasa, where I now live, and you may hear:
“Vipi?”
instead of the familiar Nairobi:
“Sasa?”
The conversation still works.
But the texture has changed.
And that texture matters.
Geography can become language
This is one of the things I find fascinating about African urban culture.
We often think of geography as something physical.
East.
West.
North.
Coast.
City.
Neighbourhood.
But geography also produces social and linguistic environments.
People develop expressions around their communities.
They borrow from neighbouring languages.
They respond to music.
They respond to migration.
They respond to technology.
They respond to youth culture.
They respond to what is happening around them.
Language becomes a living map.
You can sometimes hear where someone comes from without asking.
And when you move, you adapt.
I experienced that myself moving between Nairobi and Mombasa.
The difference isn’t simply vocabulary.
There is a different rhythm.
A different cultural atmosphere.
Different references.
Different ways of expressing familiarity.
Different expectations around conversation.
You learn to adjust.
You code-switch.
You adapt.
And most of us do this without consciously thinking about it.
We are all already designing for context
Think about how you speak to:
Your mother.
Your closest friend.
Your employer.
A customer.
A child.
A stranger.
A government official.
A WhatsApp group.
You probably don’t use exactly the same language in all those environments.
You adapt.
The vocabulary changes.
The tone changes.
The formality changes.
Sometimes the language itself changes.
That is contextual interaction.
And it happens naturally.
Yet many digital interfaces behave as though everyone interacts with a brand in exactly the same way.
The website says one thing.
The WhatsApp message says another.
The social media account sounds completely different.
The chatbot sounds robotic.
The customer service representative speaks naturally.
The result is a fragmented experience.
The brand has no consistent linguistic personality.
Code-switching is not confusion
Code-switching is sometimes treated as evidence that someone isn’t speaking “properly.”
I see it differently.
Code-switching can be a sophisticated communication behaviour.
It allows people to move between identities, contexts and communities.
In Kenya, someone might move between English, Kiswahili, Sheng and an indigenous language within the same conversation.
Sometimes within the same sentence.
And everyone understands what is happening.
The switch itself communicates something.
It can signal:
Familiarity.
Humour.
Authority.
Intimacy.
Distance.
Professionalism.
Playfulness.
This is incredibly relevant to UX.
Because digital interfaces are also constantly asking us to establish relationships.
The question becomes:
Can an interface adapt its language to the relationship it is building with the user?
What would a culturally intelligent interface sound like?
Imagine a Kenyan financial service.
Its homepage speaks formal corporate English.
Its Instagram account speaks casually.
Its customer support uses Kiswahili.
Its WhatsApp agent uses a mixture of English and Sheng.
Its AI assistant speaks like a generic international chatbot.
There is nothing technically broken.
But the experience feels disconnected.
Now imagine a system designed intentionally around language context.
The formal website maintains clarity and credibility.
The educational content can use accessible Kiswahili where appropriate.
The social experience becomes more conversational.
The WhatsApp journey becomes naturally human.
The AI assistant understands local terminology and common expressions.
The brand doesn’t necessarily use Sheng everywhere.
That would be forced.
Instead, it understands when different forms of language are appropriate.
That is the difference between translation and cultural intelligence.
Don’t make every website speak Sheng
This is an important distinction.
Cultural Interaction Design should never become:
“Put some Sheng on the website and call it African UX.”
That would be superficial.
A bank may need to communicate authority.
A hospital needs clarity.
A petroleum company needs technical precision.
A luxury brand may need restraint.
A youth fashion brand might have much more linguistic freedom.
A government service needs accessibility across very different audiences.
The cultural question is therefore not:
“How much Sheng should we put on the website?”
It is:
“What language, tone and vocabulary will make this experience meaningful to this particular audience in this particular context?”
Sometimes the answer will be Sheng.
Sometimes Kiswahili.
Sometimes English.
Sometimes a combination.
Sometimes none of them.
Context is the design material.
This also changes how we think about SEO
This is where my work in SEO and AEO becomes particularly interesting.
People don’t always search the way businesses describe their products.
A business might call something:
“Automotive Liquefied Petroleum Gas Solutions.”
A customer might search:
“AutoGas near me.”
Or:
“LPG for my car.”
Or:
“Where can I convert my car to gas?”
Or something even more conversational.
The words people use reveal how they understand a problem.
Search is therefore not only a technical optimisation exercise.
It is also a language research exercise.
If we want businesses to be discovered, we need to understand the language people actually use when they are looking.
And this becomes even more important as search moves toward conversational AI.
People increasingly ask questions rather than type fragments.
They describe situations.
They explain problems.
They use natural language.
AI systems interpret those questions and attempt to connect them with useful information.
Which means the future of search visibility is increasingly connected to understanding human language and intent.
AEO and AI make this even more interesting
Answer Engine Optimization and AI Search Visibility aren’t simply about adding keywords.
They are about helping systems understand:
What does this business do?
Who is it for?
Where does it operate?
What problems does it solve?
What terminology does its audience use?
What questions does that audience ask?
What makes this business authoritative?
And importantly:
What context surrounds those questions?
A business in Mombasa may need to be understood differently from a similar business operating in Nairobi.
A tourism company targeting international travellers speaks differently from one targeting local residents.
A youth brand communicates differently from a corporate institution.
The language of discovery is part of the experience.
Language should influence UX writing
This also means we need to rethink microcopy.
A button isn’t just a button.
Its words create an emotional instruction.
Compare:
Submit
with:
Send enquiry
with:
Talk to us
with:
Let’s get started
with:
Tupatane
Each creates a different relationship.
Each assumes something about the visitor.
Each creates a different emotional temperature.
This is why I increasingly believe that UX writing should be treated as interaction design.
Words determine what people understand.
They influence what people expect.
They can reduce anxiety.
They can create confidence.
They can make an experience feel human.
Or they can make it feel like a machine.
The AI interface has a language problem too
We are now entering another interesting phase.
Interfaces are becoming conversational.
Chatbots.
AI assistants.
Voice interfaces.
AI search.
Digital agents.
The interface itself is beginning to speak back.
That creates an enormous opportunity.
But it also creates a danger.
If every AI system speaks in the same polished, generic global English, we may simply replace one kind of digital homogenisation with another.
What does an AI assistant sound like when it understands Kenyan context?
What does it know about local expressions?
How does it respond when someone switches between English and Kiswahili?
What happens when someone asks something using Sheng?
Can it understand the intent without awkwardly translating everything into formal English?
These are not merely AI engineering questions.
They are interaction-design questions.
Designing for a multilingual Africa
Africa contains thousands of languages and enormous cultural diversity.
There is no single linguistic solution.
That is precisely why context becomes so important.
A good cultural interaction designer needs to ask:
Who are we designing for?
Where are they?
How do they speak?
What words do they use?
What references do they understand?
What makes them trust something?
What tone feels natural?
When do they switch languages?
What changes when they move from one context to another?
These questions belong in the design process.
Not after it.
From translation to interpretation
I think there is an important distinction between translation and interpretation.
Translation asks:
What is the equivalent word in another language?
Interpretation asks:
What does this person actually mean, in this context?
Digital experiences need both.
But culturally intelligent experiences need the second one much more.
Because culture is rarely a one-to-one translation.
A phrase can be grammatically correct and still feel completely wrong.
A corporate sentence can be perfectly written and still feel distant.
A translated interface can be technically accurate and culturally awkward.
The goal isn’t simply to speak another language.
The goal is to speak in a way that makes sense to the person experiencing it.
This is where my own journey is taking me
The more I work across web design, interaction design, SEO, AEO, AI, prompt engineering and storytelling, the more I see these disciplines converging.
They are all, in different ways, concerned with one thing:
meaning.
What does a business mean?
What does a user mean?
What does a search query mean?
What does a cultural expression mean?
What does an interaction mean?
What does a brand sound like when it speaks?
And how do we translate all of that into a digital experience?
This is the direction I am increasingly exploring through KYTCH and Cultural Interaction Design.
I want to build websites and digital experiences that don’t simply function.
I want them to feel contextual.
I want to understand the language around the business.
The people.
The place.
The behaviour.
The culture.
The stories.
And then use those things as design inputs.
The web doesn’t have to sound the same everywhere
Perhaps that is the larger point.
The internet has given us incredible access to global design patterns.
But global access can also create global sameness.
The same layouts.
The same terminology.
The same stock photographs.
The same corporate tone.
The same interfaces.
The same experience.
I don’t think the answer is to reject global design.
I think the answer is to add context to it.
A Kenyan website can be world-class without sounding like it was written in Silicon Valley.
A Mombasa brand can be technologically sophisticated without losing its local character.
A Nairobi startup can speak globally while still retaining the linguistic personality of the people who created it.
And an African digital experience can be modern without becoming culturally anonymous.
The question I am now asking
I started building websites because I was fascinated by technology.
Today, I am still fascinated by technology.
But I am even more fascinated by people.
How we speak.
How we move.
How we understand.
How we connect.
How we adapt.
How we make meaning.
And how all of that can become part of the digital experiences we create.
So perhaps the question isn’t:
“How do we make websites speak African languages?”
The deeper question is:
“How do we make digital experiences understand the people who speak them?”
That is the question I am carrying into the next phase of my work.
Not simply designing websites.
Not simply optimising websites for search.
Not simply building AI systems.
But exploring how language, culture, behaviour and technology can come together to create a web that feels more human, more contextual and more alive.
KYTCH: turning identity, culture and intelligence into distinctive digital experiences.
