Fourteen years in Southeast Asia, and what they taught me about growth

1

One sentence, four ways

Singapore has four official languages. Grow up there and you learn something early: a sentence is never just a sentence. The same words can be warm in one room, rude in another and meaningless in a third.

I lived in Singapore from 2007 to 2020, and spent about fourteen years in Southeast Asia in all. I went to school and university there, finishing a Business Administration degree in 2018. It was the background of my life, the way weather is the background of a town.

It turned out to be the most useful thing I'd ever learned about marketing. Before you decide what to say, find out who is listening, and what they will hear. Everything else in this story is that question, asked in a new place.

What I took from itA message doesn't have one meaning. People decide the meaning, so I start with the people.

2

My first jobs happened on a screen

The question followed me into work, and it got harder. My first marketing jobs were remote. The first was for a company in the Netherlands: social media, copy, market research and campaign support. Then came a business publisher, where I worked on content, search, email newsletters and events, and won a Rising Star Award in 2021.

Distance brought a puzzle. What does a customer think when you can't see their face? The answer I found was to read people by what they leave behind: what they click, what they skip, what they ask support. A search term is a sentence a stranger said out loud when they thought nobody was listening.

3

Four creators and a USD 300,000 target

Then came the first real test of whether the habit would scale. A nail polish brand wanted to sell to mothers and beauty fans in the US and Canada over one holiday season. The platform I worked for was in the Netherlands. The target was USD 300,000 in sales across four months.

I wasn't one of those customers. What I had was a habit: write down who they are, let creators speak to them in their own way, and let the numbers overrule me. I directed the audience personas and the creative. We worked with four creators, tested 120 versions of the ad copy, and moved the paid budget to whichever version was selling.

The campaign sold USD 1.26M. That's 319% above the target, with a 6.3x return on ad spend on Meta.

What I took from itYou don't have to belong to a culture to serve it. You have to listen to it faster than your assumptions can get in the way.

4

The launch that never launched

Not everything I planned reached a customer. At a Web3 gaming studio I planned the marketing for two games, with teams in Indonesia, Singapore and India. One game shipped. The other was cancelled before release, in the crypto winter.

I still learned from the one that never launched. Three countries meant different time zones, different habits and one plan that had to hold them all. With teams in several countries, the plan has to be written so clearly that it survives translation and time zones, and it has to say who decides.

5

The page that had to earn a stranger's trust

Trust is the hardest thing to carry across a screen. A telepsychiatry company in New York had patients in seven US states, about USD 20,000 a month in paid media, and one stubborn number: only 1.8% of visitors booked a consultation.

Think about what that page asked a stranger to do. Trust a psychiatrist they had never met, over a screen. More budget would not have fixed that. So I built a landing page for each condition, so the page matched what the person had searched for. I ran A/B tests on trust. I moved 40% of the Meta brand spend into search, where people were already looking to book. The booking rate reached 5.0%, and daily qualified calls went from 12 to 20. The case study has the detail.

What I took from itTrust looks different in every category and every country. The only way to learn what it looks like is to test it.

6

A market with no price list

The next lesson ran the other way. For a while I worked independently with early-stage clients on budgets under USD 1,000 a month. I cut their acquisition cost by 20% by fixing the funnel before adding spend, and one of them hired me full-time.

Then a games publisher in Singapore asked me to find studios in Indonesia, and the scene had no going rate for a deal. Nobody could tell either side what a game was worth. Until then I had mostly listened to customers. This time I had to listen to the people across the table.

I built a model to price each deal, and I went to meet the studios. Two campus events at a university in Surabaya and an online pitch event brought most of them to us. We vetted fifty and signed five.

7

The tools changed the job

By then I was using AI in my daily work, as I had been since 2023, and it felt like a new language: fluent, confident and sometimes wrong. So the old question came back in a new form. Who is listening when a machine writes? A person has to be.

At a B2B agency with clients in the US and Europe, my client book grew from two accounts to five, and we kept all five. I led a growth team of around 30 people through three senior managers. Behind the numbers, weekly competitor and account research went from four hours to one, using Claude, n8n and Make, and we launched new AI agents for existing clients.

What I took from itAI is another language to learn. Learn where it's strong, put a person where it's weak, and write the weekly work down before you automate it.

8

What I'm still learning

Eight years into marketing, three questions keep coming back.

How should AI change planning itself, the research, the structure and the first draft, while a person still makes the calls? How do you trust the result of a test when AI lets you run ten times as many? And what will teams that sell across borders look like when the repeat work belongs to machines?

I'm working through them in public, with the planning models I use laid out in plain words and the industry articles.

Singapore taught me that one sentence can land four ways. The question I learned there is still the one I start with: who is listening, and what will they hear?

If you're building a team that sells across languages and markets, I'd like to hear what you're stuck on. adityamg96@gmail.com.

The years, roughly