Most growth marketing advice quietly assumes you already have something to work with. An email list. A pixel with six months of data on it. A few thousand followers. Some baseline you can optimise.
I didn’t have any of that.
I work on a cross-border money transfer app. People living abroad send money home to Vietnam and the Philippines. When I started, the user database was empty in the literal sense. No list, no retargeting audience, no reviews, no community, no organic traffic. Zero.
Getting your first users in that situation is a completely different job from growing a product that already works. There is nothing to optimise. There is only outreach, one person at a time.
This is the first post in a series on what I actually did about it. Not a framework I read somewhere. The specific steps, in the order I took them, including the parts that felt slightly embarrassing at the time.
Your First User Should Be You
Before I messaged a single stranger, I used the product myself. Properly. I downloaded it, went through KYC like anyone else, and sent real money to my own family.
This sounds obvious. Most marketers skip it anyway, because you can write perfectly good ad copy off a feature list and nobody will know the difference.
But you will know. And you will miss things:
- How long the verification step actually takes when your ID photo is slightly blurry
- What the app does when you type a recipient bank name in Vietnamese with no diacritics
- Whether the person receiving the money understands what just landed in their account
- The exact moment where you, a person who works on this product, feel a flicker of doubt
That flicker of doubt is the most valuable thing you will find in your first week. Write it down. It is the same doubt your users will feel, except they will not push through it the way you do.
Your Next Few Users Come From Your Own Network
My second move was the least scalable thing possible: I messaged people I already knew who live abroad. Cousins, former colleagues, university friends, friends of friends.
Not a public post. Individual messages, one at a time, written like a person and not a campaign.
Twenty personal messages will outperform one Facebook post to five hundred people, every single time. The post gets three likes and no replies. The messages get you five or six real users and a handful of honest reactions.
Here is the honest caveat, though. Your own network is biased. They like you. They will download the app to be supportive, say nice things, and never mention the part they found confusing. So use them for one thing only: finding bugs and breakages before strangers hit them. Do not use them to validate whether the product is any good. They are not qualified for that, because they are your friends.
No Friends Abroad? Find Your First Users on TikTok
This is the part most people get stuck on. If your whole network lives in the same city as you, and your product is for people on the other side of the world, where do you even begin?
You go to where your first users already talk about their lives. For the Vietnamese diaspora, that is TikTok.
Say you want to understand Vietnamese people living in the UK. Search vietnam uk. Then search it the way they would actually say it: người Việt ở Anh, du học sinh Anh, cuộc sống ở UK, làm nail ở Anh. Each search gives you a wall of real accounts belonging to real people who fit your audience exactly.
Then read the comments. The comment section under a video about working in the UK is a better research document than any market report you can buy, because nobody in there is performing for a survey.
And then you do the unglamorous thing. You message them. One by one.
What the First Message Should Look Like
Do not pitch in the first message. Nobody downloads an app because a stranger sent them a link.
Keep it short, human and specific to their content. Something close to this:
Hi chị, em xem video của chị về cuộc sống ở UK. Em đang làm marketing cho một app chuyển tiền về Việt Nam và đang tìm hiểu xem mọi người bên đó thường gửi tiền kiểu gì. Chị cho em hỏi vài câu được không ạ? Em không bán gì cả, chỉ muốn nghe thật.
The last line matters more than the rest of it combined. The moment someone believes you are not selling, they will tell you things they would never put in a survey.
Set your expectations correctly. Your first users arrive slowly. Out of a hundred messages, you might get one person who becomes a user. That number sounds brutal until you remember what you are actually buying at this stage. You are not buying installs. You are buying understanding, and one honest forty-minute conversation is worth more than fifty installs from people you know nothing about.
Your First Users Should Have Jobs, Not Followers
When you pick who to message, filter for occupation before you filter for follower count.
A student with 50,000 followers is a great creator and a poor user, because students receive money from home rather than send it. The people who send money regularly are the ones with steady income and family waiting on the other end: nail technicians, restaurant staff, care workers, factory workers, delivery drivers, engineers, nurses.
Their jobs also tell you their constraints. Someone finishing a nail salon shift at 9pm will not do KYC during business hours. Someone on a factory rota gets paid on a fixed day, which means their transfer happens on a predictable day too. That is a scheduling insight you cannot get from analytics you do not have yet.
Start with ten people. Not a thousand. Ten real conversations will teach you more than any dashboard, and ten is small enough that you will actually finish.
The Download Is Not the Point. The Conversation Is.
Once someone replies, you are not there to close them. You are there to understand them.
These are the questions I keep coming back to:
- Do you send money home? How often, roughly?
- How much does a typical transfer look like for you?
- What do you use right now to send it?
- How did you first find that service?
- What made you pick it instead of the others?
- Who receives the money, and how do they collect it?
- Has anything ever gone wrong? What happened?
- What would have to be true for you to switch to something else?
The fourth and fifth questions are where the real answers live. People rarely choose a service for the reason the service thinks. They chose it because a cousin recommended it, or because it was the only one their mother’s bank branch recognised, or because a friend’s transfer failed once on a competitor and the story spread through the whole community.
None of that shows up in a feature comparison. All of it decides your market.
Walk Them Through the First Transfer Yourself
When someone agrees to try the app, do not send a link and disappear. Stay with them.
Message them while they sign up. Ask how it is going. Be there when they hit the recipient details screen. This feels wildly inefficient and it is exactly the right thing to do at this stage, because every hesitation they have is a hesitation ten thousand future users will have with nobody there to help.
Keep a running document. Every time someone pauses, asks a question, or gets confused, write down the exact screen and the exact words they used. After ten users you will have a list of friction points ranked by how often they came up, built from real behaviour rather than opinion. That list is worth more to your product team than any competitor analysis, and you built it by talking to ten people.
Then Do It Again in a Different Country
Here is the mistake I want you to avoid. Do not treat “the Vietnamese diaspora” as one audience. It is not.
Vietnamese people in the UK, Germany, the Czech Republic, Japan and Korea live different lives. Different visa situations, different jobs, different banking habits, different messaging apps, different levels of trust in financial services, different reasons for sending money home and different amounts. Filipino workers in the same city as Vietnamese workers behave differently again, with their own norms around who sends, how often, and what the money is for.
So run the same process per country. Ten conversations in the UK. Ten in Germany. Ten in Japan. Get as granular as you can stand, because the granularity is the whole asset. Generalising here is how you end up writing one message for everyone that lands with no one.
What You Actually Have After a Month
Your first users will not fill a dashboard. Your install numbers will be small enough to count on your hands. That is fine, because you walked away with four things:
- A small group of real users you can message directly, forever
- An insight bank broken down by country, job and sending behaviour
- A ranked list of product friction built from watching people, not guessing
- Their actual vocabulary, which becomes your ad copy, your app store description and your onboarding text
That last one is quietly the most valuable. When you eventually run paid ads, you will not be writing copy. You will be repeating back sentences your users already said to you, which is a completely different thing and performs like it.
Coming in Part 2
Collecting insight is the easy half. The harder question is what you do with it once the notes pile up: how you turn thirty scattered conversations into a segment, a message, a first campaign and a repeatable loop that grows without you personally messaging every single person.
That is the next post. Follow along and I will walk through exactly how I turned these conversations into the first real growth engine, including the parts that did not work.