Kamal Noori
Scaling Internationally Without Breaking the System
The brand that opens five countries at once almost always ends up running zero well.
Scaling internationally isn’t about how many markets you can launch. It’s about how many you can keep healthy at the same time. Those are very different numbers, and the gap between them is where most expansions break, spread too thin, every market mediocre, fires in three languages, and nobody who can put any of them out.
This is the last post in the series, and the discipline it asks for is the same one from the first: patience as a system, not a personality trait.
When to scale: prove, then pour
Don’t scale a market because it’s new and exciting. Scale it because it’s earned it. A market is ready for more budget when:
- New-customer payback works at the current spend. Not blended, new customers, paying back inside a window you can fund. If a market isn’t profitable on new customers small, it won’t become profitable big. Volume doesn’t fix bad unit economics, it multiplies them.
- Per-market MER is holding as you nudge spend up. If MER drops as you add budget, you’ve found the ceiling for now, that’s the auction telling you efficient demand is tapped.
- Operations aren’t strained. Fulfillment, returns, and support in that market are smooth. Scaling spend scales orders, which scales the ops load. Scale into a strained operation and you scale complaints and refunds.
The signal isn’t a good day. It’s a stable, profitable few weeks on the metrics that come from your bank, not the platform.
Read the signals, not the noise
A single great day is noise. A platform’s reported ROAS spiking is noise. Real signals are slower and come from the till: per-market MER, new-customer CAC versus margin and LTV, return rate, repeat-purchase rate. Watch those over weeks. (Why platform numbers mislead, especially abroad: Tracking Across Markets and ROAS Is Lying to You.)
When you do scale, scale in steps, not leaps. Raise budget gradually so campaigns don’t get knocked back into the learning phase, and so you can watch whether efficiency holds at each level. A big overnight jump resets learning and teaches you nothing clean.
One market before the next
Here’s the rule that prevents most disasters: stabilize one market before opening the next.
Get country one profitable, smooth, and boring. Document what worked, the structure, the creative that traveled, the local quirks. Then open country two, applying the playbook. Each market you stabilize makes the next one faster and cheaper, because you’re replicating a proven system instead of improvising five at once.
Open them all simultaneously and you can’t tell what’s working, you can’t give any of them real attention, and when one breaks you don’t have the bandwidth or the playbook to fix it. You’ve turned one solvable problem into five unsolvable ones.
The psychology: patience is the actual edge
There’s a tempting story in expansion: “Move fast, plant flags everywhere, win on speed.” It’s almost always wrong. Speed without stability isn’t growth, it’s spreading the same effort thinner until none of it is enough to matter anywhere.
The operators who win internationally are patient on purpose. They treat each market as a system to be stabilized, not a flag to be planted. They resist the pull to open the next country until the current one is genuinely healthy, because they know an unstable market doesn’t get better with more budget, it gets more expensively unstable.
You are not exporting flags. You’re exporting a working machine, one clone at a time, and only when the last clone is running clean. That patience feels slow. It’s the fastest sustainable path there is.
What to do next
- Scale on new-customer payback and per-market MER, held over weeks, not days.
- Raise budgets in steps, don’t reset the learning phase with leaps.
- Stabilize one market fully before opening the next. Document the playbook each time.
- Protect operations, scaling spend scales the ops load too.
Build the machine, make it boring, clone it patiently. That’s the whole game, at home and abroad.