Vietnam, Cambodia and Thailand on One eSIM: Routes, Borders and Costs

A three-country route through Vietnam, Cambodia and Thailand breaks the assumption that one plan covers one trip. The data you bought for Hanoi does not follow you over the Moc Bai border, and the card you buy in Phnom Penh does not reach Bangkok. Most travelers discover this at the worst possible moment: a border town, at dusk, with a bus waiting.

This guide covers the crossing itself rather than the sightseeing. Three borders, three registration regimes, and the arithmetic that decides whether one regional plan beats three local cards on a 21-day route. It also covers the two failure modes that catch people: a regional plan that throttles halfway through, and a phone that cannot hold two profiles without losing the verification code sent to the first one.

The short answer: if you cross borders more than once, a regional Asia plan wins on convenience and usually on price. If you spend two weeks in one country and three days in another, buy locally and accept the ten minutes at each counter. The break-even point is at two crossings, and the rest of this page is about where that arithmetic comes from.

Three borders, three sets of SIM rules, one phone

The three countries do not share a registration system, and that is the first thing that makes a multi-country plan attractive.

Read the middle column as time you spend three times on a three-country trip, and the last column as value you cannot carry across a border. A Thai tourist SIM’s number works in Thailand; a Vietnamese one works in Vietnam; neither follows you.

A regional plan replaces all three purchases with one profile and one price. What it does not replace is the number: a regional eSIM is data-only, so if you need a local number in any of the three countries, you are buying a local card anyway. That is the honest boundary, and the Vietnam number guide goes through what a number is and is not good for.

Moc Bai to Bavet: what happens to your data at the crossing

The bus from Ho Chi Minh City to Phnom Penh takes about six hours and crosses at Moc Bai on the Vietnamese side and Bavet on the Cambodian side. It is the most common land route between the two countries and the one worth understanding in detail, because the other crossings behave the same way.

Here is what happens to your connection, in order:

1. The last Vietnamese tower. Signal holds until the border zone and then drops as you pass through. If your plan is Vietnam-only, this is where it ends.
2. The gap. Between the two checkpoints there is no network you can use, on any plan. It lasts a few minutes, or longer if the queue is bad.
3. The Cambodian tower. A regional plan reattaches automatically, usually within a minute once you clear immigration. A Vietnam-only plan shows “no service” and stays that way.
4. The first useful checkpoint. Money, a local SIM if you want one, and a working connection if you have a regional plan. This is the ten minutes where the difference between the two approaches is most visible.

Two practical notes. Vietnamese and Cambodian immigration both want the passport in hand rather than in a bag, so keep it accessible rather than packed. And a border town is the worst place to troubleshoot: the Wi-Fi at the roadside cafe usually works, and the staff usually know which network to point you at, but you are solving a problem you could have solved at home.

The other land crossings behave similarly. Ha Tien to Prek Chak, on the coast route toward Kep and Kampot, is smaller and faster. The river route from Chau Doc to Phnom Penh is the pleasant one and the slowest, with no usable signal for part of the journey. None of them is difficult; all of them reward having the data question already settled.

The Asia regional plan covers all three, and this is its catch

A regional Asia plan is a single profile that attaches to a partner network in each country. The one sold on this site lists 26 countries including Vietnam, Thailand, Cambodia and Laos, which covers every route in this guide plus the detours people actually take.

Three things it does well. It removes the border decision entirely, because the profile reattaches on its own. It removes the registration queue, because a roaming profile is not a local subscription. And it prices per trip rather than per country, which is the whole point on a multi-border route.

Three things it does less well, and they are the reason this is not an unconditional recommendation.

It is data-only. No local number in any of the three countries. If a guesthouse in Cambodia wants to call you, they cannot.

Routing can add latency. Many regional profiles route traffic through a hub outside the region, which adds a round trip. On a video call to Europe you will not notice. On a low-latency game you will.

The partner network is not always the best local one. In each country your profile attaches to whichever operator the seller has an agreement with. That is usually the largest, but not always, and in rural areas the difference shows.

None of those three is a dealbreaker for a normal trip, and all three are worth knowing before you rely on the plan for something that matters. The comparison of Vietnam plans covers the same trade-offs for a single country, where the arithmetic is simpler.

Three local cards against one regional plan, costed for 21 days

This is the comparison that decides most trips, and it has a clear answer at the two-crossing mark.

Take a 21-day route: ten days in Vietnam, five in Cambodia, six in Thailand.

The third row is the one people default to and should not: buying three separate country eSIMs costs more than one regional plan in almost every case, because each purchase carries its own minimum and its own small-data premium. The regional plan exists precisely to avoid that.

The first row deserves respect rather than dismissal. Three local cards give you three numbers, three local-rate connections and usually more data per dollar than any regional plan. On a long trip with heavy use, that is a real advantage, and the cost is three queues and three registrations.

The deciding variable is not money. It is how many times you want to solve the same problem. At two crossings the regional plan wins on effort and roughly ties on price. At zero crossings, buy locally and read the Vietnam SIM card guide instead, because that is a different decision.

Cambodia’s networks are not Vietnam’s, and your card knows it

Cambodia runs on a different set of operators, and the practical consequence is that signal quality changes at the border even on a regional plan.

The three networks that matter are Smart, Cellcard and Metfone. Coverage in Phnom Penh and Siem Reap is good on all three. Coverage on the islands and in the northeast is variable, and coverage on the road between them is what you would expect from a country with long stretches of low population density. Cellcard and Metfone both publish coverage maps of their own, which is more use than any comparison written from a distance.

Two Cambodia-specific points worth planning around:

The temples are not a connectivity problem. Angkor is close enough to Siem Reap that the networks reach it, and the constraint there is the heat and your battery rather than the signal. Downloading an offline map is still the right move, because the app’s tile requests are what drain a phone in the sun.

The islands are a connectivity problem. Koh Rong and its neighbors have coverage in the developed parts and gaps elsewhere. If you are going for three days, treat the connection as optional and download what you need before the boat.

For a proper operator-by-operator breakdown, the Cambodia carrier guide covers which network holds signal where, and it is worth ten minutes before you cross rather than after.

Thailand wants your passport at Suvarnabhumi, and takes a photo

Thailand has the most procedural tourist SIM market of the three, and the process is worth knowing because it is the one that most often goes wrong at an airport counter.

Tourist SIMs are sold at the airport and in city shops, require a passport, and the counter photographs the buyer as part of registration. The package usually includes a Thai number and a fixed validity, commonly 15 or 30 days, with a data allowance that is generous by regional standards.

Three things to know before you join the queue:

• Airport counters are slower here than in Vietnam. The photo step adds a minute per customer, and the queue at a wide-body arrival bank is long. If your flight lands with three others, expect to wait.
• The validity cap is real. A 15-day package bought on day one of a three-week Thailand stay expires before you leave, and topping up a tourist package is not always possible.
• The 7-Eleven route exists but is not a shortcut. Some branches sell starter packs; activation and registration still happen, and the staff are not there to do it for you.

Package names and validity differ between the three Thai operators, and the airport counters quote whichever one they are paid to move. AIS and True publish their tourist packages on their own sites, and comparing two of those against the counter offer takes about three minutes on airport Wi-Fi.

For a route that ends in Thailand, the regional plan removes all of this. Whether that is worth it depends on whether you want a Thai number, which brings it back to the same question as Vietnam: a data plan is a data plan, and a number is a separate purchase.

A 21-day Hanoi to Bangkok route, day by day

An itinerary makes the data question concrete. This is a common route, and the notes are about connectivity rather than sightseeing.

Days 1–4, Hanoi. City coverage is dense on every network. The old quarter is loud, crowded and well covered. Data use is low: maps, Grab, messages, the occasional video call.

Days 5–7, Ninh Binh and the coast road south. Coverage holds along the main roads and thins on the smaller ones. Boat trips on the rivers lose signal for stretches, which is the point of the boat trip.

Days 8–10, Ho Chi Minh City. The densest coverage of the trip and the highest data use, because this is where people upload everything they have been sitting on.

Day 11, the Moc Bai crossing. Refer to the section above. This is the day the regional plan earns its price and the country plan stops working.

Days 12–14, Phnom Penh and Siem Reap. Good coverage in both cities. The road between them is long and the signal comes and goes.

Days 15–16, the crossing at Poipet. The bus from Siem Reap to Bangkok takes most of a day including the border, and the border itself is the busiest of the three crossings on this route. Expect queues, expect touts, and expect your regional plan to reattach on the Thai side within a minute.

Days 17–21, Bangkok. Coverage everywhere, high data use, and a phone that has now been running on a Vietnamese, a Cambodian and a Thai network in three weeks without a single settings change. That is the entire argument for the regional plan, expressed as an experience rather than a table.

Two profiles on one phone: switching without losing the code

A multi-country route usually means two profiles: the regional data plan and your home line, kept alive for bank codes. Getting the arrangement right before you fly prevents the two failures that ruin trips.

The setup, in order:

1. Install the regional profile at home and leave it switched off, set as the data line for later.
2. Turn off data roaming on your home line. Background data on a roaming home line bills at international rates even when the regional plan carries your traffic.
3. Turn on Wi-Fi calling on your home line if your carrier supports it in the region. Codes then arrive over the regional plan’s data connection at domestic rates.
4. Name both lines clearly. “Home” and “Asia” in the settings, so you never answer on the wrong one or send a message from the wrong number.
5. Screenshot the QR code and the order confirmation before you leave, and store them where you can reach them without data.

The two failures this prevents: a home line that quietly bills for background data, and a phone that switches its default data line at a border without telling you. The second one looks like a broken plan and is a settings problem, which is why the dual-SIM troubleshooting list starts with the data-line assignment rather than with the network.

A phone that holds only one active line changes this. Then the choice is between being reachable on your home number and having data, and data wins almost every time. Check messages during Wi-Fi moments and treat the code as a morning task.

Border towns, slow boats and the offline hour

Every land route on this itinerary has a stretch where no plan works, and planning for it is more useful than shopping for a better one.

The gaps are predictable: the zone between two checkpoints, the middle of a river crossing, the inside of a tunnel on the train, and the last hour of any road that climbs into hills. None of them lasts long, and all of them feel long if you are relying on a map.

What to have on the phone before each crossing:

• An offline map of the next two cities. Downloaded in advance, not fetched at the border.
• The address of your next accommodation in the local language, as text you can show a driver.
• A screenshot of the booking confirmation, because email will not load in the gap.
• Cash in both currencies. The border is where the ATM is broken and the card machine is “not working”.

That list costs nothing and covers the two hours a year when a good data plan does not help.

Islands and mountains: where the signal genuinely stops

Two kinds of destination on this route will not have the connection you expect, and neither is a reason to change plans.

The Cambodian islands. Koh Rong and the smaller islands have coverage near the main piers and resorts, and gaps on the beaches and the boat routes. Three days of patchy data is the normal experience, not a fault.

The northern Vietnamese mountains. The roads toward Sapa and the loops beyond it pass through valleys where coverage drops for minutes at a time. On a regional plan you are on whichever partner network the seller uses, which may not be the one with the best coverage in that specific valley. This is the one place on the route where a local Vietnamese card on the largest network genuinely beats a regional plan, and the Vietnam carrier guide maps which operator holds which province.

The honest framing: a data plan solves the cities, which is where you spend most of the trip, and does not solve the places you go to escape connectivity. That is usually the trade travelers want, and it is worth naming rather than pretending otherwise.

Night buses and the reunification line: eleven hours of nothing

The long-haul legs are where a data plan’s limits are most obvious, because there is nothing to attach to for hours at a time.

The sleeper buses. Vietnam’s open-tour buses and Cambodia’s overnight coaches run through areas with intermittent coverage. You will get signal in towns and lose it between them, which means a nine-hour leg is really four hours of connection in fragments.

The train. Vietnam’s north–south line covers roughly 1,700 kilometres between Hanoi and Ho Chi Minh City, and the full journey takes more than a day. Coverage follows the towns, which means long stretches of nothing in the middle sections. On the coastal stretch between Hue and Da Nang the run is short enough that the tunnels are the only real gaps.

The flights. If you fly instead, the connectivity problem disappears and the border problem becomes an airport problem: immigration, then a taxi, then a data connection that already works because you set it up at home. On a three-country trip with two flights, the regional plan is even easier to justify, because there is no counter to visit at all.

What to do with the offline hours: download a series, download maps, and treat the gap as the reason you took the night bus. A power bank matters more than a second data plan on these legs, because a phone searching for signal drains a battery faster than a phone watching video.

What changes if you fly between all three instead of crossing by land?

The flight version of this route is a different product decision, and it is worth separating because the arithmetic changes.

Flying Ho Chi Minh City to Siem Reap, then Siem Reap to Bangkok, removes the three border crossings and replaces them with two airports. Airports are easier in every way: immigration is a queue rather than a negotiation, the arrivals hall has Wi-Fi, and your data plan either works or it does not, with no middle ground to troubleshoot.

The cost is money, and the saving is time. A land crossing plus a bus is a day of your trip; a flight is three hours plus transfers. On a 21-day route the difference is roughly one day, which is worth a flight if the day matters.

On the data side, flying changes nothing about which plan to buy. What it changes is when you need it: at a land border the profile reattaches while you are standing in a queue, and at an airport it reattaches while you are walking to baggage claim. Neither takes a setting change on a regional plan, and neither happens at all on a country plan.

When the regional plan throttles on day nine

Any unlimited regional plan has a fair-use boundary, and on a 21-day route you will find it. What happens next depends on how you bought.

Three patterns, in order of how often they cause a problem:

A daily high-speed allowance. The most common structure. Each day resets and each day you get the same speed until the allowance runs out. On this route that means heavy days are throttled in the evening and the connection recovers overnight, which is manageable if you know it is coming.

A total high-speed allowance across the whole plan. Less common and more disruptive, because the plan does not reset. If the allowance is 20GB over 21 days, day nine is exactly when heavy users hit it.

A throttle that never lifts. Rare, and disclosed in the terms rather than the marketing. Read the terms.

The defense is the same one that works for a single country: buy based on your heaviest week rather than your average day, and know the number before you fly. If the plan throttles, how to tell a throttle from a coverage gap matters, because the two look identical on a phone and behave completely differently.

Buying at each border instead, and who that suits

There is a legitimate case for ignoring everything above and buying a card at each crossing, and it is worth stating fairly.

It suits three kinds of traveler. Someone staying two weeks in Vietnam and four days in Cambodia, where the second purchase is a small one and the effort is trivial. Someone who needs a local number in each country, for whom a regional plan is useless by definition. And someone on a slow trip with no fixed schedule, for whom the ten minutes at a counter is part of the experience rather than a cost.

It does not suit anyone crossing twice in three weeks, which is the majority of people searching for a three-country eSIM. For that traveler the regional plan is cheaper in effort and usually in money, and the only thing given up is the number.

If you do buy locally, buy on the operator that owns the ground you are standing on rather than the one with the best sign at the counter. In Cambodia that distinction matters more than in Vietnam, and the operator-by-operator network notes for Cambodia are the shorter read of the two.

If you want to check whether your handset can run two profiles at once, which is the setup both approaches need for keeping your home number alive, the compatibility check takes about a minute, and a short free trial proves it about your specific device rather than about the model in general.

Three weeks of this route in gigabytes, counted rather than guessed

Every plan decision above comes down to a number most travelers estimate badly. Here it is, in the order a normal day spends it.

Two rows decide the answer, and neither is the one people watch. Cloud backup is the biggest line on a normal day and it is entirely optional until you are on guesthouse Wi-Fi; switching automatic backup off for the trip removes a third of the total on its own. Streaming is the line that breaks unlimited plans, because an hour a night is 15 GB over three weeks before anything else happens.

The rows that never change are the ones that matter for navigation. Maps, ride-hailing and messages together come to under 3 GB for the whole trip, which is why a dead plan is an inconvenience and a dead map is a problem. If you keep backup manual, watch video on the room television rather than the phone, and put the phone in airplane mode on long bus legs, 10 GB genuinely covers 21 days.

Sizing follows from that: a plan with a single total allowance should be bought from the top two rows, not the median day, because the surplus does not carry. A plan with a daily reset can be sized from the daily column.

Three currencies, three payment apps, and the one card that keeps working

A working connection is half of being functional here. The other half is paying for things, and the tools that solve it are national rather than regional.

The pattern holds across all three: the wallet needs a local bank account, and the bank account needs residency or a long-stay visa. That rules the wallet route out for a three-week trip, which is why the practical answer everywhere is a card plus cash.

Four things that catch people out. Cambodian prices are in dollars and change comes back in riel, so a $3 coffee can return a handful of notes worth about $1.80 — spend the riel rather than carry it home. Small Vietnamese restaurants often add a few percent for cards, and some decline them below a threshold. Thai street food never takes a card, and neither do most markets in any of the three. And a card transaction that triggers a one-time code will fail if your home line has no signal, which is the single strongest argument for keeping the home number alive on Wi-Fi calling.

Withdrawing cash is where the fees hide, and they are charged per withdrawal rather than per amount. Take out larger sums less often, decline the machine’s offer to convert into your home currency, and treat the fee as the price of the withdrawal rather than a reason to walk to another machine.

Five ways the plan fails at a border, and the fix for each

Almost every “my eSIM died at the border” report on this route is one of five things. All five are visible on the phone within a minute of crossing.

The first row is the most common and the least alarming. A phone holding two profiles often leaves the data line on the home SIM after a network change, and the symptom is not “no service” but “everything is slow and nothing loads”, which people read as a broken plan and try to replace.

The fourth row is the one misdiagnosed as a faulty eSIM. Two active profiles on some handsets renegotiate repeatedly for the same network and produce a connection that works for a minute at a time. Switching one off entirely, rather than switching off only its data, ends it.

Row two is the only one you cannot fix in settings, and it is the failure this whole article exists to prevent. At any border, open settings before opening a wallet.

The Laos detour, and why a fourth country changes the plan

A good share of three-country routes quietly become four-country routes, because the Mekong makes it easy: the slow boat from Chiang Rai down to Luang Prabang, a bus from Hanoi, or the crossing from northern Cambodia at Stung Treng.

If Laos is on your route, the regional plan earns more than convenience. Lao tourist SIMs are cheap and registration is quick, but coverage outside Vientiane and Luang Prabang is thin, and the slow boat spends two days on a river with no usable signal at all — so the second purchase buys very little actual connection for the money.

The regional plan covers Laos because it sits on the same country list as the other three, and the practical difference is that the profile reattaches at the river rather than at a counter in a town you reach at dusk. Two caveats. The partner network in Laos is one operator rather than the best one in every province, and the northern mountains behave exactly like the Vietnamese ones, with gaps in the valleys that are not a fault.

In data terms Laos adds three to five days and almost nothing else, because there is nowhere to spend a gigabyte. That argues for sizing the plan from the rest of the trip and letting those days ride along.

Can two people share one regional plan, and should they?

Couples and pairs ask the same question two ways: can we share one plan, and should we.

Technically yes. Every phone on this route that runs eSIM also runs a personal hotspot, and a second device can tether to it. Practically, three things go wrong.

The battery. A phone acting as a hotspot runs hot and drains roughly twice as fast. On a temple day or a six-hour bus, the phone carrying the hotspot is the one that dies, and it is also the one holding the map.

The allowance. Hotspot traffic comes out of the same bucket as the host phone, so one person streaming at night spends the data the other needs at noon.

The split. Two people who separate for a morning — one to a market, one to a museum — leave one of them offline, which is precisely when a message matters.

The arrangement that holds up is two cheap profiles rather than one shared plan. Each phone stays independent, and the total is closer than it looks, because a second small plan usually costs less than the difference in allowance on a single bigger one. Sharing makes sense for a family with children, where the adults stay together and the children’s tablets should not each need a plan.

FAQ

Does one eSIM really work in all three countries? A regional Asia plan does, because it attaches to a partner network in each country. A country plan does not; it stops at the border. Check that the plan you are buying lists all three countries before you pay, because “Asia” plans vary in coverage.

Is a regional plan cheaper than three local cards? On a 21-day route with two crossings it usually ties on money and wins clearly on effort. For heavy data users staying long in each country, three local cards can be cheaper per gigabyte, at the cost of three registrations.

Will my home number still receive bank codes while I travel? Yes, if your home line stays active with data roaming off and Wi-Fi calling on. The codes arrive over the regional plan’s data connection at domestic rates, which is the arrangement most long-route travelers run.

What happens at the border if my plan does not cover the next country? The phone shows no service and stays that way until you buy a local card or a new plan. That is the failure the whole article is about avoiding, and it is why the crossing at Moc Bai is the moment to have already decided.

Do I need a Thai or Cambodian number for anything? Rarely. Grab works with a foreign number, hotels confirm by email or messaging apps, and the main case for a local number is a business that only phones. If that is not on your itinerary, a data plan covers the trip.

Can I buy the regional plan after I arrive instead of before? You can, over hotel Wi-Fi, in about six minutes. Doing it at home is better for one reason: a failed install with a week to spare is an inconvenience, and a failed install at a border is a day.

How much data does three weeks on this route actually need? About 10 GB if you keep cloud backup manual and watch video on guesthouse Wi-Fi. Add roughly 15 GB if you stream an hour a night, and add 4–13 GB if you let photos back up over mobile data. The navigation itself costs under 3 GB for the whole trip.

Does the regional plan cover Laos and Malaysia as well? Coverage varies by seller, so read the country list before paying rather than after. The plan sold here lists 26 countries including Laos, Thailand, Cambodia and Vietnam; a cheaper “Asia” plan with eight countries often covers only the first three or four.

Can two people travel on a single regional plan? Only over hotspot, and it is rarely worth it. The host phone drains about twice as fast, the second person eats from the same allowance, and one of you goes offline as soon as you separate. Two small profiles cost a little more and remove all three problems.