What to settle before your first campaign in Vietnam

Most Vietnam entry plans start with the media plan. Which platforms, what budget, which agency, what the first quarter looks like.

That work matters, but it is not where launches go wrong. They go wrong in four decisions that are made quietly or not at all, usually because they look administrative next to a media plan. Each one is cheap to settle before launch and expensive to revisit after six months of data has been collected under the wrong assumption.

1. What counts as revenue

Cash on delivery is ordinary in Vietnam, not a fringe method. A customer confirms an order, a parcel goes out, and the transaction completes only when someone accepts it at the door.

So a confirmed order and money received are two different events, separated by a rate that varies by category, by price point, and by how the customer was acquired. A buyer who arrived through a deep discount refuses delivery more often than one who arrived through search.

Decide now which of the two the business runs on, write it into the reporting template, and require every party to use it. Deciding later means recalculating every efficiency figure you have, and discovering that some channel comparisons reverse.

Either answer can be defended. Running on confirmed orders is simpler and gives the operating team a faster feedback loop; running on accepted deliveries is the figure the business actually banks. What cannot be defended is having both in circulation, with the agency reporting one and finance reporting the other, and nobody reconciling them until the first bad quarter.

2. Who owns the advertising accounts

The accounts should be registered to your company, with agencies granted access.

This is not about trust in any particular partner. Historical performance data and audience lists gain value with time, and that value only exists if it can move when the relationship does. In Vietnam, accounts held in an agency’s name are common enough that nobody will flag it as unusual — which is precisely why it needs to be raised in week one rather than discovered in month twelve.

3. Whether marketplaces are a channel or the channel

For many consumer categories in Vietnam, marketplaces are not one channel among several. They are where discovery and comparison happen, and a brand site plays a supporting role rather than a central one.

This changes the shape of the plan rather than just the budget split:

Getting this wrong is not a tuning error. It means the whole plan is aimed at the wrong step of the buying journey.

4. How sales that close in chat get recorded

A significant share of Vietnamese commerce is agreed in Zalo or Messenger. The customer sees something, asks a question, negotiates, and confirms — inside a conversation.

Standard tracking sees the click that opened the conversation and nothing after it. The predictable consequence is that channels whose job is to start conversations look weak, channels that close on a website look strong, and budget moves toward the second group for reasons that are not real.

Connecting the chat platform to the order record is not difficult, but it has to be someone’s job, named before launch. After launch it competes with everything else.

What it involves, concretely: a source tag attached to the conversation when it opens, and an outcome recorded against that conversation when it closes — sold, not sold, and briefly why. No tool supplies an outcome nobody entered, which is why this is a staffing decision rather than a software one.

5. Which operating structure the plan assumes

The four decisions above all sit inside marketing. This one sits outside it and changes all four, which is why it belongs in the same conversation.

Teams often treat this as a corporate matter to be resolved in parallel, and it is — but the marketing plan cannot be finalised without knowing the answer. A plan built for a local entity and executed cross-border will miss on the one number that matters most in the first year, because a longer delivery promise moves the acceptance rate, and the acceptance rate sits underneath every efficiency figure in the deck.

The distributor row carries a specific and easily missed cost: when the distributor holds the customer relationship, the first year produces revenue but very little of the data that the second year’s plan would be built from. That can be the right trade. It should be a stated trade, with a review date, not a default.

Three assumptions carried in from the last market

Entry plans are rarely wrong because of ignorance. They are wrong because something true elsewhere was carried across without being re-tested.

“Customers will search for us.” For a large part of consumer goods here, discovery happens in feeds, in live selling and inside marketplace recommendation surfaces. A plan weighted toward capturing existing demand finds less of it than forecast, and the usual conclusion drawn is that the category is smaller than it is — which channels matter treats this at length.

“Email will carry the follow-up.” The holding pattern between interest and purchase runs through messaging rather than email. A nurture programme designed around an email sequence will show open rates that look survivable and a conversion contribution close to zero.

“One agency covers the market.” Marketplace operations, creator programmes and live selling are separate disciplines from platform media buying. A single contract covering all of it either leaves parts unserved or subcontracts them invisibly — the questions that expose which is happening are in choosing and overseeing an agency.

Each of these is testable in the first quarter for a small fraction of the media budget. None of them is testable at all once the whole plan has been built on top of it.

The four numbers a board should see in year one

Entry reporting tends to grow into a deck. Four lines cover what a board can actually act on, and every one of them depends on the decisions above having been made:

• Accepted orders, not confirmed orders. The revenue definition from decision one, applied consistently.
• Cost per accepted order, by acquisition source. Not cost per click, and not a company average — the average hides the source producing orders that are never kept.
• Share of orders that began in a conversation. This is the number that tells you whether the reporting is missing a channel.
• Contribution after acquisition cost, with the marketplace fee stack subtracted, since fees and promotions reshape it every month.

What should not be in year-one reporting is a target return copied from another market. The break-even threshold is a function of local margin after local fees, and it has to be calculated here before it can be met here.

What each decision costs to reverse

The argument for settling these before launch is not tidiness. It is that the cost of reversing them rises sharply with the amount of data collected under the old assumption.

None of the four is irreversible. All four are cheaper now, and the first column is measured in hours.

Why these four get skipped

It is worth asking why decisions this cheap go unmade, because the answer points at the fix.

None of them belongs cleanly to anyone. The revenue definition sits between finance and marketing. Account ownership looks like procurement. The role of marketplaces reads as commercial strategy. Chat-to-order tracking looks like an IT task, and is actually a sales-operations one. In an entry team assembled from a regional marketing lead, a corporate development manager, a finance controller and a first local hire, each of the four falls just outside somebody’s remit — and things that fall just outside four remits do not get decided, they get assumed.

The second reason is that all four are invisible in the artefact everyone is reviewing. A media plan has slides. A revenue definition is a sentence, and a sentence does not survive a review process built around decks.

The fix is unglamorous and works: a single page listing the four decisions, each with a named owner and a date, reviewed at the same meeting as the media plan. Not a workstream, not a project — one page, five minutes on the agenda, before the budget is approved.

One more point about sequence. These decisions have a natural order, and it is not the order of urgency. The revenue definition comes first because everything else is measured against it. Account ownership comes next because it becomes harder every week. The marketplace question comes third because it determines the shape of the plan. Chat tracking comes last of the four, and still before the first campaign goes live.

The sequence that works

• Settle the revenue definition. One sentence, in the reporting template.
• Register the accounts to your company.
• Decide the role of marketplaces for your category, and structure the plan around that answer.
• Assign ownership of chat-to-order tracking to a named person.
• Then book the media.

Each step also has a plain test of whether it is really done. The revenue definition is done when finance and marketing quote the same number for last month. Account ownership is done when someone at head office can log in and read the administrator list. The marketplace role is done when the budget split in the media plan can be explained from it. Chat tracking is done when an order closed in chat can be traced to its source in the order record.

None of the first four requires a large budget or a long project. They require a decision, taken in the right order.

Settling them is the first half of a market entry; sizing the opportunity they apply to is the other half. Where the category is genuinely large enough, and who is buying it, comes out of the research phase of Go-to-Market Strategy. Turning that into a sequence — which channels, in what order, at what budget, and what has to be true before scaling — is its second phase.

Where MWY sits in this

MWY does not run campaigns, does not sell media, and takes no commission from any platform or agency. The work is setting these decisions up correctly before spending starts, then reading the numbers back each month on the company’s side of the table.

What MWY reconstructs is profit after advertising and acquisition cost — recorded revenue, less media spend, less the acquisition costs sitting outside the ad accounts. True profit requires cost of goods and inventory, which sit outside independent marketing oversight. Where a margin figure appears in any MWY analysis, it is one the company supplied.

Once spending is live, the continuing version of this work — monthly reading, challenging the plans, and reporting to the head office in English — is Vietnam Marketing Advisory & Oversight.

What MWY will not do is forecast a return before the measurement exists to check it against. A projected return built on another market’s assumptions is not a forecast; it is a number that will be quoted back at everyone for a year.