Sourcing from Vietnam: A Practical Guide

Vietnam is one of the strongest manufacturing bases in the world for a specific set of consumer products, and an awkward one for others. The thing that separates the two is not the assembly cost. It is how deep the domestic supply of materials and components runs behind whichever factory you are talking to — because that depth sets your lead time, your minimum order, and whether your goods qualify for preferential tariff treatment when they reach the EU.

A country guide invites confident-sounding claims about wages, capacity, throughput and lead times. This one stays with what can be checked against a primary source, and where a useful figure does not exist in one it describes the mechanism and names what sets the number instead. Every figure here comes from a source listed at the end.

What Vietnam is genuinely strong at

The most honest evidence of what a country makes at scale is what it ships, and Vietnam's General Department of Customs publishes that monthly by commodity. These are the preliminary totals for January to November 2025.

An export figure measures what is made and shipped, overwhelmingly for very large accounts. It does not tell you a factory in that line will take your order, or at what quantity. What it tells you reliably is whether the industry exists and at what scale — the first question, and the one most often answered from impression.

Who owns the factories, and why it changes your reception

The same customs release splits each export line by foreign-invested enterprises, and that split predicts the kind of company on the other end of the email.

The percentages are ours, calculated from the two customs tables cited at the end. A high foreign-invested share tends to mean large plants built by regional manufacturing groups for very large accounts, with minimums and onboarding to match. A lower share means more domestically owned firms of varied size — usually where a mid-sized order is welcome rather than tolerated.

Read practically: in footwear you are mostly approaching plants built around a handful of global brands, where a first order in the low thousands of pairs is a small conversation. In wood and furniture the ownership picture is far more mixed and the company sizes wider. Garments sit between, which is one reason apparel is where a mid-sized brand most often finds a workable partner here.

Category fit: the economics are not the same in every category

Treating a country as uniformly good or bad is the most expensive simplification in sourcing. Vietnam is not one industrial base but several, at very different depths, and a decision obviously right for a hoodie can be obviously wrong for a jewellery line. The table reads each category through what the industrial base actually supports.

This is a judgement about industrial structure, not a ranking and not a quotation. Its evidence is the export composition above plus how far upstream each category reaches for inputs. Individual factories beat generalisations — the point is to show where you will have to look harder, not to tell you no.

Where Vietnam is not the obvious base

Four structural conditions make Vietnam a poor first choice, and none is about the cost of labour or the competence of the factories.

• Your product is mostly bought-in components from many small vendors. The value of a dense component ecosystem is not the price of any one part; it is changing a part in a week. Where that density is thinner, every substitution becomes an import.
• You need a very low first order. A supplier base weighted towards large export plants and a materials base weighted towards imported inputs both push minimums up, in the same direction.
• You need fast tooling iteration. Where a mould has to be cut, tested, corrected and recut several times, you are buying access to a tooling cluster rather than to a factory.
• Your product needs a documented input chain you do not control. Where the substantive risk sits in ingredients, formulation or a legality dossier rather than in making the thing, choose a base where that chain is domestic and auditable.

The supplier ecosystem, and why upstream is the whole difference

Look again at three lines in the export table. Textiles and garments: 35.9 billion. Yarn: 3.9 billion. Textile, leather and footwear materials and auxiliaries: 2.2 billion. Finished goods leave in enormous volume; the materials that go into them are, as an export line, a rounding error beside them. A country can be a world-class converter of materials into product without producing the materials, and Vietnam is largely the former. That does not make a Vietnamese garment or shoe worse — the export figures argue the opposite — but a meaningful part of its material content typically arrives from elsewhere before production starts. Three things follow, and a first-time buyer usually discovers them in the wrong order.

1. Your lead time may contain an international leg before cutting begins — not a customs formality but an actual shipment of material, with its own booking, transit and clearance.
2. Where the material comes from decides your tariff treatment, because the EVFTA origin rules for many products turn on where the material was made, not only where the product was assembled.
3. Changing a material late is harder and dearer than with a domestic mill down the road, because the substitution is an import decision rather than a purchasing one.

MOQs and what actually sets them

There is no honest country-level MOQ figure, in Vietnam or anywhere, because a minimum order is not a national characteristic — it is three separate calculations stacked on top of each other.

• The factory's line economics. Setting up a line, training operators on a new style and running the first pieces slowly cost the same at 300 units as at 3,000. The minimum is where that setup is worth absorbing, and it rises with complexity and with how large the plant's usual accounts are.
• The material lot economics. A dye lot, a weaving run, a mould run each have a smallest quantity worth setting up. This one is set by the material supplier, not the factory, and is very often the higher of the two.
• The shipping unit, where material is imported. Below some quantity, freight per unit becomes absurd. This third minimum exists only where the upstream is not domestic — which, per the section above, is common here.

The method is the same everywhere: get all three numbers, convert them into finished units, plan against the largest. The apparel sourcing guide works this arithmetic through for fabric and garments, and the logic transfers to a moulded part or a coated textile. What moves the answer down here specifically: a stock material the supplier already holds, a domestically produced material, a domestically owned factory rather than a large export plant, and fewer colours and sizes rather than a lower total quantity. What moves it up: a custom material, a custom component, a custom tool, a long variant list.

Lead times, and the leg most calendars miss

A Vietnamese factory quoting a production window is quoting the part of the calendar it controls. The parts it does not control are where delivery dates actually move, and in Vietnam the largest of those is material.

1. Specification and quotation

   Set by how complete your specification is. An incomplete pack is a slow start disguised as a fast one: every open question returns as a sample round.

2. Sampling and approval

   Set by the number of rounds and courier time both ways. On a multi-round approval, shipping samples often takes longer than making them.

3. Material sourcing and, where applicable, material import

   Set by whether the material is domestic or imported, stock or made to order, and by the material supplier's own order book. The leg most first calendars omit.

4. Material arrival and inspection at the factory

   Set by transit, clearance and incoming inspection. Nothing downstream can start early to compensate for this being late.

5. Production

   Set by quantity, complexity and the capacity actually allocated to you — a different number from the capacity the plant has.

6. Inspection, packing and freight booking

   Set by your sampling plan, by what happens if an inspection fails, and by sailings from the port serving that factory.

One national factor is worth planning around explicitly: the Tết Lunar New Year holiday interrupts production and shipping, with a rush before it and a slow restart after. Its dates move every year and are set by official public-holiday announcements, so check them for your production year and treat the weeks either side as reduced capacity.

Wages, and what a minimum wage does and does not tell you

Vietnam sets statutory minimum wages by region. The current figures come from Decree No. 293/2025/NĐ-CP, issued on 10 November 2025 and effective from 1 January 2026, which replaced Decree No. 74/2024/NĐ-CP and raised the minimums by an average of 7.2%.

Which region applies is set by the locality in which the employer operates, with branches following their own locality. Where an industrial park, export processing zone, high-tech zone or concentrated digital technology zone spans localities in different regions, the highest applicable region governs.

The part that matters more than the numbers: a statutory minimum is a legal floor, not the wage being paid. Skilled operators in an export plant are generally paid above it, often on incentive or piece-rate systems, and total employment cost includes statutory insurance contributions, overtime premiums and allowances the headline does not show. A wage comparison between countries is also not a cost comparison. Labour is one line in a cost sheet and in most consumer goods it is not the largest — material usually is — so a country with a lower minimum and a thinner materials base can easily deliver a higher landed cost than one with a higher minimum and a domestic mill.

EVFTA: preferential tariff treatment is conditional, not automatic

The EU–Vietnam Free Trade Agreement entered into force on 1 August 2020. The European Commission describes it as eliminating 99% of all tariffs, and reports Vietnam as the EU's largest goods trading partner within ASEAN and its fourteenth largest overall. That headline is real. It is also not a rate you can put in a cost sheet, for one reason: a preference applies to originating goods, claimed with a valid proof of origin, and nothing about that is automatic.

A product originates in Vietnam on one of three bases — wholly obtained there, produced exclusively from materials that already originate there, or produced from non-originating materials while satisfying the product-specific rule for its tariff heading. Most manufactured consumer goods are in the third case, so everything depends on the rule written for your heading. Those rules were updated to the HS 2022 nomenclature by Decision No 2/2024 of the EU–Vietnam Trade Committee, published in OJ L 838 of 12 March 2024 and applying from 1 January 2024, so a rule copied from an older annex may no longer be your rule.

For textiles and clothing the rules generally require a double transformation — two manufacturing steps in Vietnam — so sewing a garment from imported fabric is not enough: the fabric has to be made from yarn in Vietnam, or the equivalent step for the product. Printed fabrics have their own printing rule. This is where the materials asymmetry above turns into a tariff question rather than a logistics one. Two mechanisms can soften it. Bilateral cumulation lets EU-originating materials count as Vietnamese, and fabric originating in Korea may also be used, applicable from 23 December 2020; the other cumulation types the agreement contemplates are not in effect, and its ASEAN provision covers squid and octopus, not textiles. A tolerance rule allows a limited value of non-originating material — 10% of the ex-works price for industrial products other than textiles and clothing, while HS chapters 50 to 63 are governed by textile-specific notes that work differently. Tolerance never rescues a wholly-obtained requirement.

The proof, and the timing

Origin on a Vietnam-to-EU flow is proved either by a EUR.1 movement certificate issued by Vietnam's Ministry of Industry and Trade, or by an origin declaration made out by any exporter for consignments below EUR 6,000; the registered-exporter route, where an approved exporter self-certifies without a value limit, requires Vietnam to have notified such a system to the EU, and it has not. Vietnamese EUR.1s carry a manual stamp and signature — there is no electronic certificate. Timing runs asymmetrically: the Union Customs Code lets a preference be claimed after release on imports into the EU, but Vietnam does not allow a claim after importation, so on a flow into Vietnam the paperwork has to be right at the moment of declaration. Goods must also satisfy the non-alteration rule — nothing done to them in transit beyond preserving them.

• The order of work on an origin question
• Classify the finished product — the heading selects the rule, so a classification error invalidates everything above it.
• Read that rule in the current annex, HS 2022 version, literally, including any note it points to.
• Evidence where each material the rule cares about is produced. A verbal assurance is not evidence.
• Agree who issues the EUR.1 or declaration before shipment, and cost the order at both the preferential and the normal rate.

None of this is a route around anything. If the rule is not met the goods pay the normal rate — a costing input to plan for, and the only correct responses are to change where the material is made, change the product, or accept the rate.

Ports and logistics

Vietnam has two deep-water gateway systems at opposite ends of a long country: Haiphong in the north, serving Hanoi and the northern industrial provinces, and the Cai Mep area in the south, serving Ho Chi Minh City and the southern manufacturing belt. Which one your goods leave from is decided by where the factory is, and the inland leg between the two is long enough to be a real cost and calendar item.

On measured performance, both do well. The World Bank and S&P Global Container Port Performance Index 2025 — sixth edition, measuring vessel time in port — places Cai Mep 11th in the world with a score of 122 and Haiphong 13th, also on 122, in its top-20 table. Haiphong also appears among the twenty most-improved ports between 2020 and 2025, rising from 70 to 122.

• Establish which port serves your factory and how long the inland leg is before comparing freight quotes — a rate to one gateway is not comparable to a rate to the other.
• If you are consolidating from more than one supplier, check they are in the same region. A north–south consolidation adds a domestic leg no ocean quote includes.
• Agree the Incoterm, and so who books, who pays origin charges and who carries risk at each handover.
• Confirm who produces which document and when — invoice, packing list, bill of lading, origin proof. A late origin document is a delayed clearance.

Validating a Vietnamese supplier

The discipline is the same as anywhere. You are establishing that the entity you contract with runs the plant, that the plant can make your product at your quantity in your season, and that whatever leaves the site is named.

• What to establish before you place a first order
• The enterprise registration certificate and enterprise code, and that the name on it matches the party you will contract and pay.
• For a foreign-invested plant, the investment registration certificate, and which group actually controls the site.
• That the address on the documents is the factory you visited or had audited — not a trading office.
• Export history in your specific product, not in the category. Capability in one construction is not capability in another.
• Capacity in your production window as lines or hours allocated to you, rather than total plant capacity.
• Which operations are subcontracted off-site — printing, embroidery, washing, plating, moulding — with the subcontractor named.
• Where each significant material comes from, in writing, with country of production.
• Audit history with report date and scope, and any product testing held, with its scope and expiry.
• A named merchandiser or account contact, and what happens when that person is on leave.

The factory vetting guide sets out the full process, and how to vet an overseas manufacturer covers what each document proves. On the audit line: amfori, which runs the widely used BSCI programme, states plainly that BSCI is not a certification and that an audit is a point-in-time snapshot which cannot guarantee compliance with laws or standards. Treating a grade as a warranty is a category error.

Quality and compliance: two country-specific wrinkles

Quality management here is the same discipline as anywhere — a written standard, a sealed reference sample, a stated inspection plan and someone on site — and the manufacturing quality control guide covers the programme. Two things are worth handling differently, and both come back to imported materials. First, incoming material inspection carries more weight: where the cloth, hardware or component arrives from another country, the factory did not make it and cannot fix it, so the check on arrival is the last cheap moment to catch a problem. Second, name the off-site operations — printing, washing, plating and moulding are commonly subcontracted, and a plan covering only the assembly site does not cover where the defect will happen.

On product compliance, keep the categories apart. EU product-safety, substance-restriction and labelling obligations attach to you as the importer placing goods on the market, and no supplier certificate or audit grade discharges them — see manufacturing compliance for the EU market. A golden sample settles what "correct" means before the argument rather than during it, which matters more, not less, at this distance.

Working practice: what actually decides whether an order runs smoothly

The strongest predictor of a smooth Vietnamese order is not the country, the city or the plant size. It is the merchandiser assigned to you — the person who turns your specification into instructions on the floor, chases the material supplier, and tells you early when something has slipped. A good one makes a mid-sized factory feel excellent; a poor one makes a world-class factory feel unresponsive.

• Write decisions down and number them. A decision agreed in a call and never written into the specification will not survive into the second order.
• Number your sample rounds and state what each one approves. "Approved" without a scope is the most expensive word in sourcing.
• Ask for the material supplier's name and the material's status rather than for reassurance about the schedule. Reassurance is not information.
• Expect English fluency to vary by role — strong in merchandising, limited on the floor. Specify visually: dimensioned drawings, marked photographs and physical references beat prose.
• Agree a single escalation contact, and account for the time difference when you set expectations about response speed.
• Visit, or have someone visit for you, before the first large order. No document set substitutes for seeing the floor.

Landed cost, and the number that actually matters

The FOB price out of Vietnam is not what the product costs you. Freight, origin and destination charges, duty, clearance, insurance and inland delivery sit on top, and they land very differently on a light garment than on a piece of furniture. The duty line in particular is not a fixed property of the product — it turns on classification and on whether you can document origin, which is why the origin work above belongs in the costing exercise rather than in the shipping paperwork. Two habits keep this honest. Build the landed cost per unit before comparing suppliers, because a lower FOB from a factory further from a gateway, or one whose material chain will not satisfy an origin rule, can be the more expensive option. And treat a preferential rate you cannot yet document as unavailable. The landed cost guide sets out the full build-up.

When Vietnam is right, and when it is not

• Vietnam is likely a strong choice when
• Your product sits in a category with a large export base — garments, footwear, wooden furniture and home goods, bags and travel goods, sports and outdoor.
• The main value added is skilled conversion of material into product, not assembly of many bought-in components.
• Your materials are domestic, held as stock, or stable enough that an import leg can be planned rather than reacted to.
• You can meet the origin rule for your heading, or you have costed the order without assuming the preference.
• Your quantities suit the plant you are approaching, and you have compared its minimum against the material minimum.
• Your factory is near the gateway that serves it, and consolidation happens in the same region.

• Look elsewhere first where any of the four conditions above applies — a bought-in-parts product, a very low first order, fast tooling iteration, or regulated risk in an imported input chain.
• Do not decide on a country at all until you have a specification good enough to quote from. A comparison run on a vague brief compares guesses, and the guesses differ more than the countries do.

How Library of Trade approaches a Vietnam brief

The work runs in the same order as anywhere: turn the brief into a specification and a bill of materials, source materials and upstream suppliers against it, match factories on capability, capacity, price, quality and compliance, develop and approve samples, then manage production, quality and freight through to delivery. What changes by country is where the difficulty concentrates, and in Vietnam it concentrates upstream — in materials and their origin, which is where the early work goes.

You contract with your suppliers directly; this is not a marketplace or a directory, and the relationships stay yours. Software structures the specification, compares options across suppliers and holds one version of what was approved. The judgements — whether a capacity claim survives your season, whether a material chain will satisfy an origin rule, whether a cost sheet is honest — are made by people.