Robot adoption patterns differ between markets for structural reasons rather than cultural ones. Several features of the Vietnamese context shape where robots work here and where they do not, and they are worth understanding before planning a deployment.
A business mix weighted toward small operations
The most consequential factor.
What it means. A very large share of commercial activity happens in small premises with few staff — shops, cafés, family restaurants, service businesses.
Why it matters for robots. Small premises rarely have the distances that make delivery robots worthwhile, and small teams rarely have someone available to look after equipment. Both are structural obstacles rather than attitudinal ones.
Where this points instead. Software rather than hardware. A messaging assistant handling customer enquiries costs a fraction of a robot and addresses a problem small businesses genuinely have — a constant stream of repetitive enquiries arriving at all hours.
The implication for suppliers. Selling robots into this segment on operational grounds is difficult. Selling them on attention grounds works for a subset, and it should be positioned honestly as such.
The implication for buyers. If you run a small operation, the question worth asking first is whether your problem is distance walked or enquiries answered. The second is far more common and far cheaper to address.
A production mix weighted toward variety
Relevant to industrial applications.
What it means. Much manufacturing here runs many product variants at moderate volume rather than few variants at very high volume.
Why it matters. Traditional automation economics depend on spreading fixed setup costs — tooling, programming, fixtures — across large runs. Many variants at moderate volume is the hardest case for that model.
What works instead. Investment in changeover rather than in cycle time. Quick-change fixture bases, stored programmes, standardised presentation. These move the viable batch size down substantially and they are frequently under-invested.
The related observation. Where a single operation is common across many products — deburring, marking, packing — automating that one operation serves everything and multiplies effective volume.
The practical advice. Look for the shared operation rather than the highest-volume product. This is where automation cases hide in a high-variety business, and it is routinely missed because analysis is done product by product.
Building stock and physical constraints
An underrated practical factor.
Narrow frontages and multiple floors. The shophouse form is common, and it is close to the worst case for a mobile robot — narrow, vertical, with stairs.
Thresholds. Raised thresholds at bathroom and entrance doors are standard, and many robots cannot cross them.
Mixed floor levels. Steps between areas within a single storey are common.
Lift availability and age. In older buildings, lifts may not be connectable to a robot's control system, which limits a delivery robot to one floor.
Dust and outdoor exposure. Many premises open directly to the street, so sensors accumulate dust faster than in enclosed buildings.
What this means. Site survey matters more here than in markets with more uniform modern building stock. Measure the narrowest passage and the highest threshold before evaluating any product, because these frequently rule out mobile robots entirely.
Labour market conditions
The factor most often cited and most often oversimplified.
What is true. Certain roles are genuinely difficult to fill and retain — heavy repetitive work, night shifts, hot or dusty environments. This is a capacity constraint that cannot be resolved by pay alone, and it is the strongest driver of automation interest.
What is less true. That general labour cost makes automation obviously worthwhile. Full deployment costs of a robot cell frequently exceed what a simple wage comparison suggests, and the comparison must include integration, content, maintenance and the owner's time.
What is often overlooked. Seasonal variation. Periods following major holidays produce staffing gaps that create real operational pressure, and equipment that keeps running through them has value that annual averages do not capture.
The practical framing. Automate where hiring is hard rather than where labour is cheap. The first is a constraint; the second is a comparison that often does not survive full costing.
What this suggests about sequencing
Drawing the factors together into a practical order.
Software before hardware. For the majority of businesses here, an assistant handling enquiries addresses a bigger problem at a fraction of the cost, and the content built for it transfers to a robot later.
End-of-line before mid-process. Palletising and packing are relatively self-contained, high volume regardless of product variety, and physically demanding — the strongest industrial case.
Attention applications where they fit. Events, showrooms, launches. Real value, honestly budgeted as marketing, and best served by rental.
Changeover investment before cycle-time investment. In a high-variety context this is where the return is.
Maintenance capability before scale. The constraint on how many machines an organisation can usefully run is how many it can keep running, not how many it can buy.
That last point is the one worth ending on: in this market, capability to operate is a scarcer resource than capital to purchase, and planning that recognises it produces better outcomes than planning that does not.
Frequently asked questions
Why are robots a difficult fit for small premises here?
Small premises rarely have the distances that make delivery robots worthwhile, and small teams rarely have someone available to maintain equipment. Both are structural rather than attitudinal obstacles.
What works better than automation for high product variety?
Investment in changeover rather than cycle time — quick-change fixture bases, stored programmes, standardised presentation. And looking for an operation shared across many products rather than the highest-volume single product.
Which physical constraint most often rules out mobile robots?
Raised thresholds and steps between areas, plus narrow multi-storey premises. Measuring the narrowest passage and highest threshold before evaluating products frequently settles the question immediately.
How should the labour argument be framed?
Automate where hiring is hard rather than where labour is cheap. Difficulty filling night shifts and heavy repetitive roles is a capacity constraint; a wage comparison often does not survive full costing of integration and maintenance.
More in AI robots for real work and Myths and questions.
