Production capacity · Yerevan · screen print · DTG · embroidery

Local print capacity in Armenia: what 2026 can and cannot do

Local production is the right answer more often than buyers assume, and the wrong answer more abruptly than they expect. The switch is not gradual and it is not about quality in the abstract. It happens when the order crosses a machine-hour ceiling, when the dye lot cannot be repeated, or when the blank itself has to be imported anyway. This page sets out where the real ceilings sit in Yerevan production, what lead time to expect at each volume band, and how to work out before you order whether your project fits local capacity or has already outgrown it.

Check capacity for your run All B2B notes

What local capacity actually is in Yerevan

The phrase local production covers four quite different operations, and conflating them is the source of most bad capacity expectations. A Yerevan workshop that owns a four-colour carousel press and a heat press can print apparel in-house while sourcing every blank from an importer, and it can laser-engrave metal in-house while every piece of steel arrives from abroad. The workshop that appears to be the local answer on a website may be a local answer for exactly one step of a five-step supply chain. Any capacity discussion that does not name which steps are local is not a capacity discussion.

The second thing worth knowing is that local capacity is measured in machine-hours, not in units. A garment decoration line running at a realistic rate of 400 to 900 garment-prints per working day cannot be asked for 5 000 pieces in a week without either a second shift or a subcontractor. An embroidery line on two multi-head machines running at a realistic 600 to 1 200 garment-equivalents per working day has the same shape of ceiling. A direct-to-garment printer producing at a realistic 250 to 500 garments per working day is smaller still. These are ordinary figures for the equipment class, and the useful question for a buyer is not whether the workshop is capable but whether the arithmetic of their own order fits the ceiling.

The third thing is that the ceilings move with the season. Late autumn through early spring carries festival, corporate gifting and conference season simultaneously, and every workshop in the city is full in the same weeks. This is the period in which local lead times extend most sharply, and it is also the period in which the temptation to accept a longer quoted lead time is strongest. A buyer who places an order in November for a December event without a buffer is not negotiating hard; they are accepting the queue.

Screen print ceiling: 400–900 garment-prints per working day on a standard carousel line, colour passes included
DTG ceiling: 250–500 garments per working day on a single industrial unit, before pre-treatment throughput
Embroidery ceiling: 600–1 200 garment-equivalents per working day on two multi-head machines
Hard-goods decoration: Laser and pad capacity is ample; blank supply is the constraint, not the machine
Practical comfort band: Up to roughly 1 500 decorated units in a single local run before subcontract risk rises sharply
Peak-season effect: Lead times stretch most between late October and mid-December, and again in March for spring campaigns
Common silent subcontract: The blank itself, and any embroidery above the in-house machine count
Lead time, 100–500 units: 7–12 working days from artwork approval, peak season 12–20
Lead time, 500–1 500 units: 12–18 working days, peak season 18–30
Lead time, above 1 500 units: 20–35 working days locally; import routing becomes competitive

Four operations, four different ceilings

The method you choose determines which ceiling you hit first, and the four methods do not fail in the same way. Screen printing fails on colour passes, digital printing fails on hourly throughput, embroidery fails on machine count and digitising backlog, and hard-goods decoration almost never fails on the machine at all. Understanding each failure mode is what lets a buyer redesign a brief so it fits local capacity instead of discovering the ceiling after the order is placed.

Screen printing: colour passes are the real capacity

A screen printing line is not constrained by garments, it is constrained by passes. A one-colour one-location print consumes one pass per garment; a three-colour two-location print consumes six; a six-colour design at two locations consumes twelve. On a four-colour carousel, a twelve-pass design is roughly six times the press time of a two-pass design, and that ratio is exactly what the quoted per-unit price is reflecting when a low-volume and a high-volume quote for the same artwork differ so much. The practical ceiling locally is around 1 500 to 2 000 garment-prints per garment-position per working day on a single four-colour carousel at realistic efficiency, which means a 2 000-unit order with a six-colour design can consume an entire machine for a day and a half before the second production step even starts. Above roughly 3 000 units on a complex design, the local line stops being economical and a second pass or an overflow subcontract becomes the honest answer. Buyers who reduce a six-colour back print to a two-colour front print and a single-colour sleeve print get the same visual effect at a fraction of the pass count, and that redesign is usually available if it is raised before prepress rather than after.

DTG and digital printing: no setup ceiling, an hourly one

Direct-to-garment printing inverts the screen printing constraint almost completely. There is no screen, no colour separation, no pass count, so an eleven-colour photographic design costs the same machine time as a two-colour flat design and the setup cost is close to zero once the file is pretreated and loaded. What replaces the screen constraint is throughput: an industrial DTG unit produces in the region of 250 to 500 garments per working day in a realistic production pattern, and the pre-treatment and curing steps around it add their own queue. The practical consequence is that DTG is the right answer for short runs with complex artwork — fifty to 300 units, many colours, photographic content, different artwork on every unit — and the wrong answer for anything that needs a genuinely large quantity of a simple design. It is also the right answer for the run that cannot be screen printed at all, such as a design with continuous-tone photography or a gradient, and for personalised runs where each garment carries different content, since the per-unit setup is effectively nil. Buyers who try to use DTG for a 2 000-unit one-colour run are choosing a slower and more expensive route and usually discover it in week two.

Embroidery: machine hours and the digitising queue

Embroidery is the operation most often quoted with the least honest lead time, because two different queues sit behind the headline figure. The first is machine time, which depends on stitch count: a 3 000-stitch cap runs at roughly 20 to 25 minutes on a single head, a 6 000-stitch chest logo runs at 40 to 50 minutes, and a 12 000-stitch large-format back print can exceed 90 minutes. On a two-head machine doing single-colour work with batched thread changes, that translates to somewhere between 600 and 1 200 garment-equivalents per working day, and it halves when the design needs a thread change per unit. The second queue is digitising, which is a one-off per design but sits behind a specialist and is the reason a new complex logo often adds a week to an otherwise routine run. The ceiling arrives well before most buyers expect it: a 1 500-piece cap run with a 6 000-stitch front design is already a multi-day machine commitment, and a 3 000-piece hoodie run at that stitch count is a fortnight on one workshop. Multi-head machines reduce the linear part of the queue but do not remove the digitising or the thread-change overhead, so the honest answer for a large embroidery run is still subcontracting, not optimism.

Subcontracting: the step that is usually not in the price

The fourth block is the one that decides whether a local answer is a local answer at all. Most local quotes cover a chain rather than a workshop: the blank comes from a knitwear importer, the embroidery above a certain head-count goes to a facility in another city, the packaging is sourced from a supplier of boxes rather than made in house, and the laser engraving on a cylindrical item is done on a subcontract line. None of this is problematic, but it must appear in the quote and in the lead time, because it is where the schedule actually goes. A buyer who assumes a Yerevan workshop prints, cuts, sews, embroiders and packs in one building has misread the offer, and the gap shows up as a committed date that slips when the subcontracted step runs late. The three questions to ask before accepting a local lead time are simple: which steps happen in Yerevan, which happen elsewhere, and who owns the handover between them. A supplier who answers that clearly is managing the chain. A supplier who answers vaguely is discovering the chain after the order is released, and the discovery lands on the deadline.

Lead time bands and where they break

Lead time in this category is not one number with noise around it. It is four sequential stages, each with its own queue, and the total is the sum rather than the maximum. The stages are artwork preparation and prepress approval, sourcing and goods-in of the blank, production, and finishing including folding, packing, labelling and delivery. In a well-run local shop with stock blanks the first stage is one to three days, the second is zero to three days if the blank is held, the third is one to eight days depending on volume and method, and the fourth is one to two days. That gives the familiar 7 to 12 working day band for a 100 to 500 unit run, and the number is almost entirely consumed by the two queue stages rather than by production.

The bands break at predictable places. Below 100 units the lead time is dominated by setup and prepress, not by volume, so a 30-unit run and a 90-unit run take almost the same time. Between 500 and 1 500 units the production stage becomes the largest block and lead time grows roughly in proportion. Above 1 500 units on a complex screen design, local production stops fitting inside a normal working window and the honest lead time moves to 20 to 35 working days, at which point an imported blank with local decoration, or a fully imported finished item, becomes competitive on both time and cost. In peak season every band stretches by a factor of roughly one and a half, and the stretch is worst at the front of the queue rather than in the middle of production.

Two practical rules follow. First, the committed date should be built backwards from the event date by the total band plus a buffer of three working days, not by the optimistic figure a supplier gives when the order is still an enquiry. Second, a quoted lead time that is shorter than the sum of the four stages is usually a lead time that assumes blanks are in stock; confirm the blank position in writing, because a stock assumption that turns out to be wrong is the most common cause of a missed date in local production, and it is not always the supplier's error.

Doing the capacity arithmetic before you request a quote

A buyer can run the capacity check themselves in about ten minutes, and doing so converts a conversation about availability into a conversation about a specific plan. The arithmetic has four inputs and one decision, and it is worth writing it down because it also becomes the basis for deciding whether to change the specification rather than the supplier.

Input one: units and SKUs. Total units per SKU, not total units across the order, because capacity is consumed per design position. Input two: colour passes. Colours multiplied by locations per garment, which is the screen printing consumption figure. Input three: stitch count. Total stitches per garment for embroidery, converted to machine minutes using roughly 140 stitches per minute as a practical single-head rate and adjusted upward for thread changes. Input four: the calendar. Which weeks the run must occupy, and whether those weeks sit inside the autumn-to-spring peak.

The calculation is then: pass count times units divided by daily ceiling gives screen printing days; stitch minutes divided by head count divided by minutes per day gives embroidery days; units divided by daily DTG output gives digital days. Add the fixed stages, then divide by working days available before the event date. If the result leaves fewer than three working days of buffer, the order does not fit local capacity as specified, and there are exactly four levers: reduce the pass count, split the delivery into two runs, move a method to a faster one, or import. That decision, made deliberately at the quotation stage, costs nothing. The same decision made in week three of production costs the campaign.

When importing becomes rational rather than a compromise

Importing has a reputation problem in this category that it does not deserve, because it is the better answer in a specific and fairly wide band of cases. The test is simple arithmetic: compare the delivered cost per unit of an imported finished item against the delivered cost per unit of a locally decorated item with an imported blank. The locally decorated option looks attractive because it promises a short lead time, but it carries the import cost of the blank, the local setup cost, the local decoration cost, and the local finishing cost, and each of those is charged at small-batch pricing.

There are four situations where import wins outright. A large single-SKU run above roughly 3 000 units, where factory setup is amortised across enough units to bring the per-unit figure well below what local pricing can reach at any volume. A multi-colour photographic design at scale, where screen printing would need an impractical number of passes and local DTG is too slow for the volume. A tightly specified technical product — a specific fabric weight, a certified substrate, a branded hardware item with a published specification — where the blank simply is not available locally at any price. A deadline that is genuinely tight and a volume above the local ceiling, where the question is not cost per unit but whether the date can be met at all.

Import carries three real costs that should be priced honestly into the comparison: a longer and less certain lead time, freight and customs that add a share to the landed figure, and less opportunity to intervene when something goes wrong. Against that, an imported finished item brings a single accountable party for the finished result, which for a bank or a large corporate campaign has real operational value. The decision is rarely binary. A common and effective pattern is to import the bulk standard item — the plain cap, the tote, the notebook — and decorate only the small hero quantity locally, which keeps the programme inside local capacity while getting factory economics on eighty per cent of the units.

Quality variance, dye lots and the thing local cannot repeat

The honest weakness of local production at volume is not defect rate, which is generally comparable, but repeatability across separated time. An apparel blank from an importer arrives as a set of dye lots. A 2026 campaign order and a matching 2027 re-order drawn from a different lot will not be the same colour, and no amount of care at the decoration stage prevents that, because the difference is in the substrate. Screen print ink can be matched to a Pantone reference precisely; the garment underneath it cannot be re-dyed to match a previous run. This is the single most common reason a corporate merchandise programme becomes expensive after two years, and it is worth building a mitigation into the plan rather than discovering it in the third season.

The mitigations are unglamorous and effective. Specify a named dye lot and require that the lot number is recorded on the delivery note, so the same lot can be reserved or re-ordered while stock lasts. Order a little more than the programme needs and hold the surplus unwashed in a controlled store so that a later replenishment comes from the same lot. Where the programme includes a highly visible colour, approve the new lot against a physical reference sample under daylight rather than under shop lighting. And where consistency matters more than cost, this is one of the clearest cases for importing the blank from a single source and decoration locally, because a single import lot held in store removes most of the variance for years.

The same reasoning applies to printed hard goods. A Pantone reference on a ceramic mug, a bottle body or a pen barrel will hold across batches because the glaze or coating is applied to the decoration directly. A Pantone reference on a dyed textile is a request, not an instruction. Buyers who understand that distinction will not hold a textile supplier to a colour-match tolerance that the substrate cannot deliver.

Using local capacity properly across a year

The mistake is not ordering locally. It is ordering locally without a calendar. Capacity is a shared resource with a known seasonal shape, and a programme that treats production as an on-demand service will pay peak prices and accept peak lead times without noticing. The alternative is a simple rolling plan with four fixed anchors: the volume committed by month, the blank order placed far enough ahead to avoid stock-outs, the production slot reserved rather than requested, and the delivery scheduled against the campaign calendar rather than against the completion date.

In practice the annual shape looks like this. January to February is the deep low season locally, with the shortest queues and the most capacity for a production slot, which makes it the right month for the blank order and for any run that must be ready before spring. March to May builds toward the conference season and the lead times start moving. June to August is stable for mid-volume work and is the right window for the autumn gifting programmes, which need to be finished well before the demand appears. September and October tighten quickly as festival and year-end orders land together, and this is when a planned slot is worth more than a good price. November and December are the sharpest period in the local calendar, and anything committed here should be placed in August.

A supplier who will reserve a production slot, say what the blank position is, and confirm the dye lot in writing is giving the buyer something more valuable than a discount: control of the queue. Buyers who plan against that calendar typically find their cost per unit improves as well, because the workshop is filling empty slots rather than bidding for overflow work at short notice.

Questions buyers ask about local production capacity

What is the largest printed apparel run a Yerevan workshop will usually take on in one go?

Around 1 500 to 2 000 garment-prints of a simple one-colour one-location design is comfortable on a single carousel line. A three-colour two-location design consumes six passes per garment and effectively divides that. Beyond roughly 3 000 units on a complex design, the workshop is normally right to route the overflow to a partner line, and a buyer who is told that up front is dealing with a supplier who understands the constraint rather than one who is guessing. The practical threshold to plan against is volume multiplied by pass count: keep that product comfortably inside a working day if the date is fixed.

Is it cheaper to have items printed locally or imported finished from a factory?

It depends almost entirely on volume and design complexity. For a few hundred units with a simple design, local production is cheaper once setup is amortised and freight is removed from the landed cost. From roughly 3 000 units upward on a single SKU, factory economics pull the other way, because the factory setup is spread across a much larger base and the freight share of landed cost falls as the order value rises. Designs with continuous-tone photography or many colours favour DTG locally at small volume and factory printing at large volume. A mixed pattern is often cheapest of all: import the standard blanks in bulk and decorate the hero quantity locally.

How far in advance should an Armenian company place a large local merchandise order?

For 100 to 500 units, seven to twelve working days from artwork approval is realistic outside peak season and twelve to twenty inside it. For 500 to 1 500 units, allow twelve to eighteen working days, or eighteen to thirty between late October and mid-December. For anything above 1 500 units, plan twenty to thirty-five working days locally and consider import routing from the start. The critical qualifier is the blank position: if the garment or cap is not held in stock locally, sourcing adds three to ten working days that are not visible in the production figure, so that position should be confirmed in writing before a date is committed.

Is Corp-Merch a partner or official supplier of the printers, workshops or manufacturers referenced on this page?

No. Corp-Merch is an independent B2B merchandise supplier. References to equipment classes, production methods, capacity ranges, lead times and quality-control practice describe general characteristics of the category as we understand them, and imply no partnership, sponsorship, endorsement or official-supplier relationship with any workshop, machine manufacturer or trade body. Nothing on this page is a technical standard, a production guarantee or a quotation; every figure is indicative for planning and must be confirmed against a written specification, a written capacity check and a signed contract.

This page is written for procurement, marketing and production teams evaluating where a corporate merchandise order should be produced. Corp-Merch is an independent B2B merchandise supplier. The capacity ceilings, daily throughput figures, lead time bands, dye-lot behaviour and costing relationships described here are general characteristics of this production category as we understand them; they are not engineering specifications, machine specifications, certification claims or a commitment about the capacity of any particular facility. Real throughput depends on the equipment actually installed, the material selected, the design, the season and the queue. Every order is priced and scheduled individually against a written specification, a written capacity confirmation and a signed contract, and any figure on this page should be treated as a planning estimate until it appears in a confirmed offer.