Internal programmes
Most companies run merch as an event. The better model is a small internal service with a catalogue, an intake route, named approvers and a delivery rhythm, so that a branch manager in Vanadzor can order a badge reel without emailing your marketing team first.
It usually starts with a booth. A team needs 200 staff shirts for a trade event, someone sends a logo by email, a supplier quotes, the shirts arrive six weeks later, and half of them are wrong. By the time the next show is booked, nobody remembers which vendor made the last batch, which specification file was used, or whether the 2024 print looked washed out because a different screen shop mixed the ink differently.
The failure is not the shirts. It is that every request is treated as a bespoke project with a fresh price negotiation. Administrative load per order stays flat while the number of orders grows, so the merch budget quietly consumes the marketing team's month. In our experience with Armenian banking and telecom clients, the request volume roughly doubles between the first and second year, and almost all of it comes from regional branches rather than head office.
A service model changes the unit of work. You stop selling custom projects to your own departments and start publishing a catalogue they can use. Regional staff get a fixed product list, a fixed price list, and a known lead time. Your marketing team stops being the production manager and becomes the owner of a standard. That is the shift: from purchasing artwork to governing standards.
Look at your last twelve months of requests before designing anything. In our books the request pattern is very stable: around 60% of units are apparel, 15% drinkware, 10% paper or desk items, and the rest is event-specific giveaway volume that disappears once the event is over. Build the catalogue out of the top six or eight items, not twenty. A catalogue is a speed tool, and every extra line slows the request down.
For each catalogue item you need a locked specification: the product, the decoration method, the artwork template, the colour options, the minimum quantity, the unit price at that minimum, and the lead time in working days. Once that specification exists, the person in the branch does not have to describe anything. They pick an item, pick a colour, pick a quantity, and submit.
Practical advice: keep three approval levels rather than one. Staff can order up to a set value without a second signature, team leads up to a higher value with their own approval, and anything above that goes to marketing or procurement. Without clear thresholds, the service either becomes bureaucratic enough that people route around it with personal contacts, or it becomes so loose that brand consistency disappears.
Eight to twelve SKUs, each with a locked specification, a unit price and a stated minimum. Prices in AMD including the 20% VAT treatment your finance department needs. The point of publishing the number is that nobody has to ask for a quote, and nobody has to wonder whether the last quote was fair.
One form, one mailbox, or one ticket category. The service commits to a response inside one working day, a proof inside three, and a delivery date confirmed at order acceptance. People use a service they can predict. An SLA you miss once costs you more trust than a price that is slightly above market.
Each catalogue item carries its own approver. Apparel goes to the brand owner, giveaway volume goes to marketing, anything with a logo lockup goes through the design reviewer. Branch managers order their own teams' apparel up to their threshold without waiting for head office, which is what makes the service feel like a service rather than a favour.
Decide in advance when people may reorder: once a quarter for core catalogue items, once a year for seasonal sets, and never inside a four-week window before a live campaign. Then archive every completed run, including the final artwork and the supplier record. The archive is the single most valuable artefact the service produces.
There is no platform licence. The cost is the catalogue, the form, and the time of whoever maintains the specifications. Realistically budget a day per quarter to keep artwork templates and price lists current, and a standing relationship with one or two production partners who already know your file setup.
The unit economics improve sharply once the catalogue is stable, and the improvement comes from three places. Setup fees amortise across repeats rather than being repaid on every new design. Screen printing at 12,000 to 18,000 AMD per colour stops being a line item you negotiate each time it appears. And production slots get easier to hit because the supplier can plan a week of identical work instead of a queue of one-offs.
Be honest about what a service does not fix. It will not make a bad product good, and it will not remove the need for decisions about quantity and timing. If your company runs three hundred events a year with a different audience each time, a fixed catalogue will not cover it and you need both models: the catalogue for the repeatable volume, and a project track for the exceptions.
Start with one department that has the highest request volume and the least brand exposure risk. Internal communications, field sales, or the customer service floor are good candidates. Publish eight catalogue items, run one quarter, and count what happened: how many requests arrived through the route, how many bypassed it, how many were refused, and how long each order actually took from request to delivery.
Those four numbers are the whole business case for scaling the service. If the bypass rate is high, your thresholds are wrong or the lead time is too long. If the refusal rate is high, the catalogue is missing something people obviously want. If request-to-delivery is longer than your stated SLA, the problem is production planning, not the process you designed.
The ownership question gets decided by accident more often than by design, and the outcome matters. If nobody owns it, the catalogue drifts: a colour gets substituted when an item is out of stock, a new vendor runs a rush job with a different ink, and within a year the company owns four different shades of the same navy. Assign a named owner. Marketing usually fits, because the catalogue encodes brand standards. Finance fits too, because the thresholds and the VAT treatment are finance questions.
Expect the owner to spend roughly a day a month on supplier coordination plus a day a quarter on catalogue maintenance. That is the real cost of the service, and it is smaller than the cost of the ad-hoc version it replaces. When we rebuild a client's process, the audit usually shows that the informal model consumed between four and nine staff days per month in quoting, chasing proofs and reconciling invoices, which is not counting the money spent on quantities that nobody needed.
Price the catalogue honestly rather than cheaply. A service priced below replacement cost gets treated as a favour, and people stop reading the rules because the rules clearly do not add up. Publish the unit price, the minimum, the lead time and the delivery charge separately, and let the person ordering see what they are spending. Transparency is what stops the internal politics.
Apparel is the easy half of a service catalogue because a person's size determines the quantity and the reorder logic is obvious. Giveaway items are harder, and they are where programmes lose money. A giveaway item ordered at a quantity nobody uses is not a gift, it is inventory in a cupboard, and in our experience it is still there three years later.
Two rules help. First, giveaway volume moves to a project track with a named budget holder and a written justification, because unlike a polo shirt, a stress ball for 600 people needs a reason to exist. Second, agree the distribution method before production, not after. If the item goes to a booth, you need a count of attendees. If it goes to existing customers, you need a mailing list. The number you print should be the number you can hand out plus a margin you have decided deliberately, not a round number somebody liked.
Keep giveaway items out of the standing catalogue entirely. They belong in the exception track, reviewed each time, because the audience and the objective change per campaign and a fixed item on a fixed rhythm is usually wrong for one of them.
Collect every request that went anywhere in the past year. Product, quantity, method, who paid, who approved, and what it actually cost including internal time. This audit is usually more persuasive than any proposal, because the total number surprises the people who approved it.
Pick eight to twelve SKUs from the audit data. Lock each specification, get real production quotes at the stated minimums, and publish prices. Write the approval thresholds at the same time, because they are two halves of the same decision.
Open the intake route, run the stated SLA hard, and log every exception. A pilot that breaks visibly in month two is worth more than a careful launch that fails quietly in month six.
Extend to all departments, add the seasonal set on the annual cadence, and archive the pilot records. Review request-to-delivery time and bypass rate quarterly, and adjust the catalogue by dropping what nobody orders rather than adding what one person asked for.
Eight to twelve. Below that, requests that fall outside the catalogue go around the process, and above that, people cannot find the item they want and give up. Most of our running service programmes for banking clients live in that band, with a seasonal replacement each year rather than growth.
Let branches order their own repeat business directly, up to a value threshold, and route anything novel or high-value through head office. In our experience regional teams order apparel for their own staff more than a dozen times a year, and central approval on those requests is where internal services usually die.
Locked artwork templates, a limited colour list per product, and one reviewer for anything that changes a lockup. The catalogue is what enforces this: if the only orderable version is the approved one, there is nothing to police later.
No. Corp-Merch is an independent merchandise production and supplier company. We work with corporate buyers in the market on their own branded programmes, and we are not a partner, sponsor or authorised supplier of any bank, telecom operator or industry body.
Urgent work goes through a fast lane with a named approver and a stated premium, and the quantity floor rises, because a rush order under fifty units usually costs more per piece than the catalogue price and there is no way around that. We would rather quote the real rush cost than hide it inside a delivery date.
Yes. Delivery across Yerevan, Vanadzor, Gyumri and regional offices is routine, and one consolidated shipment to a central store often beats many small drops. The design question is whether branches hold stock or draw from head office, and for most programmes a small central buffer plus quarterly replenishment is simpler than local inventory.
That is a signal, not a rebellion. People route around a service that is slower or more expensive than the alternative. Check the published lead time and the price against what a branch can get on its own, and fix whichever one loses. A catalogue that respects the stated terms gets used without enforcement.
Collect sizes at request time rather than assuming a standard distribution, then publish a size ratio that your workforce actually matches. Armenian offices vary a lot by team, and the usual medium-heavy skew we see is not universal. Where data is missing, order a small adjustment run and correct the ratio for the next order.
Corp-Merch is an independent merchandise supplier. This page describes an operating model we have set up with corporate buyers; it does not represent any client, bank, event organiser or industry body, and no programme described here implies a partnership, sponsorship or official supply relationship.
Tell us roughly how many requests your teams made in the last year and what they asked for. We will come back with a draft catalogue structure, realistic lead times and what it should cost at your volumes.