The short answer
- PromoStandards is a shared API specification, so one integration reaches hundreds of suppliers
- Onboarding a compliant supplier can take about 24 hours — you need their endpoint URLs and credentials
- Products, colours, sizes, images, inventory and pricing all sync; purchase orders can go back the other way
- SanMar publishes a daily invoice file at 6am Pacific on FTP, separate from the live API
- Integration removes rekeying. It does not remove the need to check before a store opens
Most decorators sell other people's blanks. That means the catalogue on your storefront is a copy of somebody else's data, and every hour it goes uncorrected it drifts further from what can actually be bought. A supplier integration closes that gap automatically. What it does not do — and this is the part shops get wrong — is remove your responsibility to look.
What is PromoStandards, and why does it matter?
PromoStandards is a shared specification for how suppliers expose their data. Rather than every supplier inventing its own API, the compliant ones all answer the same set of calls for product data, inventory, pricing and order status. Several hundred suppliers now follow it, which is what makes a single integration worth building.
The practical effect is speed. For a supplier that already follows the specification, onboarding is usually a matter of supplying their endpoint URLs and credentials, and can be live in around 24 hours. That is a very different proposition from a bespoke integration per supplier, which is what the industry did before.
The large apparel suppliers sit slightly apart from this. SanMar and S&S Activewear both offer their own web services alongside standards-based access, and both are commonly available as pre-tested integrations. If those two are the bulk of your blanks, check that they are connected directly rather than assuming the generic route covers them well.
What actually syncs?
| Data | Direction | What it saves |
|---|---|---|
| Products, colours, sizes | In | Building the catalogue by hand |
| Images | In | Downloading and uploading product shots |
| Inventory | In | Selling something that is not there |
| Pricing | In | Quoting on last season's cost |
| Purchase orders | Out | Rekeying the order into the supplier's site |
| Order status | In | Phoning to ask where the blanks are |
The two lines worth dwelling on are the ones going outward and coming back. Automated purchase orders mean the order a customer placed becomes the order you place, without a person retyping sizes — which is where transcription errors enter. Order status coming back means the answer to "where is my order" exists before anyone asks it, which is the whole basis of removing status calls.
How fresh is the data really?
Not as fresh as "real time" suggests, and the difference matters. Live inventory calls reflect the supplier's stock at the moment of asking, but your storefront is not calling on every page view — it is caching, on whatever schedule your platform runs. Ask what that schedule is rather than assuming.
Some feeds are explicitly not live at all. SanMar generates a daily invoice file at 6am Pacific on their FTP server, which is a batch mechanism rather than an API call. That is perfectly usable and many shops rely on it, but a shop that believes it has minute-by-minute data when it actually has a once-a-day file will be surprised eventually.
An integration tells you what the supplier said. It does not tell you when they said it.
Set your expectations from the slowest link. If inventory refreshes nightly, then a style selling out at 9am is sellable on your site until the following morning. That is usually acceptable for a normal order and unacceptable for a team store about to commit two hundred people to a colourway.
What still needs a person?
Three things, and none of them are covered by any integration.
Discontinuations. An integration will stop offering a discontinued style, eventually. It will not tell you that the style you built a customer's whole programme around is going away, or suggest what replaces it. That remains a conversation with your rep.
Substitutions. When a colour disappears mid-order, only a person can judge whether the nearest match is acceptable for that customer and that job. What you can automate is having agreed the rule in advance, so the decision is quick rather than fraught.
The check before a commitment. Before a store opens or a large order is confirmed, someone should confirm the specific styles and colours are current. It takes under a minute and it prevents the single most expensive failure in this trade.
Questions worth asking your platform
- Which of my suppliers are connected directly, and which through a generic route?
- How often does inventory refresh, and is it an API call or a scheduled file?
- Do purchase orders go back automatically, or do we still rekey them?
- What does my storefront show when a supplier feed is unavailable?
- Is there a per-supplier or per-integration fee?
- Our supplier catalogue covers this across multiple suppliers from one place
That fourth question is the one that separates a considered platform from a hopeful one — a store that keeps serving a cached catalogue when the feed fails will cheerfully sell something that no longer exists. It is worth putting in front of a vendor alongside the other questions a demo will not answer. The rest of the suppliers archive assumes something other than memory is checking your catalogue.
What does an integration cost?
Rarely nothing, and the pricing shape varies enough to matter. Some platforms include supplier connections in the subscription, some charge per supplier connected, and some charge per order routed through the integration. The third of those turns a fixed cost into a variable one, which is fine at low volume and expensive at scale.
Ask for the number in writing against your actual supplier list, not against a general answer. A platform that includes two suppliers and charges for the rest is a different proposition to a shop using six.
Weigh it against what it replaces. If someone spends an hour a day rekeying purchase orders and correcting size transcription errors, the integration has a clear payback. If you place four supplier orders a week from two suppliers, the honest answer may be that a well-organised manual process is cheaper — and knowing that is worth more than buying an integration because it sounds modern.



