One catalogue, many storefronts.
Most product businesses do not choose a channel mix so much as accumulate one. A webstore, then a marketplace, then a wholesale portal somebody asked for. Each holds its own idea of what is in stock, and somewhere in the middle a person is reconciling them by hand. Cin7 Core is where that stops being a person’s job, provided the rules underneath it are set properly first.
Not everything connects the same way.
Which channels you can put behind Cin7 Core matters rather less than how each one gets there, because that is what decides who fixes it when it stops. Every channel below is a built-in connector, which is not true of everything else in our stack: elsewhere there are apps that reach you only through another platform and never touch Cin7 Core at all.
Built into Cin7 Core and configured from its own Integrations area. You turn one on, map your products and locations, and Cin7 maintain the connection at their end.
- Amazon Seller CentralThe marketplace channel, natively connected
- BigCommerceA native storefront connector
- eBayA native marketplace channel
- EtsyA native marketplace channel
- FAIREWholesale orders arriving as a channel
- LeafLinkA native wholesale marketplace channel
- LightspeedRetail point of sale as a channel
- Magento 2.0 / Adobe CommerceEnterprise storefronts, current and legacy
- ShopifyThe channel most of our clients start from
- SquareA sales channel in its own right
- WalmartA native marketplace channel, US-facing
- WooCommerceWordPress storefronts, natively connected
Shopify has a page of its own, because its B2B and point-of-sale products make it three channels rather than one. If most of your volume sits there, start there
Square appears twice in our stack: a sales channel in its own right, which is what it is doing above, and the processor sitting behind Cin7 Core’s own point of sale, which is a different connector and a different licence. What taking the money involves
Amazon is on both sides of this: a marketplace channel above, and FBA or Multi-Channel Fulfilment on the shipping page. Orders arriving is half of it, and parcels leaving is the other half. Getting the orders out
Four things go wrong, and they are always the same four.
None of these is an obscure edge case. Cin7 have a help article for every one of them, which is the tell: this is what a multi-channel setup does to you when the rules were left on their defaults.
Two channels selling the same unit
Stock is pushed out from Cin7 Core to a channel, never back the other way, and it goes out per location. So the answer to “why did we oversell?” is almost always a mapping question: which locations feed which channel, and whether the product was ever mapped to a location at all. A buffer quantity, set per channel or per product, is the usual safety margin. It should be a considered number rather than the same three units everywhere.
The order that ships from the wrong warehouse
Cin7 Core can pick the fulfilment location from the delivery address, matching on postcode, then city, then state, then country. It is genuinely useful and it is fussy: the address fields have to be spelled the same way on both sides, and there is a separate setting for what should happen when the nearest location has run out. Get either wrong and orders route somewhere plausible but expensive.
Selling what you have not got yet
Backorder can be allowed for a whole channel or switched off product by product, and it depends on how the channel has been set to capture and process orders. It is a commercial decision more than a technical one, and it is worth taking deliberately per channel: what a marketplace tolerates on a late dispatch is not what your own store’s customers will.
One SKU, three listings
Cin7 Core insists a SKU is unique. Storefronts generally do not, and third-party apps happily create a second listing on the same SKU at a different price. Cin7 Core can import duplicates, but it does it by building a kit product with an auto-assembly behind each one, so what looks like a tidy import leaves a product list that needs explaining. Cleaning the catalogue first is usually cheaper than living with the workaround.
A launch is not a normal trading day.
For most of the year the sync keeps up and none of this needs thinking about. The exception is a drop or a flash sale where a product is meant to sell out: orders arriving milliseconds apart, stock updates going the other way, and both systems briefly disagreeing about how many are left. That is when overselling happens, and it happens at the moment it costs the most goodwill.
Our usual advice is to stop the conversation for the duration. Count the stock, push the true figures out, then turn the automatic updates off and let the channel run down its own number while orders keep flowing back into Cin7 Core. Afterwards you reload orders over a date range starting before the sale, check what is left sitting in the connected locations, and turn the sync back on.
The trap is running several channels off one warehouse. Freeze the sync and each of them believes it owns the whole quantity, so three channels with a thousand units between them behave as though there were three thousand. The stock has to be split across them by hand, and put back afterwards.
- ✓ Worth doing only when a product is expected to sell out under real load
- ✓ Stocktake first, because every figure downstream comes from it
- ✓ Split the quantity between channels sharing one location
- ✓ Reload orders from before the sale started, then reconcile
- ✓ Bring the quietest channel back on sync first, the busiest last
We decide the rules, then connect it.
Turning a connector on takes an afternoon. Deciding what each channel is allowed to see, and what happens when two of them want the same unit, is the part that takes the time and the part that is painful to change later.
Decide what each channel sees
Which locations feed which channel, what stock is held back, where a product is allowed to go on backorder, and how a marketplace order is told apart from a wholesale one once it is inside Cin7 Core.
Get the catalogue honest first
Duplicate SKUs, half-listed variants and products mapped to no location cause most of the mess. Sorting them before the connection goes live is cheaper than sorting them after, and considerably less confusing.
Connect it, then test the awkward cases
Partial shipments, returns, bundles, a customer on a special price, and the order that arrives while the stock is being counted. The connector handles the ordinary path on its own. It is the rest that needs proving.
What we are not.
We are not a growth agency, and we would rather say so plainly than let a first call run twenty minutes before it becomes obvious. Which channels are worth being on, what a listing should say, how to price against the competition on it, whether the product has a market at all: those are real questions and they are not ours. We can tell you what a channel and Cin7 Core do to each other. We cannot tell you whether the channel is a good idea.
We are not your storefront agency either. Themes, templates and front-end build are somebody else’s trade, and we would rather work alongside whoever does yours than pretend otherwise.
And we are not your accountant. We should be able to make sure channel sales land in the accounting system in a shape that reconciles. How any of it is treated for tax is a conversation for the person who signs off your books.
Us, on camera, on the connections themselves.
What people ask before starting.
How many channels can we run off one Cin7 Core account?
There is no number that breaks it. What limits you in practice is how clearly you have decided which stock belongs to which channel, because that decision has to be made once per channel and then held to.
The setups that struggle are usually the ones running several channels off a single pool of stock with nothing held back. That is workable, but it needs buffers and it needs somebody watching them.
Our channel is not on your list. Can it still connect?
Often yes, through the Cin7 Core API or a middleware party, and we would look at that on a first call. What we would not do is tell you a connection is straightforward before we have seen what is at the other end of it.
The list above is what we have done the work on and can speak to properly. Anything outside it is a scoping conversation rather than a no.
Do we have to clean the product list before we start?
In our experience it is the thing most likely to decide how the project goes. Duplicate SKUs, variants that exist on one channel and not another, and products that were never mapped to a location all survive the migration and turn into support tickets afterwards.
It does not have to be perfect. It does have to be understood, and we would rather spend a phase on it up front than find it during go-live.
How long does connecting a channel take?
The connection itself is short. The catalogue and the rules are what take the time, and they scale with how many locations, variants and per-customer prices you have rather than with how many channels.
We quote in phases at £99 an hour, as estimates rather than a fixed scope, and we give a range at scoping. We would rather set it wide than miss it.
Something is connected already and going wrong. Can you look at it?
Yes. Sync failures between Cin7 Core and a sales channel are one of the three things people call our support service about most.
If the list of symptoms is long, a system audit is usually the cheaper start, because it finds the decision underneath the symptoms instead of treating them one at a time.
What we have written about connections and the way they fail.
Tell us where you sell.
List the channels, however untidy the list is. Book a call and we will tell you which of them Cin7 Core connects to natively, which need another route, and what would have to be true of your catalogue before any of it is worth doing.
please check your timezone and am/pm when booking
What to expect
- ✓ A straight conversation, no scripted sales pitch
- ✓ An honest view on whether you need this yet
- ✓ Estimates by phase before you commit to anything
- ✓ A clear split of our work and your storefront agency’s