Kate Callender, the CFO of BLUNT Umbrellas, had two options.
Hire a finance team big enough to match thousands of Shopify transactions against a wall of bank accounts by hand, or import summary totals and give up transaction-level detail. Neither worked. She runs a New Zealand company on NetSuite OneWorld with multiple country entities and a distributor network across the region, plus a high volume of Shopify transactions running through Afterpay, PayPal, and AmazonPay. She's quoted in the Zone & Co case study describing the choice before she added reconciliation tooling. She'd already run into the problem many APAC OneWorld operators eventually face: the storefront, payment, bank, and subsidiary reality was more complex than the standard connector story.
Not a BLUNT-specific problem. Most multi-subsidiary APAC operators on OneWorld eventually run into some version of it.
We've watched the same patterns across Hong Kong, Singapore, Kuala Lumpur, and Sydney enough times that we wrote it down. Off-the-shelf connectors, the kind you install from the SuiteApp marketplace and expect to "just work," tend to fail on APAC configurations in four ways, roughly in this order.
1. Presentment currency breaks day one
If you run Shopify Markets, a customer in Hong Kong might see prices in HKD, pay in HKD, and have that amount settled to your bank in USD at Shopify's FX rate. The order record in Shopify contains both numbers. A well-behaved connector separates them: it preserves the order currency for revenue and tax treatment, while routing the settlement amount into the right clearing or cash flow path. The exact accounting depends on your merchant-of-record model and revenue recognition setup, but the principle holds. Two numbers, two destinations.
Off-the-shelf connectors tend to collapse both into a single figure. Sometimes they pick the presentment amount and call it settlement. Sometimes they pick settlement and call it revenue. Either way, the month end reconciliation is wrong by whatever the cumulative FX delta happens to be, which on thousands of transactions is rarely a round number anyone wants to eyeball.
This is the first thing that breaks, because it breaks on the first order. You see it the morning after go-live. We've had finance teams call us in a panic on day two with a Shopify report and a NetSuite report that don't agree at the penny level, and the answer is almost always that the connector lumped presentment currency into a single column.
A 0.01 rounding difference per order across a few thousand orders a month produces enough noise to take two days to reconcile by hand. And that's before anyone talks about returns.
2. Multi-subsidiary routing is next
OneWorld supports hundreds of currencies and subsidiaries. What it doesn't do automatically is tell a connector which subsidiary a given order belongs to. If your Shopify store sells into Hong Kong, Singapore, Australia, and Malaysia from a single storefront your connector needs a rule. Shipping address and tender currency are convenient signals, but the actual routing rule has to reflect seller of record, inventory location, and nexus registration. Sometimes a HKD order to a Causeway Bay address is a straight HK sale. Sometimes it's an AU-entity sale fulfilled from Sydney and shipped in. The connector has to know which.
Most off-the-shelf connectors map a store one-to-one with a subsidiary. Fine for a Texas DTC brand that only sells to the US. Useless for an operator running HKD, SGD, MYR, AUD, and TWD through the same storefront.
Wrong-subsidiary posting is worse than a regional-P&L misreport. It breaks inter-company consolidation, throws off VAT and GST filing, and forces the finance team to hand-jot corrections every month. We inherited one implementation where Singapore sales had been booked into the HK subsidiary for months before the Q-end IRAS filing surfaced it. Cleanup ran into amended GST review, journal entries to unwind the inter-company receivables, and a conversation with the client's tax advisor about whether the HK returns needed refiling.
Celigo's Shopify flow handles this if you configure it correctly. NetSuite's own Shopify connector handles a subset of cases. Magento connectors vary by vendor. BigCommerce partners have a cleaner story because BigCommerce multi-storefront maps naturally to subsidiary-per-store. Subsidiary routing is never free. Assume it's a configuration project.
3. B2B pricing tiers stop the connector pricing sheet from helping
Week three or four into a rollout, someone asks when wholesale is going live. That's when the third failure surfaces.
Shopify B2B layers customer-specific catalogs, tiered pricing, net-30 terms, and company-level accounts onto the same platform as DTC. NetSuite has the same concepts: contract pricing, customer-specific price lists, credit limits. Connecting the two is where the off-the-shelf story stops. NetSuite's own Shopify connector was built around DTC flows and doesn't handle Shopify B2B cleanly. Third-party B2B-capable tiers typically run 2 to 4 times the price of the B2C versions and still don't map every field you need.
APAC merchants with serious wholesale requirements usually end up building custom workflows around the connector. We built one for D1 Milano, a Milan-based luxury watch brand running NetSuite and Shopify through Workato. A wholesale portal where distributors enter their own orders without pinging internal staff.
Faster than fighting the connector, and cleaner over 24 months than paying for a higher B2B tier and then customizing around its gaps anyway. If wholesale is on the 12-month roadmap, scope it into the original integration. Retrofitting B2B onto a live DTC connector is the most expensive rework pattern we see.
4. Returns break around week six
A clean same-currency full refund on an unfulfilled order clears every connector we've touched. A partial return, a return on a multi-currency order, a return split across two warehouses. That's where the connector silently starts creating orphaned records.
The usual failure modes: partial captures that don't unwind cleanly (a partial refund posted against the wrong original transaction, or a settlement batch that no longer maps 1:1 to the order), gift card redemptions where the liability release doesn't post against the correct deferred revenue subaccount, restocking fees hitting the wrong GL, and PayPal chargebacks arriving weeks later with no clean match to the original NetSuite transaction.
Six weeks in, the finance team opens the returns reconciliation and finds a pile of records that don't match anything. Unwinding takes days. Every month.
Most off-the-shelf connectors don't try to solve this. They handle the clean case and leave edge cases to the implementation partner. Fair enough for a $5,000 SuiteApp. Worth knowing before you sign the PO.
Where off-the-shelf connectors break
A typical APAC OneWorld + Shopify rollout
Presentment
currency
Shopify and NetSuite totals stop matching at the transaction level.
Connector collapses presentment + settlement into one figure.
FX deltas across thousands of orders. Two days of manual reconciliation.
Multi-subsidiary
routing
Orders post to the wrong entity, breaking consolidation and tax filings.
Connector maps one store to one subsidiary — no routing logic.
Hand-journaled corrections monthly. GST/VAT cleanup. Possible amended filings.
B2B and
wholesale logic
Customer-specific catalogs, tiered pricing, and net terms hit the connector's ceiling.
B2B-capable tiers run 2–4× the DTC price — and still leave gaps.
Wholesale rollout stalls. Custom workflow built around the connector anyway.
Returns and
edge cases
Refund exceptions and chargebacks turn into a recurring reconciliation backlog.
Partial refunds, multi-currency returns, late PayPal chargebacks.
Orphaned records pile up. Days of unwinding, every month.
Off-the-shelf connectors are built for single-entity merchants in single-currency markets.
Why APAC hits all four
Everything above happens to US merchants too. The reason it bites harder in APAC is that the four failures compound against conditions US merchants mostly don't face.
First, the currency stack. HKD, SGD, MYR, TWD, AUD on the same OneWorld tenant. Each with its own base currency lock. Once a subsidiary is set up, changing its base currency is either impossible or effectively so, depending on the exact vintage of your NetSuite instance — either way, a wrong pick at go-live follows the entity for its whole life on your books. Each subsidiary also carries its own tax regime. GST at 10% in Australia. GST at 9% in Singapore (up from 8% on 1 January 2024). SST in Malaysia. No VAT in Hong Kong, just profits tax on the entity. Taiwan runs Business Tax as VAT for general industries, with a separate GBRT track for financial institutions and specific F&B carve-outs. The tax math alone defeats connectors built for single-country merchants.
Then the payment gateway question. Among SuitePayments-certified paths, Adyen's NetSuite plugin is the one with meaningful APAC coverage: Alipay, Alipay HK, WeChat Pay, GrabPay, GCash, Touch'n Go, Malaysian and Thai online banking, and a handful of others. Stripe has its own Stripe Connector for NetSuite, a direct SuiteApp that pulls Stripe activity into the ledger without routing through the storefront (though its APAC method coverage is thinner than Adyen's). For merchants already on either, that's a reasonable route. Beyond those two, there's no equivalent NetSuite-side bridge. 2C2P, standalone PayMe, FPX, eNETS, and Razer Merchant Services don't appear on the SuitePayments directory. If your customers want to pay with those, you route through Shopify or Magento, let the storefront handle the gateway, and use NetSuite's connector to pull the transaction data in. Oracle's own NetSuite Connector documentation confirms this pattern: "NetSuite Connector does not have any direct payment gateway integrations. Almost all ecommerce systems such as Shopify, Magento, and eBay already integrate directly with payment gateways."
And the marketplace question.
If you sell on Shopee, Lazada, Qoo10, Tmall, JD.com, or Rakuten Japan, the answer lives with regional specialists, not the mainstream US middleware conversation. Shearwater has a published SuiteApp for Alibaba, Tmall, and Taobao. Sunoan publishes a NetSuite connector that covers multiple APAC marketplaces. TCT China's Orderin markets a connector across Lazada, Shopee, Tmall, JD, and CaiNiao. Celigo added Lazada and Shopee platform connectors in its 2023.9.1 release. OneCart out of Singapore covers Shopee and Lazada from a regional angle. These exist. They're also almost entirely invisible in the US-centric Netsuite integration content most operators first encounter.
For Tmall and JD specifically, even with a SuiteApp in hand, Trigger Networks and Shearwater Asia both note that a per-merchant application with the marketplace itself is still part of the path. Technology is only half the work. Qoo10 and Rakuten Japan are the genuine gaps: no published SuiteApp or Celigo template, which means custom or a regional partner engagement.
Pick a US NetSuite integration roundup at random. DCKAP's top-ten list, dated January 2026, names Magento, Shopify, HubSpot, Salesforce, Freshdesk, Adobe Commerce, Amazon, BigCommerce, shipping, and marketing automation. No Tmall. No Shopee. No Lazada. Nothing that looks like a real APAC marketplace stack. If your business is registered in Hong Kong or Singapore, the mainstream NetSuite integration conversation is not about you.
What to do instead
If you're a US single-entity merchant on Shopify, Celigo or the NetSuite Connector is fine. Go install one.
If you're running HKD, SGD, MYR, or TWD on OneWorld, the order of decisions matters. Scope subsidiary routing before you pick the connector, because the routing rule constrains which connector you can use. Model returns before go-live: walk every permutation, including partial returns, multi-currency returns, gift-card redemptions, and chargebacks that arrive weeks after the original settlement. If your connector's docs don't describe how it handles each, assume it doesn't. Decide whether B2B is on the 12-month roadmap before wiring the DTC integration; retrofitting is where budget really disappears. And keep payment gateways in the storefront layer with NetSuite consuming transaction data downstream. The exceptions are Adyen and Stripe, which each have a direct NetSuite path if you're already standardized on one of them.
Across the Celigo rollouts we've done in APAC, roughly 80% of use cases fit the pre built pattern. The remaining 20% (split shipments across APAC warehouses, custom fulfillment logic, multi-subsidiary routing) needs middleware customization. Budget for that 20% up front rather than discovering it in month three.
Base currency locks in early in the subsidiary lifecycle, and several consolidation settings are similarly hard to undo. Slow the go-live checklist down enough to make those choices consciously.
For APAC marketplaces, the layer question also decides itself: Shearwater or Orderin for Tmall and JD, Celigo's newer Lazada and Shopee connectors or OneCart for SEA, custom or partner work for Rakuten Japan and Qoo10. For DTC Shopify-to-NetSuite, Celigo plus thoughtful configuration covers most ground, though a custom SuiteScript build still beats it on complex return flows and multi-subsidiary routing.
Typical live-in-production timelines on APAC Shopify integrations run 4 to 8 weeks - Celigo pre-built flows on the faster end, custom SuiteScript at 6 to 8. That range holds even for larger and more complex builds: WobbleWorks, makers of the 3Doodler, moved from Google Sheets to NetSuite plus Workato and now run Shopify and Amazon through the same backbone.
Back to BLUNT. What made her stack work in the end wasn't a different connector; it was reconciliation tooling layered on top of one, matching the storefront traffic to the bank feeds she couldn't otherwise close on. That's the pattern for most APAC OneWorld operators. The connector is the plumbing. The reconciliation, the subsidiary routing rules, the returns flow, and the B2B path are the actual project.
Want a straight answer on whether any of this applies to your setup? We've seen it all. Book a 30-minute call with our team - no pitch deck, just a chat.
