PAHI

• The fix Celigo recommended on 27 August was applied to the custom form rather than to the Return Authorization records. Three attempts, all on the screen layout.
• The error has since moved on. The condition blocking refunds today is not the one discussed on the calls.
• All seventeen errors on the refund flow occur when writing into NetSuite. Reading from Shopify succeeds every run.
• The contract is non cancelable and runs to 30 August 2027 at $3,248 a month for NetSuite alone.
• A $13,500 Shopify connector that advertises refund sync is already licensed and has never been switched on.

Source: NetSuite (1).pdf

• Cancellation: the order form is non cancelable and the fees are non refundable. There is no early exit to negotiate.
• End date: the start date field was left blank, so the terms make service start on the signature date. That fixes the end at 30 August 2027.
• Renewal: current pricing carries a fifty percent discount. Discounts of that size win accounts rather than keep them, so renewal will be a repricing conversation.
• Scale: the licence covers 5,000 orders a year and up to 100 users. Ten seats were purchased. Current volume runs at roughly a quarter of the order allowance.

Around $50,000 a year leaves the business until 30 August 2027 under every option in this document. That figure is fixed and no recommendation can change it.

The connector already licensed

The paid Shopify connector you mentioned appears on the order form as line item three: NetSuite Premium Plus eCommerce Connector, $13,500 over 36 months. Its description names Shopify and includes real time order sync, full product, price and quantity sync, coupon and discount sync, settlement sync, and refund sync.

Celigo was purchased separately, at roughly $7,800 a year, for the same work. The Celigo refund flow is the one that has not run since June.

Click to zoom

There may be a sound technical reason the bundled connector was rejected. If there is, we want it, because a constraint strong enough to rule it out will probably affect any replacement too. What is not in doubt: one connector is paid for and idle, a second is paid for and failing at the job the first one advertises.

The spend to August 2027 is committed. The open question is what PAHI holds on 30 August 2027, when the renewal lands. Four options, all viable.

Assessment

• Option A, for: an Oracle connector naming Shopify and advertising refund sync is already licensed and idle. If it works, Celigo is cancelled, roughly $7,800 a year comes out immediately, and the returns problem may resolve using software already bought. No migration risk, no parallel running.
• Option A, against: renewal arrives with no alternative and no comparison. Every further build inside NetSuite deepens the commitment, starting with B2B.
• Option C, against: leaving early costs the same as staying, because the contract cannot be cancelled. The only purchase is fewer months of NetSuite, paid for with a compressed migration, a cutover pushed toward peak, and no buffer.
• Option D, against: the renewal arrives before the analysis does. At that point the practical choices are to sign again, or to attempt in weeks what this plan spreads across eleven months.

Recommendation

Option B, for three reasons.

• It opens with the same free work as Option A, including the connector question and the refund fix. Nothing is lost if PAHI later decides to stay.
• $2,000 to $5,300 across the whole remaining term, against $50,000 a year, is close to a rounding error.
• What that buys is a choice. On 30 August 2027 both answers are genuinely available.

Option A is a defensible decision. If appetite for change is low this year, Option A done well beats Option B done badly. The first six weeks are identical either way, so the choice does not have to be made this week.

Eleven months, working back from the contract boundary. Two fixed points shape it: nothing cuts over during peak season, and the ledger moves at a fiscal year boundary.

The refund flow is fixed in the first fortnight under every option. It is NetSuite configuration, it costs nothing, and it ends the manual exchange work now. Nothing about it waits on the larger decision.

NetSuite currently performs eleven distinct jobs. All eleven are accounted for.

The same work, for roughly one tenth of the cost. The saving begins at contract end on 30 August 2027, and the work to be ready for it begins now.

Requirements not met by standard applications

• Third party logistics. Inbound and outbound both run through NetSuite today. Most 3PLs now connect to Shopify directly through the fulfilment orders API, which would cost nothing. Whether Barrett does is the single item most likely to change the shape of the plan.
• Landed cost. Shopify's purchase order receiving records quantities, not landed cost per unit, so duty, freight and clearance do not reach the cost of goods on their own. For an importer of footwear that determines whether margin is known or estimated.

Three ways to close the landed cost gap, in the order we would try them:

1. Prediko, if its purchase order costing is sufficient. Tested against a real inbound container before it is recommended, not against a demo.
2. An operations layer, if Prediko falls short. Cin7 Core at $349 a month adds full purchasing, landed cost, warehouse and 3PL handling. Roughly a sixth of NetSuite, and another system to run.
3. A purpose built application, if neither fits. A small Shopify app covering landed cost and factory purchase orders, and nothing else. Source control, tests, reviewed changes, documentation alongside. Owned by PAHI and exportable.

The third option is earned rather than assumed. Building software you did not need costs more than buying software you did not need.

• The licensed connector. Why was the bundled Premium Plus eCommerce Connector set aside in favour of Celigo?
• Barrett. How the 3PL receives orders and posts receipts, inbound and outbound, in enough detail to scope a replacement.
• Wholesale pricing. One shared price list across accounts, or negotiated individually? This determines how catalogues are built on either platform.
• Wholesale terms. Whether any account trades over EDI or through a supplier portal rather than by email.
• Sandbox access. NetSuite and Celigo, so the refund fix can be tested rather than described.
• The returns process. Undocumented at handover. We would write it down as part of fixing it.
• A working session. You offered to walk through the business process flow and open items, and sent the exchanges plan outline. Yes to both.