Transactional Order Messaging Integration

Broadly, a transactional message conveys factual information about something the customer has already done, such as confirming an order, advising dispatch with tracking, notifying that a collection order is ready, or confirming a refund. Marketing promotes products or offers. The distinction matters because commercial electronic messages carry obligations under the Spam Act 2003 including consent, sender identification and a functional unsubscribe facility, and adding promotional content to an otherwise factual message can bring it within that scope. Since the boundary depends on the actual content of your templates, it is worth having them reviewed rather than assuming a dispatch email is automatically exempt.

Whichever one you decide, provided it is exactly one per message type and the decision is documented. The common and sensible pattern is that the commerce platform owns pre fulfilment messages such as order confirmation, the fulfilment or warehouse system owns dispatch and delivery updates because it knows what actually shipped, and the point of sale or store system owns collection readiness. What matters far more than the specific allocation is that duplicates are switched off at source. Two systems both sending a dispatch notice with slightly different wording is worse than either one doing it alone.

Explicitly, in the message design, rather than letting each shipment generate an identical looking notice. The customer needs to know what has shipped, what has not, and when to expect the rest, ideally in one coherent message per dispatch that references the whole order. Batching helps where parcels leave within a short window, so three shipments in an hour produce one update rather than three. Where a line is short supplied or backordered, tell the customer before or at dispatch rather than letting them discover it when the box arrives, because that is the point at which a minor issue becomes a complaint.

Both, chosen by message type rather than picking one. Email suits detail: order confirmations with line items, tax invoices, dispatch notices with tracking links. SMS suits time critical and short messages where immediacy matters, most obviously collection readiness, delivery windows and delivery attempt notifications. Sending everything by SMS is expensive and quickly feels intrusive, while sending time critical messages only by email means they are read too late to be useful. Capture customer preference where practical, and be mindful of sending times, since an automated SMS at two in the morning is technically correct and commercially unwise.

It depends on the message. A hard bounce on an order confirmation is operationally significant, because that customer has no record their order exists and may dispute the charge later, so it should raise a task for someone to make contact another way. A soft bounce is usually worth retrying. An invalid mobile number should update the customer record so it is corrected rather than repeatedly failing. The general principle is that failures on operationally important messages need a person, while the rest need monitoring in aggregate, and neither should end up in a provider log that nobody opens.

Yes, in the sense that routing messaging through a platform means customer contact details, order contents and delivery addresses now exist there too. That is personal information you hold under the Privacy Act 1988 and the Australian Privacy Principles, so it needs to be secured, covered by your privacy policy, and locatable and deletable when someone requests it. Practical steps are to send the minimum each message genuinely requires rather than the full order record, set a deliberate retention period on message logs, ensure the provider arrangement is appropriate for where the data is held, and make sure deletion requests reach the messaging platform as well as your core systems.

We will not quote a figure without looking at your data, and any number quoted without that is invented. What is consistent is that order status enquiries are among the most common contacts for retailers, and a large share of them are people who were not told something they needed to know, or were told something that turned out to be inaccurate. The useful approach is to categorise a sample of recent contacts, identify which were preventable with a timely accurate message, and use that as the baseline. It also tells you which messages to fix first, which is usually more valuable than the headline number.