Skip to main content
Card class: Hero • Category: Shipping & Courier

​
At a glance

Absolute count of Australia Post consignments where the delivered scan landed after the published estimated delivery window for the service tier, in the trailing 7 days. The count is the operational workload behind the On-Time Delivery Rate percentage. A 95 percent OTD on 6,800 weekly parcels is 340 late deliveries, which is roughly 340 customer-service tickets in flight.
What it countsCOUNT(shipments WHERE delivered_at > estimated_delivery_date) over the trailing 7 days. Deliveries past 14 days roll into RTO/lost territory, see Returned to Sender.
API endpointGET /shipping/v1/shipments and GET /shipping/v1/track/items/{trackingNumber}. Reads service_code, consignment_id, estimated_delivery_date, delivered_at. In-transit consignments not yet delivered are excluded; only consignments with a final delivered scan and a missed aim count.
Service-tier scopeAll tracked services (Express Post, Parcel Post, eParcel Premium, StarTrack). The aggregate is the merchant’s total workload; per-tier splits live in siblings.
Geographic OTD varianceLate counts skew heavily to remote and regional postcodes where transit-time variance is widest. A merchant with 15 percent volume into outback lanes typically sees 35 to 50 percent of late shipments coming from that 15 percent of volume.
Returns / RTOOutbound only. Returns leg is excluded.
Climate handlingCyclone, bushfire, flood events spike the count for affected weeks; not auto-excluded. The card reports the actual surge so the operations team can staff customer service accordingly.
Peak-period seasonalityLate counts roughly 2 to 3 times normal during the 1 to 23 December peak window. Click Frenzy and Black Friday weeks add a 30 to 50 percent surge for 7 to 10 days.
Time window7D (rolling 7 days). The 7-day window is the typical customer-service ticket review cadence.
Alert trigger>5% of total weekly shipments. For a 6,800 parcel/week merchant, alert fires at 340 late shipments.
Rolesowner, operations

​
Calculation

Calculated automatically from your Australia Post data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.

​
Worked example

The same Brisbane outdoor adventure brand from the OTD Rate example, 6,800 outbound parcels per week, mixed Australia Post service mix. Reading taken at 09:00 AEDT on 12 Mar 26 for the trailing 7 days (06 Mar 26 to 12 Mar 26).
Service tierShipments (7D)LateLate %
Parcel Post3,7402526.7%
Express Post2,040753.7%
StarTrack Premium680162.4%
eParcel Premium34092.6%
All AP tracked (this card)6,8003525.2%
The card reads 352 late shipments for the trailing 7 days. The alert at >5% of total is just tripped (352/6,800 = 5.2 percent). Three observations:
  1. Parcel Post is the workload driver. 252 late Parcel Post shipments are 71 percent of the total late count, despite Parcel Post being only 55 percent of volume. The customer-service queue this week will be dominated by Parcel Post WISMO (“where is my order”) tickets.
  2. The remote tail compounds during wet weeks. Splitting by destination, 134 of the 352 late shipments (38 percent) went to remote/outback postcodes which are 15 percent of volume. Tropical Cyclone activity north of the Tropic of Capricorn this week added roughly 80 of those 134.
  3. Express Post late count is structural baseline. 75 late Express Post shipments per 7 days is the level for this merchant outside peak. It comes from non-Express-Post-Network postcodes paying the premium but receiving 2 to 4 day delivery, not from network failure on the metro corridors.

​
Sibling cards merchants should reference together

CardWhy pair it with Late ShipmentsWhat the combination tells you
On-Time Delivery RateThe percentage view of the same population.Percentage tells you the customer-felt rate; this count tells you the operations workload.
Exception RatePredictor at 24 to 72 hour lag.Spike in exceptions today predicts a spike in lates over the next few days.
OTD by RouteWhere the late shipments are going.Confirms whether the spike is regional or network-wide.
Avg Transit (days)Mean transit creep is a leading indicator of late count.A 0.5 day transit increase usually precedes a doubling of late count within 7 days.
Failed DeliveriesDifferent failure mode (attempted but not delivered).Distinguishes “couriered late but landed” from “couriered on time but no one home”.
Cross-connector: shopify.unfulfilled_ordersUpstream input. Slow fulfilment makes late count worse.A backlog 2 to 4 days ago is now a late-shipment surge today.
Cross-connector: shopify.refund_rateDownstream consequence.Late count spike at day 0 typically drives a refund rate spike at day 7 to 14.

​
Reconciling against the vendor’s own dashboard

Where to look in Australia Post’s own portal: Australia Post Business → MyPost Business → Reports → Shipment Status → Filter: delivered late. Larger eParcel customers find the same view in eParcel Customer Centre → Performance → Late Shipments Report. Why our number may legitimately differ from Australia Post’s report:
ReasonDirectionWhy
Timezone (AEDT/AEST/AWST)Boundary day offPortal reports in account timezone; card stores UTC. A late shipment delivered at 23:30 AEDT shows different “day” depending on which side you read.
Customs lag for internationalOurs higherInternational EMS consignments awaiting customs clearance overseas can sit “in transit” for 5 to 14 days; the portal sometimes pauses the SLA clock during customs, the card does not.
Climate-driven exception spikesEitherPostcodes under official network impact notification (cyclone, flood) are sometimes excluded from the portal’s “late” tally; the card always counts them.
Peak-period batch processingEitherNovember to December the portal can show a 5 to 7 day lag in the late-shipment tally while batches catch up; the card surfaces what the API has actually returned.
In-transit-but-already-lateOurs lowerThe card counts only consignments with a final delivered scan past aim. The portal sometimes flags in-transit consignments past their aim as “late” already; the card calls those “exceptions” not “late”.
Cross-connector reconciliation:
CardExpected relationshipCauses of legitimate divergence
shopify.unfulfilled_ordersUpstream input.Webhook delays, B2B/pre-order flows.
shopify.refund_rateDownstream sentiment.Refund rate has many drivers.

​
Known limitations / merchant FAQs

The OTD Rate dial says 95 percent and the late count says 352 in 7 days. Which do I act on? Both, for different teams. The percentage is the merchant-facing dial that tells you whether the customer experience is acceptable. The count is the customer-service workload that tells you how many WISMO tickets are about to land. A team running customer support will care about the count; a team negotiating service credits with Australia Post will care about the percentage. Why is the late count so much higher in December and January? December is the Australian Christmas peak (3 to 4x normal volume); January slows but the carry-over from late-December backlog can keep late counts elevated for the first 2 weeks of January. Australia Post protects Express Post capacity but Parcel Post compresses 4 to 8 OTD points during peak, which doubles or triples the absolute late count. My MyPost Business portal says 290 late and Vortex IQ says 352. Why? Two common reasons: (1) the portal sometimes excludes postcodes under network impact notifications during cyclone/flood; the card never excludes; (2) the portal classifies in-transit-past-aim consignments as “late already” while the card waits for the delivered scan and counts them at delivery. The two should agree to within 5 to 10 percent on a 7-day window outside weather events. How do I split out climate-related lates from process-related lates? Use OTD by Route. Climate-driven lates are concentrated in tropical postcodes (4XXX, 4870 to 4895, 0800 to 0899) and Western WA; process-driven lates are spread evenly across the network or concentrated in your highest-volume metros. If your late count is up 30 percent and the increase is all in northern Queensland, it is the wet season; if it is uniform, it is your warehouse or Australia Post’s network. Customs adds days to my international EMS shipments. Are those counted as late? Yes, against the published EMS aim-by date. Australia Post’s EMS aim is destination-country-specific (3 to 5 business days to NZ, 5 to 8 to UK/EU, 5 to 10 to US); customs clearance time is included in the aim. If a parcel sits in destination customs for 14 days, it counts as late on this card. International is typically 5 to 15 percent of late count for cross-border-active merchants. StarTrack is barely showing in my late count. Should I move more volume to it? StarTrack’s network is structurally faster but only on lanes its fleet covers (mostly metro and major regional). Moving non-metro volume to StarTrack does not gain reliability because StarTrack subcontracts those lanes back to Australia Post. The cost premium is 60 to 100 percent over Parcel Post; the OTD gain on metro is 4 to 6 points. Worth it for high-value or B2B; not worth it for low-margin DTC. Why does the count spike on Mondays? Weekend processing. Australia Post does not do Saturday or Sunday delivery for Parcel Post; consignments scanned in Friday afternoon often deliver Tuesday rather than Monday, which means anything with a Monday aim arrives Tuesday and counts as late. The card honours the published aim, the spike is the structural artefact of the M-F delivery cadence. Tighten checkout copy to set Monday-after-Friday-order expectations. My alert just fired at 5.2 percent. What should I do first? (1) Check Australia Post’s service updates page for any active network event. (2) Check OTD by Route for geographic concentration. (3) Check Exception Rate for upstream label-data issues. If all three are clean, the spike is likely organic peak surge; the action is to staff CS for the WISMO wave and tighten checkout copy. If any of the three is hot, address that first.

​
Tracked live in Vortex IQ Nerve Centre

Late Shipments is one of hundreds of KPI pulses Vortex IQ tracks across Australia Post and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. Start for free or book a demo to see this metric running on your own data.