Most Shopify merchants find out their shipping setup is wrong in the worst possible order: the customer complains first, the invoice arrives second, and the merchant only understands what happened third. A satchel that should have been a pallet. A "cheap" checkout rate that quietly ate the margin. A delivery to a house that got billed like a business.

That order of operations is the actual product problem shipping software has never solved. Rate-comparison tools got good at asking "which carrier is cheapest for this box?" Nobody built the layer that answers the harder question first: what actually is this box, and where is it actually going?

Pack to the Future's Platform V2 is built to answer that question before the merchant has to. It's what we mean when we call it an AI shipping operating system rather than another rate-shopping plugin — it doesn't just compare prices on a shipment you describe to it, it does the describing.

The problem with "just compare carriers"

Every shipping app promises savings. Almost none of them address the actual source of shipping cost surprises, which is bad input data. A merchant who guesses at a carton size, doesn't know an address is residential, and has no idea their SKU has crossed into freight territory will get an accurate quote for the wrong shipment.

The cost of that gap shows up twice. It shows up in the invoice, when accessorial fees the merchant never priced in — residential delivery, tailgate, overlength — get added after the truck has already been booked. Freight forwarders and couriers publish these as standing surcharge schedules; TNT Australia's own domestic fee schedule lists a residential delivery fee, tail-lift charges that scale with weight, remote-area zone loadings, and dangerous-goods handling fees as separate, stackable line items, not one bundled "shipping" figure (TNT Australia surcharge schedule).

It also shows up earlier than that, at checkout, where under-declared dimensions produce an unrealistically cheap shipping quote that then explodes into a real invoice after the sale — or where the quote is accurate but too high because the default box size was never right in the first place. Baymard Institute's checkout research, drawn from large-scale usability testing, found that extra costs like shipping, tax and fees are the single leading cause of checkout abandonment: 39% of shoppers who abandon a checkout cite unexpected costs, ahead of every other reason including account-creation friction (Baymard Institute, "How to Reduce Cart Abandonment").

Platform V2 exists to close both gaps at once, before the quote is shown and before the truck is booked.

What "AI shipping operating system" actually means

We use that phrase deliberately instead of "AI-powered app" because V2 isn't one feature bolted onto a rate table — it's a decision layer that runs underneath every stage of fulfilment:

An onboarding agent that configures a live shipping stack from a conversation instead of a 12-step settings form, and starts learning the merchant's catalogue in the background from the first minute of install.
A dimension model (Product AI) that estimates packed shipping size from product text — title, description, variant — because most stores have never actually measured what they sell.
A residential-address classifier that flags the single most common source of surprise freight fees before the label is booked, not after the invoice lands.
A cubic-weight and accessorial pricing engine that quotes freight the way Australian carriers actually bill it — chargeable weight, fuel, overlength — instead of the way a rate table wishes it worked.
A checkout quoting layer that has to return an answer inside a strict time budget while still using real dimensions instead of a generic default box.
Two managed carrier networks — Pack Network for parcels, Pack Freight for anything that has outgrown a satchel — so the merchant picks a service level, not a courier.
NearShip and sustainable packaging, because the network decision and the packaging decision are both part of the same cost-and-carbon equation.

Each of those is covered in depth in its own article, linked above and again at the end of this one. This page is the map.

Why now: the market moved

Two things happened at the same time that make an agentic shipping layer make sense in 2026 rather than being a nice-to-have.

First, Australian online retail kept growing into a market where shipping decisions matter more, not less. Australia Post's eCommerce Report 2026 found that 82% of Australian households — 9.8 million of them — shopped online in 2025, spending $82.6 billion, up 14% year-on-year, with online now accounting for 24% of all retail spend. The same report found 69% of shoppers want multiple delivery options at checkout and 26% expect same- or next-day delivery for urgent purchases (Australia Post eCommerce Report 2026). Every one of those numbers raises the cost of a shipping stack that can't tell a residential address from a warehouse, or a satchel SKU from a pallet SKU, at the volume merchants are now running.

Second, the AI capability to actually fix this arrived. Gartner surveyed 509 supply chain leaders globally between July and October 2025 and found supply chain organisations are past the point of treating AI as a novelty: 55% expect agentic AI to reduce entry-level hiring needs, and 51% expect it to drive broader workforce reductions. Gartner VP Analyst Marco Sandrone was blunt about what separates the leaders from everyone else: the best-performing organisations are using it "to reinvent how work gets done," not as a headcount lever (Gartner newsroom, February 2026). McKinsey's research on agentic commerce puts a number on the shift already underway on the demand side too — its analysts size the US agentic-commerce opportunity at $900 billion to $1 trillion in orchestrated revenue by 2030, and $3–5 trillion globally, as AI agents increasingly transact on a shopper's behalf rather than just assisting them (McKinsey, "The agentic commerce opportunity"). Consumer behaviour backs this up on the Australian side too: Australia Post's report found 62% of shoppers already use AI tools somewhere in their shopping journey.

Put together, that's a market shipping faster, expecting more delivery flexibility, and increasingly comfortable with AI making decisions on its behalf. A shipping app that still asks a merchant to type in box dimensions by hand is solving 2018's problem with 2018's tools.

The loop, end to end

Every piece of V2 exists to feed the next one:

Install
  → Agentic onboarding (priorities, carriers, catalogue crawl)
    → Product AI estimates packed dims (SKU → category → LLM → merchant confirm)
      → Checkout quotes with cubic + residential in the price
        → Booking with Pack Network or Pack Freight, fees already applied
          → Dimension memory + freight profiles learn from the shipment
            → Dashboard agent tells the merchant what to do next

Nothing in that loop is a one-off automation. It's a compounding system: every label booked makes the dimension library, the residential model, and the freight profiles a little more confident for the next order — which is a very different thing from a rate-shopping plugin that treats every shipment as a blank slate.

Where to start

If you're setting up a store for the first time, start with how the agentic onboarding actually works and the Agentic Onboarding help doc (or the Getting Started guide). If you're trying to work out why your freight invoices don't match your checkout quotes, the residential fee and cubic weight articles — plus the matching help pages on residential detection and cubic weight — will show you exactly which line items are catching you out. If you're comparing shipping software options for an Australian Shopify store, the multi-carrier network breakdown, Pack Network and Pack Freight, and our existing Australia Post vs CouriersPlease comparison are the places to look next.

Full plan details, including current plan tiers and feature-flag rollout, are on the pricing page.

This article is part of a series on Platform V2. Continue with: Agentic Onboarding · Product AI & Dimensions · Residential Delivery Fees · Volumetric Weight Explained · Checkout Rate Accuracy · Multi-Carrier Networks · NearShip & Sustainable Packaging