The six-step flow, and why most of it is your existing Tuesday
In a parcel hub, every unit has a cut-off time. Miss it and the parcel loses a full day, and the customer gets an email about it. Nothing in forwarding runs to a clock that unforgiving, and the rest of the operation is more familiar than that fact makes it sound.
A parcel network is six steps. You already run four of them.
1. Injection
Volume arrives from the shipper, collected by you or dropped at your site, with a manifest that has to match what physically turned up. You do this now. What changes is frequency and file size: instead of a handful of bookings a day you're ingesting several thousand consignee records overnight, and the addresses are residential, so more of them are incomplete, duplicated or simply wrong.
2. Linehaul into the hub
Trunking full loads from origin to a central facility. This is forwarding. Nothing about it changes.
3. Sortation
The first step that isn't your Tuesday. Every unit gets separated, scanned, weighed, dimensioned and directed to an outbound lane. Whether you buy a machine or rent someone else's capacity is a capital question with well-trodden answers, and most forwarders lease in rather than build in year one.
The harder part is the discipline the machine imposes. Everything upstream is scheduled backwards from the outbound cut-offs, which leaves no slack anywhere in the chain to absorb a problem.
What that looks like in practice: a trunk from Manchester arrives forty minutes late on a Tuesday night. The Scotland lane has already cut and gone. Nobody can hold it, because holding it delays six other lanes behind it. So 900 parcels sit on the floor until tomorrow's sort, they arrive on Thursday instead of Wednesday, and the first person to know is not your night manager. It's the retailer's customer service team on Wednesday afternoon, taking calls, wondering why their carrier hasn't told them anything. Your account manager finds out from the client.
A late truck in forwarding costs you a few hours and a phone call. Here it costs a day and a service credit.
4. Trunking to depot
Outbound linehaul from hub to delivery unit, planned around volume forecasts and vehicle fill. Forwarding again, with tighter windows.
5. Last mile
The second new step, and the one that decides whether the P&L works. Last mile runs at 40% to 50% of total distribution cost for carriers. On a 100-drop round, going from 12 to 14 drops an hour takes roughly an hour out of the day, which is around 5% off your largest cost line. That's the whole game, repeated across every round you run.
Density is what you're buying. Failure is what erodes it, and the published data on failure needs care. First-attempt rates for home delivery run anywhere from 2% to 25% depending on market and carrier, largely because nobody agrees what a failure is. A parcel left with a neighbour counts as delivered in one dataset and missed in another. You'll also see a per-attempt cost of around £11.60 quoted widely in UK trade coverage, including here. Treat it as directional. A precise figure built on a definition that loose won't survive a board asking how it was derived, so model your own from your own round data first.
The cost per attempt also misses what happens afterwards. Third attempt, house still empty, parcel returns to depot. The consumer then contacts the retailer, because you're invisible to them, and that call lands on a support team whose contact volume is a metric someone senior watches monthly. Enough of it and your service level becomes a procurement conversation at the next tender.
Out-of-home collection is the pressure valve, and the direction of travel is clear. Europe now has over 646,000 active collection points, locker networks have grown 77% in two years, and CEP Research put out-of-home volumes up over 20% to more than two billion parcels in 2024, against 5% growth in home delivery. Geopost's shopper research has 46% of European shoppers now preferring it.
The honest downside: out-of-home doesn't remove cost, it relocates it. You pay a commission per parcel to the shop or the locker operator, or you fund locker capex yourself and wait for utilisation to catch up. Capacity is finite, and lockers fill in peak week, which pushes volume back to the door at the exact moment your rounds are worst. Bulky, age-restricted and high-value items still need a doorstep regardless. For a forwarder with no doorstep coverage, out-of-home gets you to a serviceable footprint faster than building rounds does. The economics over ten years are a separate question.
6. Exceptions and the reverse flow
You handle exceptions today at a volume you can work through by phone. Parcel exceptions arrive in the thousands, each one attached to a consumer refreshing a tracking page, and you have hours to resolve before it becomes a complaint. That's mostly a resourcing problem. You need people on it during the hours consumers are awake, which isn't the shift pattern a forwarding operation runs.
Returns are the part most forwarders under-price. It's a permanent second network running the opposite way, with its own collection, its own transport, and a set of decisions at the far end that forwarding never asks you to make. Every returned item needs grading and a disposition call. Back to stock, back to the vendor, discount channel, or scrapped. Get that wrong and you're either warehousing worthless stock or writing off resaleable goods.
The bit that catches people is timing. A return that isn't scanned and credited quickly generates a support contact, and that contact costs the retailer more than the return leg did. Returns handling is something retailers compete on now, so your client will be measured on it whether or not it appeared in the tender.
What the audit says
Four of six steps are capability you already own. Sortation is a capital and technology decision with partnering routes that work. Last mile is the one thing you build, learn and staff from nothing, and it's where the margin lives.
So the question isn't whether you can run a parcel operation. It's what the two unfamiliar steps cost, and whether the volume you can already see justifies them. That's an arithmetic problem, and it's a better one to solve now than in front of a board.
Working out what the build actually costs? Our Pallets to Parcels e-book maps each of the six steps against what a forwarder already has, and prices the two you don't.
