AliExpress and Correios Add Brazil Capacity for Singles Day and Black Friday

The Analysis

I had three messages about this one in my inbox before lunch on 13 October, and all three were asking roughly the same thing: AliExpress has signed something new with Brazil's post, so should we push more stock into Brazil for the fourth quarter? I get the reflex. When a platform tells you it has done something about delivery, the automatic reaction is to plan your year around it. I have also been at this long enough to know the difference between a press line and a commitment you can safely build a purchase order on, and this announcement sits much closer to the first than the second.

Here is what was actually reported, and I am going to be strict about sticking to it. As carried by Xinhua out of Sao Paulo on 12 October, AliExpress and Correios deepened their existing partnership, adding a memorandum of understanding aimed at the two biggest peaks of the year, Singles Day on 11 November and Black Friday on 27 November. The measures named are expanding parcel processing capacity and improving order tracking. Correios president Emmanuel Rondon said the post is prepared for the year-end parcel volume and talked about an integrated network connecting Brazil to global e-commerce demand. The cooperation also sits inside Correios' restructuring programme, which is explicitly leaning on cross-border growth to justify modernising the operation. That is the whole of it. No capacity number was published. No delivery-day target was published.

Sixteen days is what separates those two shopping events, and that constraint is going to drive every decision below a lot harder than anything said in the press release.

Let me deal with the elephant in the room first: the social-media version of this story reads like Brazil's delivery capacity just doubled. It didn't say anything close to that, and if you plan against a doubling you will be planning against a number you invented yourself. What you have here is two organisations agreeing to work together more closely on two specific functions. Both of those functions matter enormously in practice. Neither of them came with a unit of measurement attached. Processing capacity could mean anything from a new sorting line to a marginally longer shift pattern at peak; better tracking could mean scan events every twelve hours instead of every two days. Both are genuinely good news for anyone shipping consumer parcels into Brazil, because those are precisely the two places where Brazil-bound volume tends to come apart in December.

I have watched this film in at least two other markets, and it does not usually end the way people hope. A platform and a national post sign something in October, everybody quietly revises their expectations upward, and by the middle of December we are arguing about scan gaps and exception reports instead of counting sales. The trap is not that the announcement is meaningless. The trap is that it arrives exactly when you have to commit money, so the optimistic reading becomes the default one. My own rule is simple: nothing changes in my safety stock or my lead-time assumptions until I have watched the lane perform for at least one full cycle with my own batch numbers on it.

To understand what this means for you as an importer or exporter of these parcels, you have to separate the three things that usually get bundled together as "shipping to Brazil". There is the line-haul leg out of China, there is the clearance and sorting phase on arrival, and there is the last mile to the buyer's door. Costs and risks in those three phases behave completely differently in the fourth quarter. Line-haul pricing moves with the wider market and you can negotiate it directly with whoever consolidates for you. The arrival phase is where Brazil-specific friction lives, and it is also the part this announcement is aimed at. The last mile is driven by geography, by the density of your own parcel flow, and by how predictable your addressing data is.

That middle phase is where I would put most of my attention this year. In my experience on Brazil-bound flows, the origin leg is rarely the bottleneck; what decides whether a November campaign is profitable is how fast a containerised or bagged parcel batch gets picked up, cleared, sorted and handed to delivery once it lands. Anything that genuinely adds processing capacity there is worth real money, because every day a parcel sits in that queue is a day of working capital you will not get back and potentially one cancellation you cannot recover. But that is also a claim I want evidence for, and right now there is none in the public domain.

The compliance angle deserves a paragraph of its own, because it is the thing most sellers forget until it bites. When throughput triples in a fortnight, the number of people examining shipments does not triple. What does change is the tolerance for sloppy paperwork. Make sure your commercial invoice descriptions, your HS classifications and the values declared on the platform all tell the same story, and make sure they tell it consistently across every shipment in the batch rather than just the ones you think will be inspected. Brazil checks recipient identification and tax details on incoming parcels, with individuals normally identified by CPF and companies by CNPJ, though you should have your broker confirm the current requirement for your exact product category rather than trusting a blog post. A mismatch rate of even two percent across fifty thousand parcels is a thousand parcels sitting in a bonded area accumulating storage while your customer refreshes a tracking page.

Now the part everybody skips: transit time and the calendar. Sixteen days between 11/11 and 27/11 is not a lot of room, and door-to-door transit on China-to-Brazil parcel channels is measured in weeks, not days. Nothing you hand over on 12 November is going to reach a Brazilian buyer before Black Friday. Nothing you hand over on 12 October with slow economy service is going to reach them before Singles Day either. That means the inventory for both events has to already be on the ground, or already close to it, well before either campaign starts. This is the single most important operational consequence of the announcement, and ironically it is the one the announcement itself says nothing about.

Let me put the arithmetic on the table, with every assumption visible so you can swap in your own figures. Assume you are planning fifty thousand orders against Singles Day, and your average finished pack weighs 320 grams once it is boxed and labelled. That is sixteen thousand kilograms of actual weight. Assume your channel bills on a volumetric divisor of 6,000, and assume your typical carton dimensions produce a volumetric weight around 380 grams per piece, so you are actually paying for nineteen thousand kilograms rather than sixteen. Assume your all-in parcel rate to Brazil is 92 yuan per kilogram of chargeable weight; that figure is my own working number, not a quoted market rate, so replace it with what your forwarder actually charges you.

Run that out and the freight line for the Singles Day wave alone comes to about 1.75 million yuan, which is roughly 243,000 dollars at an assumed 7.2 exchange rate. Notice how much of that number is packaging: the three thousand kilograms of difference between actual and volumetric weight costs you about 276,000 yuan, or around 38,000 dollars, and it did not buy you one additional unit sold. That is the first line item I would attack this week, long before I argued about the rate.

Assume further that your Black Friday wave runs at around sixty percent of your Singles Day volume, which is thirty thousand orders. At the same 380 grams of average chargeable weight, that is eleven thousand four hundred kilograms, which comes to roughly 1.05 million yuan or about 146,000 dollars. Taken together you are looking at something like 2.8 million yuan of freight, close to 389,000 dollars, to move eighty thousand parcels through two peaks sixteen days apart. Sellers rarely write that number down before the campaign starts, and that is usually why the margin report in January is a surprise.

Now put timing into the same model, because timing is what will actually hurt you. Assume your economy channel takes thirty days from pickup to being available for dispatch inside Brazil, and assume your expedited channel takes eighteen days. Both figures are planning assumptions I use when I have nothing better, and both should be replaced with your own last ten batches' actuals. Counting from 13 October, you have twenty-nine days to Singles Day and forty-five days to Black Friday. On those assumptions, anything shipped on the economy channel after about 12 October simply cannot make the 11/11 campaign, which means that window closed yesterday. On the expedited channel you have until roughly 24 October. For Black Friday, the economy channel cuts off around 28 October and the expedited one around 9 November.

Here is the decision those dates force, and it is worth roughly fifty thousand dollars. Assume the expedited channel carries a thirty-five percent premium over the economy rate, again from my own planning assumption rather than from anyone's published tariff sheet. If you miss the 28 October economy cut-off for your Black Friday stock, moving those 11,400 kilograms on the fast lane instead costs you about 367,000 yuan more, which is roughly 51,000 dollars. Compare that against the cost of simply shipping early in the first place: those thirty thousand units at an assumed landed unit cost of 42 yuan tie up about 1.26 million yuan, and at an assumed monthly carrying cost of 1.5 percent that is roughly 18,900 yuan a month to hold them thirty days longer, call it 2,600 dollars. Paying 2,600 dollars to avoid a 51,000 dollar bill is not a close call. Ship early, every time, unless your cash position genuinely cannot take it.

So what do I actually want you to do, and by when? Starting today, and this should land with whoever owns your channel relationships rather than sitting in your inbox. Before 16 October, ask your parcel consolidator or forwarder for a written all-in quote valid through 30 November, and insist that three things appear on it: the volumetric divisor used, the start and end dates of any peak-season surcharge, and whether that surcharge is billed on actual weight, chargeable weight, or per piece. Those three fields are where November invoices go wrong far more often than the base rate does. If they will not put it in writing, you have learned something useful about how disputes in December will go.

During the week of 20 October, freeze your allocation split. I would work with sixty percent of your fourth-quarter units aimed at the Singles Day wave and forty percent at Black Friday, adjusted for whatever your own category actually does, and I would want at least eighty percent of the Black Friday units physically tendered before 28 October. Mark 24 October on the wall as a hard internal cut-off for anything intended to fulfil 11/11 orders on the expedited lane, with your shipping team targeting ninety-five percent tendering two days ahead of it rather than scrambling on the day itself.

Then make the tracking promise measurable rather than rhetorical. The deal explicitly mentions better tracking, so use it: require a first scan within forty-eight hours of pickup on at least ninety-five percent of parcels, and require a weekly exception report listing every parcel silent for more than five days. When you send that requirement, ask exactly which scan events will be visible and at which facilities, and get the answer in writing. A commitment to improved visibility that nobody can specify is worth very little when you are trying to answer a customer email.

There is also a customer-facing decision hiding in here. If you have been quoting delivery windows off best-case scenarios, widen them now. A storefront promise of twenty-five to thirty-five days is honest for cross-border Brazil at this time of year; promising twelve to eighteen because you read encouraging news will generate a wave of "where is my order" tickets that cost more in service time than the extra conversions are worth. And tag every order with its batch or flight identifier, so that when you sit down in January you can measure this particular channel against your alternatives instead of guessing.

On alternatives, this is roughly how I price the options. Sticking with the platform-post channel keeps you closest to what your buyer already recognises, and after today's news it is likely to be the best-managed version of itself, but it still leaves you exposed to exactly the same arrival bottleneck. A commercial parcel consolidator gives you better control of handover timing and usually clearer accountability, generally at a modest premium. Forward stocking into a Brazilian fulfilment location is the real escape from the queue, and it flips your economics entirely, but it only starts paying for itself above a fairly high monthly throughput — my rough working threshold is around eight tonnes a month of consistent volume, below which you are mostly paying rent for empty shelf space. Sea-parcel services are far cheaper per kilo and useless for either of these two events, though they are perfectly sensible for rebuilding stock in January and February when nobody cares about a forty-five day transit.

There is a pricing question hiding behind all of this that most sellers answer by accident. If your freight per kilogram in the fourth quarter is materially higher than it was when you set your autumn prices, and especially if you have been absorbing the difference to stay competitive in the cart, then you are not running a promotion, you are running a subsidy. I would rather see you raise the price by five percent and lose a little volume than ship eighty thousand parcels at a contribution margin you never signed off on. Work the number before you ship rather than after: take the switched 389,000 dollars above, divide it by the eighty thousand units, and you get roughly 4.9 dollars a unit of freight sitting inside a product that probably retails somewhere between fifteen and thirty. That single figure tells you whether your campaign economics work at all, and it is worth writing on the whiteboard before anyone books space.

Now the traps, which is where most of the money actually is this quarter. Do not treat a memorandum of understanding as a service contract; these documents almost never carry a compensation clause, so if the bags sit for nine days in November you will have nobody to claim against. Do not accept an internal target built on "the capacity doubled" when no such number was published. Check that the volumetric divisor quoted in pre-sales matches the divisor on the invoice, because they drift apart surprisingly often when volume spikes and the sales desk is quoting from memory.

Confirm whether the peak surcharge applies to bookings accepted before the start date but flown after it, since that one clause has a habit of retroactively repricing your entire October plan. Keep invoice values and platform declared values identical across the batch rather than quietly optimising some lines, because consistency is what gets you through a busy clearance window. And check whether your density numbers are real: if 380 grams of chargeable weight comes from a carton you could compress by five centimetres, you are paying tens of thousands of dollars for air.

My own position after running all this through is fairly straightforward. I believe this cooperation is real work being done by people who know what breaks in December, and I believe Brazil is a market worth investing inventory in for the long run. What I will not do is move my cutoff dates or trim my buffer stock on the strength of an announcement containing no figures. If you must lean optimistic somewhere, lean optimistic on sales and pessimistic on logistics; the cost of being wrong in one direction is unsold inventory you can discount, while the cost of being wrong in the other is fifty thousand dollars of expedited freight and a lot of very annoyed customers. I will be watching my own batch scan data every Monday from here to Christmas, and I will be very happy to be pleasantly surprised.

↑ Back to the key points

— By Leo