If you're building a Black Friday strategy for an Australian ecommerce brand this year, start with one number: contribution margin per order after Afterpay, after shipping, after GST. Not revenue. Not units. On a typical $120 order at 30% off, with a 5% BNPL fee and a real parcel cost, you're keeping around $25 β less than half what the same order returns at full price. Spend $45 to acquire that customer and the order loses money. It only turns positive if 38% of that cohort comes back and buys again at full price within the year. Most discount-acquired cohorts don't. That's the whole argument, and the rest of this post is the arithmetic behind it.
Click Frenzy has already come and gone, which means a chunk of November intent is spent and your competitors' promo emails have trained the list to wait. Black Friday lands 27 November. You're setting depth now, probably in a spreadsheet with a column labelled "what did we do last year". Replace that column.
Start from contribution margin, not from what the brand next to you is doing
Here's the full working on a single order. Numbers are illustrative, but the structure is the one we use when we sit down with a founder in October.
- AOV: $120 including GST
- GST component: $120 Γ· 11 = $10.91, so ex-GST revenue is $109.09
- Landed COGS: $33
- Payment fee: 5% + $0.30 on the GST-inclusive total = $6.30
- Shipping plus pick and pack: $14
Contribution = 109.09 β 33 β 6.30 β 14 = $55.79.
Now take 30% off. The order bills at $84. Ex-GST that's $76.36. COGS doesn't move β it's still $33. The payment fee drops slightly to $4.50 because it's a percentage. Shipping doesn't move at all.
Contribution = 76.36 β 33 β 4.50 β 14 = $24.86.
A 30% discount removed 55% of your contribution. That asymmetry is the single most misunderstood thing about BFCM pricing. COGS and freight are fixed in dollars; your discount only comes out of the margin layer, so the percentage off the sticker price is never the percentage off your profit. It's roughly double.
What Afterpay and Zip actually cost on a discounted order
BNPL merchant fees in Australia sit in a band β a percentage of the transaction plus a fixed cents component, negotiated against volume. Check your own contract rather than trusting a blog post, including this one. The structural point holds regardless of where you land in the band: the fee is charged on the gross, GST-inclusive amount, which means you pay the provider on the 10% you're about to hand to the ATO.
On the $84 discounted order above, $4.50 of fee against $24.86 of contribution is 18% of what you keep. At full price it's 11%. Discounting makes every fixed and percentage cost heavier relative to margin, and BNPL is the one most Australian merchants have the least visibility on because it's buried in a settlement report rather than itemised against orders.
The fair counterargument: Afterpay and Zip lift AOV, and on apparel and homewares the lift is real. We've seen BNPL orders run meaningfully above card orders on the same store. So don't rip it out. Do model it properly: if 40% of your BFCM orders go through BNPL at 5%, that's a 2% haircut on total gross revenue sitting outside your Shopify payouts line, and it needs to appear in the promo P&L before you sign off on depth.
Shipping to WA is real money, and free-shipping thresholds are usually set wrong
Most Australian stores set a free-shipping threshold using Sydney-to-Melbourne economics, then eat the difference on everything going west or regional. A 2kg parcel to a metro east coast suburb might cost $11 on a Sendle or Australia Post business account. The same parcel to a regional WA postcode can run north of $20, sometimes well north if it's bulky and dimensional weight kicks in.
Rerun the discounted order with $23 of freight instead of $12:
76.36 β 33 β 4.50 β 25 (freight plus pack) = $13.86.
Half the contribution of the same order shipped to Brunswick. If that order also carried $45 of paid acquisition, it's $31 underwater and needs a 59% repeat rate to justify itself. Nothing justifies that.
Two practical fixes before 27 November. Split your carriers by zone rather than picking one β Sendle is often cheaper metro-to-metro, Australia Post holds up better to remote postcodes, and Shopify handles rate logic by postcode fine without a Plus plan. And set your free-shipping threshold from your blended parcel cost, not your metro one. If 12% of your orders go to WA, NT or regional Queensland, your blended cost is higher than the number in your head, and a $99 threshold that works on paper bleeds on the tail.
The repeat-rate floor: the number that decides whether acquisition was worth it
Take the discounted order again. $24.86 contribution. Assume $45 to acquire the customer through paid social in late November, when every advertiser in the country is bidding against you for the same inventory. First order: β$20.14.
A second order at full price with no acquisition cost, less maybe $3 of email and SMS attribution, returns $52.79. To cover the $20.14 hole you need:
20.14 Γ· 52.79 = 0.38. Thirty-eight per cent of that cohort has to buy again at full price inside twelve months.
Now run it at 15% off instead. The order bills at $102, ex-GST $92.73, payment fee $5.40. Contribution = 92.73 β 33 β 5.40 β 14 = $40.33. Less $45 CAC = β$4.67. Required repeat rate: 4.67 Γ· 52.79 = 9%.
Halving the discount took the break-even repeat rate from 38% to 9%. That is the trade you're actually making when you go from 15% to 30%, and almost nobody states it in those terms in the planning meeting.
What's a realistic repeat purchase rate benchmark for D2C?
We're not going to quote an industry figure, because the ones circulating are mostly unsourced and the honest answer is that it varies enormously by category. A supplements brand and a furniture brand have nothing to say to each other on this. What matters is your number, measured on the right cohort.
Pull it yourself. In Shopify admin, isolate customers whose first order date falls inside your 2024 BFCM window, then measure what share placed a second order at 90, 180 and 365 days. Do the same for a cohort acquired in a non-promo month. In the work we do, the discount-acquired cohort almost always underperforms the full-price cohort on second-order rate, and it underperforms again on second-order AOV, because the first purchase anchored a price the customer now considers normal. If your discount cohort clears 38%, deep discounting is defensible for you. If it sits in the teens, you've been buying revenue with margin for several years and calling it growth.
One more thing to check while you're in there: how many of your BFCM "new" customers were on your email list already. Those aren't acquisitions. They're existing demand that waited for a code.
Is Black Friday worth it for small Australian brands?
For a lot of them, no β not as an acquisition event. It's worth it as a retention and AOV event.
The reason is population. Australia has roughly 27 million people. Your actual addressable segment might be a few hundred thousand. Paid acquisition in a market that size hits diminishing returns fast, and it hits them hardest in the week when every brand in the country is competing for the same impressions. You're not discovering new buyers in late November. You're paying a premium to reach people you'd have reached in February for less.
So if you're under a few million in revenue, point the offer at your list and your past customers. The economics are completely different: no acquisition cost, higher baseline conversion, and you can afford depth on a known buyer that you cannot afford on a cold one. A 30% offer to a lapsed customer with $0 CAC contributes $24.86 and needs nothing from the future to be worth doing.
Where Black Friday genuinely earns its keep for a small brand: clearing slow-moving SKUs you'd otherwise mark down in January anyway, and lifting AOV through bundles rather than straight percentage cuts.
How deep should the discount go
Work backwards from the floor rather than forwards from a round number. Decide the minimum contribution per order you'll accept, then solve for the discount that produces it. For the store in our example, holding $40 of contribution means capping the discount at about 15%.
Then stop discounting and start restructuring the offer. Things that move AOV without touching unit margin:
- Tiered thresholds β spend $150, get a gift; spend $220, get a bigger one. The gift costs you COGS, not margin percentage, and it pulls basket size up rather than price down.
- Bundles priced at a discount that only exists at the bundle level, so the individual SKU price never gets reset in the customer's head.
- Free express upgrade over a threshold. Costs you $6 of freight, reads as worth $15, and solves the pre-Christmas delivery anxiety that drives a lot of November purchasing in Australia.
A note on comparison pricing. Under Australian Consumer Law, a struck-through "was" price has to reflect a price the product was genuinely sold at for a reasonable period beforehand. Inflating RRPs in October so the November discount looks deeper is the kind of thing the ACCC takes an interest in, and it's not worth the exposure for a few extra percentage points of perceived saving. The same scrutiny applies to any sustainability or provenance claim you dress the sale in.
Black Friday or Boxing Day: which fits the Australian calendar better
Boxing Day is still the bigger cultural retail moment here, and for several categories it's the better-margin one. Three reasons worth weighing.
Advertising inventory gets cheaper once the global BFCM bidding war ends. The delivery promise disappears β nobody buying on 26 December expects it before Christmas, so you're not paying for express freight to protect a cutoff date. And the buyer is already in holiday spending mode, often with gift money, in a summer break that runs to late January. The northern calendar treats late December as the dead zone after the peak. Australia's is inverted: it's the start of a six-week window where people are off work, at the beach, and on their phones.
The cannibalisation problem is the other half of it. Black Friday sits four weeks before your Christmas peak. Every unit you discount on 27 November is plausibly a unit you'd have sold at full price on 12 December. In our experience that overlap is largest in apparel, accessories and homewares, and smallest in consumables where purchase timing is driven by running out rather than by gifting.
A defensible shape for most Australian D2C brands: a modest, list-led Black Friday at 15β20% with strong bundle mechanics, full price through the Christmas delivery cutoff, then a genuine clearance from 26 December where you take real depth on the inventory you actually need gone.
Spend the money you didn't give away on the site itself
If you hold discount depth at 15% instead of 30%, on 500 orders you've kept about $7,700 of contribution. That buys more than it would in ad spend.
The highest-return place to put it before late November is usually page speed, because every percentage point of mobile conversion applies to all traffic, paid and organic, for the whole of the quarter. Most Shopify stores we audit in October are carrying three or four apps that are no longer used but are still injecting script into every page load, plus a hero image nobody compressed. Getting LCP on a collection page from 3.1s to under 2s is a week of work, not a replatform. If you want to see where your store stands and keep an eye on it through the peak, SwiftStore scans the store, fixes what it can automatically and tracks the score as you add campaign assets. Our team also does this manually as part of speed optimisation work where the theme needs surgery rather than a scan.
What to do in the next two weeks
Build the single-order contribution model above in a spreadsheet using your actual COGS, your actual BNPL rate from last month's settlement report, and your blended parcel cost including the WA tail. Then pull your 2024 BFCM cohort's twelve-month repeat rate out of Shopify. Those two numbers together tell you your maximum defensible discount, and the answer is often five to ten points shallower than what you ran last year.
If the modelling throws up something you don't like and you'd rather have a second set of eyes on it before you commit, our Australian team runs a free store audit that covers promo mechanics, shipping rate logic and checkout performance. Bring the spreadsheet.