Should you offer free delivery

Every guide says offer free delivery over some amount and watch your average order go up. Nobody shows you the subtraction on the other side.

Free delivery isn't free and it isn't marketing. It's a discount, funded out of your margin, and whether it pays depends on one thing: does it move enough small baskets up to cover the shipping revenue you stop collecting from the baskets that were already big? For a lot of small sellers the honest answer is that it's roughly a wash, and you should choose it for the cleaner checkout rather than for the money. For some, it's actively expensive. Here's how to tell which one you are, with a $4.50 shipping charge and a hundred orders.

Three groups of buyers, and only one of them helps you

The mistake is imagining a single customer, the one who adds a second item to reach your threshold. That customer exists. So do two others, and they're the majority.

Group	What happens when you set a free-over threshold	Effect on you
Already above the threshold	Stops paying for shipping. Buys exactly what they were going to buy	Pure loss, every order
Just below, and steps up	Adds an item to qualify	Gain, worth the margin on the extra item minus the shipping
Well below, and doesn't move	Pays shipping as before	No change

Group one is the problem. It's your best customers, and a threshold hands every one of them a discount they weren't asking for. Group two has to be large enough, and step up far enough, to pay for group one.

That's the entire question. Everything below is how to put numbers in it.

Marcus, enamel pins, Chicago

Marcus makes enamel pins and sells them at $12 through Instagram and a shop link. Each pin costs him $3.40 delivered, so the gross margin is $8.60, or 71.7%.

His parcel is a padded mailer: $2.90 for the label, $0.55 for the mailer, and two minutes of his time at $22.50 an hour, so $0.75. True cost $4.20. He charges $4.50.

A normal month, 100 orders:

Basket	Orders	Average basket	Goods	Shipping collected
Under $24	42	$17.50	$735	$189
$24 to $39	33	$30.40	$1,003	$148.50
$40 and up	25	$56.00	$1,400	$112.50
Total	100	$31.38	$3,138	$450

Shipping costs him $420, he collects $450, so delivery runs at plus $30. His gross margin on goods is $2,250.

Now he switches on free delivery over $40.

The sum, done properly

What he loses. The 25 orders already over $40 stop paying $4.50. That's $112.50 a month gone, for nothing. Straight away he's in a hole.

What he gains. Say 12 of the 33 orders in the $24 to $39 band step up, from an average of $30.40 to an average of $43. That's $12.60 of extra goods each, so $151.20 of extra revenue, worth $108.41 at his margin. Those 12 also stop paying shipping, another $54 lost.

Say two of the 42 tiny orders make the jump too, from $17.50 to $41. That's $47 of extra goods, worth $33.70, minus $9 of shipping.

	Amount
Margin gained from step-ups	+$142.11
Shipping revenue lost, already-qualifying orders	-$112.50
Shipping revenue lost, stepped-up orders	-$63.00
Net	-$33.39

Thirty-three dollars a month down, for a change that felt like growth. His average order value went from $31.38 to $33.35 and he'd have posted that number proudly.

Run it again assuming the promotion works twice as well and 20 of the 33 step up. Now the margin gained is $214.38, the shipping lost is $211.50, and he's up by $2.88. Twice the response, and it's still a rounding error.

That's the real finding. For a light product with a wide margin, a well-set threshold is roughly free. Not profitable, not ruinous. Which means you should decide it on something other than the money: a checkout with no shipping line converts better than one with a surprise on the last screen, and "free delivery over $40" is a thing you can put in a caption. Those are good reasons. "It'll make me more money" usually isn't.

Where to put the threshold

There's a rule that gets you close, and it's arithmetic rather than taste.

Your threshold should sit above your average order value, by roughly the price of one typical item.

Marcus's average is $31.38 and a pin is $12, so anywhere from $40 to $45 is right. He picked $40.

What happens if you get it wrong in either direction:

• Too low (say $25). Now 58 of his 100 orders qualify automatically. He hands out $261 of free shipping and almost nobody steps up, because almost nobody has to.
• Too high (say $75). Only 6 orders qualify, and the gap from a $30 basket to $75 is four extra pins. Nobody buys four extra pins to save $4.50. The offer is decorative.

Round the number up to something that reads cleanly, then leave it alone for three months. Moving a threshold every few weeks teaches regular buyers to wait and see what next month's is.

The other shape: build it into the price

The alternative is free delivery on everything, funded by a price rise. Marcus takes his pins from $12 to $14 and never charges shipping again.

Watch what that does at the two ends:

Order	With $4.50 shipping, $12 pins	Free shipping, $14 pins	Difference
1 pin	Profit $8.90	Profit $6.40	-$2.50
5 pins	Profit $43.30	Profit $48.80	+$5.50

The single-pin buyer got worse for him. The five-pin buyer got better.

Now compare that with the threshold, which did the exact opposite: it cost him money on his biggest orders and made money on the ones that stepped up. A free-over threshold taxes your best customers to subsidise your medium ones. Baked-in pricing taxes your smallest orders to subsidise your biggest. Both are distortions. Pick the one that points at the behaviour you want more of.

For Marcus, who wants bigger baskets, baking it in is arguably the better instrument, and it also gets rid of the shipping decision entirely. The catch is that his $12 pin now shows as $14 next to someone else's $12, and a buyer comparing two Instagram shops does not read the shipping line first. That's a real cost and it doesn't show up in any spreadsheet.

When free delivery is the wrong answer

Some businesses shouldn't do this at all, and the tool-vendor version of this article never says so.

Your situation	Why free delivery hurts
Gross margin under about 30%	The shipping cost is a huge slice of what you keep. There's nothing to fund it with
A wide weight range in one catalogue	The threshold that works for the light item bankrupts you on the heavy one
Your heaviest item is your cheapest	Someone will order four of them to hit the threshold. That's the worst order in your shop
Big variation in delivery distance	A flat free rate averages a cost you can't average, and remote orders eat the lot
You sell mostly single items	There's nothing to step up to. A threshold is meaningless

If two or three of those describe you, charge honestly for delivery and compete on something else. Buyers are not as allergic to a shipping line as the internet insists, particularly when the number looks like a real cost rather than a made-up one. $4.50 reads as postage. $19 reads as a margin grab.

If you don't know your gross margin to within a couple of points, work that out before you decide anything here. Working out if you are actually profitable is the prerequisite for this whole article.

The half-measures worth considering

Free delivery is not binary. Three middles that often beat both extremes:

Free local collection. Costs you nothing, delights the 5% who live nearby, and turns them into your most profitable orders. Switch this on today whether or not you do anything else.

A reduced rate over a threshold instead of free. "$4.50 delivery, or $1.50 on orders over $40" gets most of the psychological effect at a third of the cost, and it keeps a shipping line in the checkout so the buyer doesn't stop believing shipping is real.

Free delivery as a returning-customer thing. Give it to the second order rather than the first. You're spending the same money on someone you already know converts.

Any of these is a discount, and the arithmetic of running a discount without giving away your margin is the same arithmetic as here. How to run a discount without losing money covers the general case.

Test it properly, or don't test it

Thirty days is not a test if you launched it the same week you posted a reel that did well.

Run it for a full month, and record four numbers before and after: average order value, number of orders, total shipping collected, total shipping paid. Not just the first one. Average order value going up while total profit goes down is the single most common outcome and it's invisible if you only look at the headline.

Then take the difference in gross margin, subtract the difference in net shipping, and see whether the answer has a minus sign. That's it. That's the test.

What Sailo does here, and the bit that will catch you out

Every delivery method has an optional free-over amount. Set it and any order at or above that subtotal ships free on that method; leave it blank and the rule is off. You can set a different threshold on each method, so "Standard" can go free over $40 while "Express" never does.

Here's the limitation, and it's a specific one worth knowing before you turn this on: the threshold is checked against the subtotal after any coupon discount. A buyer with a $42 basket who applies a 20% code drops to $33.60 and loses free delivery at the final step. From their side, the discount code made the total go up by $4.50 on a line they'd already seen as $0, and that reads as a bug rather than a rule.

It's defensible behaviour, since discounting your way past a shipping threshold is exactly how people game one. But it will generate messages, and you should either set your coupon minimums above your free-delivery threshold or say plainly in the coupon terms that discount codes and free delivery don't stack.

The other limit: one threshold per method, one currency, no variation by weight or country. If you need free shipping over $40 domestically and over $120 to Canada, that's two named methods, and the buyer picks.

Do this before you change anything

Open your last hundred orders and sort them by basket value. Count how many are already above the threshold you were about to set. That single count tells you what the offer costs before it earns anything, and it takes about ten minutes.

Then work out your true cost per parcel, including packaging and your own time, because every number in this article depends on it. How to work out delivery charges does that band by band. And if you're deciding this because your margins feel thin generally, check what the payment side is taking first, in how payment fees eat a small order. It's often the bigger leak.