“Free shipping” is the most requested feature in e-commerce and the most misunderstood expense. It’s never free — it’s a cost transfer, and the only question is who absorbs it and how efficiently. Most merchants get the order of operations backwards: they set the offer first and discover the cost later. Here’s the sequence that works.
ship-audit.com is reader-supported: if you sign up through our links, we may earn a commission at no cost to you. It never changes our recommendations — we recommend Pirate Ship despite earning nothing from it.
Free shipping works when you do three things in this order: (1) cut what a label actually costs you — the same parcel runs 35–50% cheaper at commercial tier than at default retail rates, and that gap is the margin that funds the offer; (2) set a threshold above your average order value so the offer pushes baskets up instead of subsidizing orders you’d have gotten anyway; (3) decide the absorption model — threshold, baked into prices, or flat rate. The stakes are real: per the Baymard Institute, 48% of shoppers who abandon a cart cite extra costs like shipping as the reason — the single largest fixable cause.
| Model | How it works | Best for |
|---|---|---|
| Threshold (“free over $X”) | Free shipping unlocks above a basket value set just past your AOV | Most stores — turns shipping into an AOV lever |
| Baked into prices | Average shipping cost folded into product prices; “free” everywhere | Consistent parcel profiles; single-item orders |
| Flat rate | One honest shipping price on every order; not free, but surprise-free | Heavy/variable parcels where free would bleed margin |
| Sitewide free | Every order ships free, no minimum | High-margin, light products only — the riskiest model |
Abandonment figures from the Baymard Institute’s published research. Rate-gap figures from our 2026 compilation of carrier-published pricing. Worked examples below are illustrative arithmetic, not statistics.
Why This Is Worth Solving: The 48% Problem
The Baymard Institute’s long-running checkout research puts numbers on what every merchant feels: roughly seven in ten carts are abandoned, and among the reasons shoppers themselves give, extra costs — shipping, taxes, fees — lead at 48%, ahead of forced account creation and slow delivery. A shopper who accepted your product price balks at the delivered price. That’s not a marketing problem; it’s a pricing-architecture problem, and free or transparent shipping is the direct fix.
But fixing conversion by destroying margin is just moving the leak. A $6 label absorbed on a $25 order with 40% product margin eats 60% of the gross profit on that order. Which is why the sequence below starts somewhere most guides never mention: the label itself.
Step 1 — Shrink the Cost You’re About to Absorb
Every free-shipping model transfers the label cost to you. So before designing the offer, make the thing you’re absorbing as small as possible. The same box, same carrier, same service costs meaningfully different amounts depending on the tier you buy at: default retail-tier setups sit at the top of the pricing ladder, while the commercial tiers surfaced by shipping platforms run 35–50% below per label, per our compilation of 2026 published rates.
Illustrative arithmetic (not a statistic — your numbers will differ): a store absorbing shipping on 150 orders a month at a $7.50 retail-tier label pays $1,125 monthly for its “free” promise. The same parcels at a $4.80 commercial rate cost $720. That $405/month gap is the funding for the offer — recovered before touching prices, thresholds, or margins. This is the step most threshold guides skip, and it’s the one that makes the rest affordable.
What would Step 1 recover for you?
Run the free shipping audit — volume, package size, destination — and see your monthly gap between default and commercial-tier pricing. That number is your free-shipping budget. No signup needed.
Run Your Free Audit →Step 2 — Set the Threshold Above Your AOV, Not At It
The threshold model’s entire logic is incremental behavior: the offer should make shoppers add something to qualify. That only happens if the bar sits above where baskets naturally land. The widely used heuristic: take your current average order value and set the threshold roughly 20–30% higher. Say your AOV is $42 — a $55 threshold invites a second item; a $40 threshold just gives away shipping on orders you were already getting, and a $90 threshold reads as unreachable and influences nothing.
Two refinements worth stealing:
- Check the margin of the “bridge” products shoppers add to qualify. If the typical add-on is a low-margin item, the extra revenue may not cover the label you’re now absorbing. The healthiest thresholds sit where a high-margin small product naturally bridges the gap.
- Mind your zones. A nationwide threshold funded by your average label cost quietly loses money on cross-country orders, where the same parcel costs most. If your order map skews to far zones, budget on your zone-weighted cost, not your average — or pair the threshold with ground services priced for distance, as covered in our Ground Advantage vs Priority guide.
Step 3 — Choose the Absorption Model Honestly
Threshold is the default for a reason: it converts the 48% objection and raises basket sizes. Start here unless your catalog argues otherwise.
Baked-in pricing suits stores where parcels are consistent — if every order ships in the same box at similar cost, folding the average into prices is clean and “free shipping” becomes a permanent storewide claim. It breaks when parcel costs vary widely: light-order customers overpay, heavy-order customers are subsidized, and competitive product pricing suffers.
Flat rate is the honest middle path merchants underrate. Baymard’s finding is about surprise costs — a clearly stated, reasonable flat rate shown early removes the surprise without absorbing the full label. For heavy or variable catalogs where free would bleed, it’s often the profit-optimal answer.
Sitewide free deserves its reputation as the margin killer. It works for light, high-margin products with tight zone distribution — and almost nowhere else. If you run it, Step 1 isn’t optional; it’s survival.
Making Step 1 Real, by Profile
- Under 50 orders/month (US) — Pirate Ship: free access to commercial USPS pricing, no subscription — the simplest way to shrink the cost you’re absorbing. We earn nothing recommending it. Our review · Start free →
- 50–500 orders/month — Shippo: commercial rates plus per-parcel rate shopping, so each order absorbs the cheapest viable label. Our review · Try Shippo free →
- 500+ or multi-channel — ShipStation: automation rules that pick the cheapest service per parcel make sitewide policies affordable at scale. Compare · Start free trial →
- Shipping from Europe — SendCloud: 160+ local carriers to route each parcel cheaply, plus the checkout delivery options EU buyers expect. Our review · Try SendCloud free →
Frequently Asked Questions
Does free shipping actually increase sales?
The strongest evidence is on the abandonment side: Baymard’s research finds 48% of cart abandoners cite extra costs like shipping as their reason — the largest fixable cause. Removing or neutralizing that objection addresses the biggest documented leak in checkout; how much it lifts your sales depends on your prices, margins and threshold design.
How do I calculate a free shipping threshold?
Start roughly 20–30% above your current average order value, so qualifying requires adding an item. Then sanity-check two things: that typical bridge products carry enough margin to cover the absorbed label, and that your zone mix doesn’t make far-away orders quietly unprofitable at that bar. Revisit quarterly — AOV moves.
Should a small store offer free shipping?
Only after Step 1. At low volume, absorbing retail-tier labels on thin margins is how free shipping kills stores. Commercial pricing first (free via Pirate Ship at small scale), then a threshold above AOV — or an honest flat rate, which removes the surprise without absorbing the full cost.
Is flat rate shipping better than free shipping?
For heavy or highly variable parcels, often yes. The documented problem is surprise cost at checkout; a clear flat rate shown early solves the surprise while sharing the cost. Free converts harder but absorbs more — the honest comparison is margin per order under each model with your real numbers.
Who pays for free shipping?
Always some mix of you and the customer: absorbed from margin (threshold, sitewide), folded into prices (baked-in), or shared transparently (flat rate). The only way to make every model cheaper is to shrink the label cost itself — which is a tier decision, not a negotiation.