Upsell vs Cross-Sell
Upsell trades the shopper up to a premium version of the same product; cross-sell adds a complementary item. Here's how they differ in placement, psychology, and AOV impact — and when to use which.
Upsell vs Cross-Sell
Upsell trades the shopper up to a premium variant of the same product; cross-sell adds a complementary item to the cart.
Upsell and cross-sell are the two foundational levers for raising average order value in online retail, and operators conflate them constantly. An upsell replaces the chosen SKU with a higher-tier version of the same thing — a larger size, a premium material, an extended warranty. A cross-sell keeps the original SKU and bolts on a complementary product — socks with shoes, a case with a phone, a primer with a foundation.
The distinction matters because the two tactics work at different funnel stages, lean on different cognitive biases, and produce different AOV curves. Treating them as interchangeable is why most Shopify merchants under-earn from both.
The cleanest test: if removing your suggestion would still leave the shopper with a working purchase, it's a cross-sell. If removing it forces them back to the original choice, it's an upsell. A 64GB-to-128GB phone prompt is an upsell. A phone case suggestion under that same phone is a cross-sell.
Psychologically they pull on different levers too. Upsells exploit anchoring and loss aversion — the shopper has already committed to the category, so the marginal cost of "a bit better" feels small relative to the base price. Cross-sells exploit completion bias and bundling — the shopper wants the job done, and a complementary item closes the loop.
Typical AOV lift by tactic and placement (Shopify apparel + beauty stores, €1M-€15M)
| Placement | Upsell AOV lift | Cross-sell AOV lift | Take rate |
|---|---|---|---|
| Product page | +8% to +14% | +3% to +6% | 2-5% |
| Cart drawer | +4% to +7% | +6% to +11% | 4-8% |
| Checkout (one-click) | +3% to +5% | +5% to +9% | 6-12% |
| Post-purchase page | +11% to +18% | +9% to +15% | 10-18% |
| Email (order confirmation) | +1% to +3% | +2% to +4% | 1-3% |
Two patterns jump out. Upsells win on the product page because the shopper is still actively choosing — switching them to a premium variant is a low-friction edit. Cross-sells win in the cart and at checkout because the buying decision is locked, and adding a complement feels like finishing the job, not reopening the decision.
When to use an upsell
Upsells work best when the price gap is small and the upgrade is easy to grasp in one sentence. A 20% premium for "double the storage", "organic cotton instead of standard", or "3-year warranty instead of 1" converts. A 60% premium for a flagship variant the shopper didn't research usually doesn't — they bounce back to compare specs and abandon.
Place upsells on the product page and on the post-purchase confirmation, not in the cart. In the cart, the shopper has mentally banked their choice; asking them to re-decide the core SKU introduces doubt and lowers checkout completion. Post-purchase is the strongest upsell slot of all — risk is gone, payment is done, and a one-click "upgrade to the larger size" routinely lifts AOV double digits. See the post-purchase upsell playbook for the mechanics.
The cart-page upsell trap
The single most common mistake we see in audits: an aggressive variant-swap upsell in the cart drawer. It reopens the decision the shopper just closed, drives a measurable drop in checkout-initiation rate, and the upsell take rate rarely compensates. If you want to lift AOV from the cart, use a cross-sell — not an upsell.
When to use a cross-sell
Cross-sells work when the complement is genuinely useful and obviously related — socks with shoes, brush with foundation, charger with laptop. The take rate collapses when the link feels like a stretch, so anchor recommendations on co-purchase data rather than "products in this category". The shopper needs to recognise the pairing in under two seconds.
Place cross-sells in the cart drawer, at checkout, and in the post-purchase flow — anywhere the primary decision is locked. Keep the cross-sell price under 25% of the basket value; above that, it starts feeling like a second purchase rather than an add-on, and take rate drops sharply. This is one of the highest-leverage AOV levers most stores under-use.
Take rate by funnel stage: upsell vs cross-sell
Upsell
Cross-sell
Upsell vs cross-sell: frequently asked
An upsell trades the shopper up to a premium version of the same product (bigger size, better material, longer warranty). A cross-sell adds a complementary but different product to the order (socks with shoes, a case with a phone). The test: if the suggestion replaces the original SKU it's an upsell; if it adds alongside it, it's a cross-sell.
It depends on placement, not on the tactic itself. Upsells win on the product page and post-purchase (8-18% lift). Cross-sells win in the cart and at checkout (6-11% lift). On a typical Shopify store running both well, cross-sell volume is higher because take rates are higher in cart-stage placements.
Yes, and you should. The two tactics work at different stages and don't cannibalise each other if placed correctly: upsells on the PDP and post-purchase page, cross-sells in the cart and checkout. Combining them adds 15-25% to AOV in most stores we audit.
In the cart drawer. Reopening the core SKU decision after the shopper has committed introduces doubt and measurably lowers checkout-initiation rate. The cart is cross-sell territory; keep variant-swap upsells on the product page or post-purchase.
Product page upsells run 2-5%. Cart and checkout cross-sells run 4-12%. Post-purchase upsells are the strongest single placement at 10-18% — risk is gone, payment is done, and the add is a single click.
Use co-purchase frequency, not category tags. Pull your last 90 days of orders and surface the pairs that already appear together — those recommendations convert two to three times better than "customers also viewed" or category-based suggestions.
A bundle is a packaged cross-sell — multiple complementary items sold as one SKU at a slight discount. It captures the AOV lift of a cross-sell upfront, removes the decision step, and tends to outperform à la carte cross-sells when the bundle composition is obvious (skincare routine, gaming setup, starter kit).
Yes, when placed in the cart or checkout. A variant-swap upsell at those stages reopens a closed decision and reliably drops checkout completion by 1-3 points in the tests we've seen. On the PDP and post-purchase page, the impact on conversion is negligible because no commitment has been made yet (PDP) or has already been completed (post-purchase).
Don't pit them against each other globally — test placement-by-placement. Run the upsell variant against the cross-sell variant in the same slot (e.g. cart drawer) with traffic split 50/50 and AOV as the primary metric. Most stores find cross-sell wins in cart and checkout; upsell wins on PDP and post-purchase.
Add a post-purchase upsell first. It's the highest-take-rate, lowest-risk placement — the order is already secured, so a failed upsell costs you nothing. Once that's live and earning, add a cart-drawer cross-sell based on your top co-purchase pair. Those two placements alone typically add 12-20% to AOV.
Test ideas before you ship them
Run unlimited A/B tests, attach hypotheses to outcomes, and build a searchable archive of what works — and what doesn't.