Every store sells out of things. That’s not a defect — it’s often the opposite, proof that something worked. The defect is what the product page does next. A sold-out page is the one place a shopper arrives with intent you’ve already paid for and can’t currently serve, and Shopify gives you exactly one good move: tell the truth quickly, and convert the intent into something you keep — a notify-me address, a pre-order with a date on it, a nudge to the in-stock alternative.
What we keep finding instead is a page that can’t decide whether the product is buyable.
Australian Shopify stores we’ve audited since mid-July 2026 showed some form of stock-state confusion on the pages we sampled — sold-out items with live buy affordances, dead variants behind a working cart button, retired products surfacing as new, or scarcity messaging that undermined itself.
The four ways the out-of-stock moment goes wrong
1. Sold out, but the buy button still works
The worst version, because both signals are load-bearing. In our published Budgy Smuggler teardown (evidence sampled 21 July), sold-out product pages still rendered size selectors and buy affordances — a shopper gets told “sold out” and invited to purchase in the same viewport. A same-day gifting brand we audited the same week showed a sold-out badge on a live product while the Add to Cart stayed active and a same-day delivery promise sat alongside both. Whichever signal is true, the other one is costing you: either a buyable product looks unavailable and you lose the order, or an unavailable product looks buyable and you lose the customer downstream — at the cart, or worse, after they’ve paid and you have to refund and apologise.
2. Dead variants behind a live page
An activewear label we audited on 22 July had five of seven sizes unavailable on a hero product, with no waitlist and no restock signal on any of them. A shopper in one of the five dead sizes gets a working page, a working cart, and no path forward — not even a “leave your email”. Three days later most sizes were back. That’s the point: stock states churn fast, and every day a dead variant offers no capture mechanism is a day of restock demand you never hear about.
3. Ghosts in New Arrivals
A community-driven fashion brand we audited surfaces retired, non-purchasable items inside contexts a shopper reads as current range. Nostalgia sections can be deliberate and even brilliant for community brands — but when a vaulted item renders next to live product with the same visual weight, the shopper’s model of “what can I actually buy here” breaks, and every subsequent page gets read with suspicion.
4. Scarcity that reads as inventory reporting
Fifty-five units is not scarce, and shoppers know it. A threshold set that loose converts a urgency device into a stocktake printout — and a shopper who clocks one inflated claim discounts every other claim on the page, including the true ones. Scarcity messaging is borrowed trust: it only works if it’s only there when it’s real.
What it actually costs
Stock-state confusion is cheaper to dismiss than the other leaks we write about because each individual dead end feels small. It compounds three ways. First, the direct loss: paid or earned traffic landing on a page that can’t complete, with no capture fallback — acquisition spend converted into a shrug. Second, the trust tax: a shopper who hits one contradictory page applies the doubt to your whole store, your shipping promise and your reviews included. Third — and least appreciated — the forecasting loss: notify-me buttons aren’t a consolation prize, they’re demand measurement. A store with no restock capture doesn’t just lose the sale, it loses the data about which restock to prioritise.
One store fixed it in days
Between our first audit pass (21 July) and a re-check on 26 July, the gifting brand from failure mode #1 shipped the correct fix: the sold-out product now disables Add to Cart entirely and shows a notify-me capture in its place. Two signals became one truthful signal, and the dead end became a list-builder. We have no inside knowledge of their numbers, and we don’t know whether our note prompted it — but the speed is the lesson: this is theme-setting and app configuration, not a rebuild. It was live within five days of us first logging the defect.
Doing it right (as observed in the wild)
The good news from the same 28 audits: the correct patterns are all in production on Australian stores right now.
- Sold out means the cart says so. Disabled buy button, plain “sold out” state, notify-me capture in the button’s place. (The five-day fix above.)
- Pre-orders carry a date. A dinnerware boutique we audited this week states the ETA right on the product page — “mid/late August” into their Sydney store — and takes the pre-order cleanly. Compare that with a page that says “available for pre order” with a blank where the dispatch date should be, which we’ve also found live this month: same feature, opposite effect.
- Restock buttons appear only when they’re true. One ingestible-beauty brand we audited showed an “Email Me When Available” button that initially looked like a bug next to an in-stock state. On re-verification the product had genuinely sold out — the button was doing exactly its job. We flag this against ourselves deliberately: the audit standard is is the page telling the truth right now, not does the page contain a pattern from a checklist. Don’t rip out a notify-me because a blog post told you restock buttons confuse people. Verify which state you’re in first.
- Scarcity thresholds are set where scarcity starts. If the counter would embarrass you read aloud — “only 55 left!” — it fires too early. Low-stock messaging earns its keep somewhere under ten units for most catalogues, and not at all on evergreen items.
Audit your own stock states in ten minutes
- Find your three most recently sold-out products (Shopify admin → Products → filter by inventory). Open each storefront page in a private window.
- Check the button. Is Add to Cart disabled? Does anything on the page still invite a purchase that can’t complete?
- Check the fallback. Is there a notify-me capture? Does it actually send a confirmation when you use it with a fresh address?
- Open a product with mixed variant availability. Pick a dead variant — what does the page offer you? A waitlist, or a wall?
- If you run pre-orders: is there a date or window on the page, and is it current? A blank or stale date is worse than none — it reads as abandonment.
- If you run low-stock counters: note the number they fire at. Read it aloud. Decide if it sounds scarce.
- Check your New Arrivals and featured collections for anything that can’t currently be bought, and decide deliberately whether it belongs there.
The honest caveats
Some of this is genuinely hard: multi-location inventory, flow-on effects from bundles, and third-party theme sections can each produce a stock-state lie without anyone touching the theme. That’s an argument for checking regularly, not for fatalism — every example above was visible to any shopper within seconds of landing. And stock-state hygiene is one leak among several: the same audits found no reviews on the buying path at most stores we sampled, nothing offered for an email address at more than half, and self-contradicting shipping promises at about half. The stores leaking here are usually leaking elsewhere, and sequencing the fixes is most of the value of an audit.
We’ll check every dead end on your buying path, not just this one.
A 40-point revenue-leak teardown — stock states, product pages, trust, email flows, shipping messages, SEO, checkout friction — delivered as a prioritised fix roadmap in 48 hours. Fixed A$695, no retainer. Fewer than 5 actionable fixes and we refund in full.