Sushinet

Catching a Wrong Price Before a Customer Buys at It

A price that dropped 90% by mistake — a typo, a bulk-edit error, a misbehaving app — sells at the wrong number until someone notices. Here's how a suspicious-change flag can catch it before that happens, and what the flag is actually checking.

use-case

The failure this is about

A variant's price gets changed by a staff member, a bulk CSV edit, or a third-party app — and it's wrong. Maybe a decimal point moved, maybe a bulk price update targeted the wrong collection, maybe an integration pushed a stale number. Nothing about the storefront looks broken; the product page loads fine, checkout works fine, the price is just wrong. Orders start coming in at that price, and nobody notices until a customer asks about it, or until someone happens to look at the product again.

What a suspicious-change flag is checking for

The signal that separates an accidental price drop from a deliberate sale is whether the compare-at price moved with it. When a merchant genuinely puts a product on sale in Shopify, the normal way to do it is to set (or keep) a compare-at price above the new selling price — that's what shows the "was $40, now $28" style strike-through to a shopper, and it's a clear signal the price change was intentional.

A price that drops sharply while the compare-at price stays exactly where it was doesn't carry that signal. Nothing marked it as a planned discount. That combination — a price fall past a set percentage, with no compare-at change in the same update — is what gets marked suspicious and surfaced in a flagged view, separate from the ordinary stream of changes.

Why it's a proxy, not a certainty

This isn't a claim that every unflagged price change is safe or every flagged one is a mistake. It's a heuristic built from an ordinary pattern: deliberate sales usually move compare-at price too, so a drop that doesn't move it is more likely to be worth a second look. A merchant who runs sales differently, or who has a specific reason for a change that doesn't fit the pattern, can adjust the threshold or ignore the flag on read.

What happens once it's flagged

The flagged change sits in a dedicated view, separate from the general list of everything that changed, so it doesn't get lost in ordinary catalogue activity. From there, seeing what the price was before the change and reverting it back is a single action — the point of flagging it quickly is to shrink the window between "the wrong price went live" and "the wrong price got fixed," since every order placed inside that window went out at the mistaken number.

The practical takeaway

A price typo or bad bulk edit is not rare, and by the time a customer flags it, the cost has usually already been paid on however many orders came in before someone noticed. A flag built around the same signal a merchant would use to eyeball it — a drop with no matching compare-at movement — catches the pattern automatically, instead of relying on someone happening to look at the right product at the right time.

Undo History watches for this automatically

Undo History records every tracked product change, flags the ones that cross a rule you set, and undoes any recorded change in one click — no workflow builder, nothing to babysit.

Questions

How does a price mistake usually get caught today?

Usually a customer notices first — either they buy at the wrong price, or a support message arrives asking whether a deal is real. By the time either happens, the mistake has often already been live for hours, and every order placed in that window is at the wrong number.

What makes a price change look like a mistake rather than a real sale?

A large drop with no accompanying discount signal. A deliberate sale in Shopify is typically set up by moving the compare-at price alongside the actual price — that pairing is what tells a shopper (and the store) 'this is intentionally on sale.' A price that falls sharply while the compare-at price stays untouched has a different shape: nothing marks it as a planned discount, which is closer to what a decimal-place slip or a bad bulk edit looks like.

Does a flag like this stop the price from going live?

No — it surfaces the change after it happens, in a dedicated flagged view, rather than blocking the update. It's a fast way to notice a suspicious change and revert it before it sits live for long, not a gate that prevents the change from saving in the first place.

Can I control how big a drop has to be before it's flagged?

Yes, the percentage threshold that counts as suspicious is a setting a merchant can adjust, not a fixed number baked into the app. A store running frequent, steep sales might want a higher threshold than a store where prices rarely move at all.

Related guides

Ready to try it?

Undo History records every tracked product change, flags the ones that cross a rule you set, and undoes any recorded change in one click — no workflow builder, nothing to babysit.

See what Undo History does · Pricing · ← All product-history guides