Put prices back when a sale ends, and catch the stale "was" prices it leaves behind
Updated · By Ecomsellertool editorial team
To clean up after a Shopify sale, save every variant's regular price before the sale starts, let a scheduled rule set the sale price and compare-at price, and put both back when the sale ends. Rules also flag compare-at prices that are zero, equal to or lower than the price, or left up too long. A person decides on any price someone changed during the sale.
- Shopify shows a product on sale when its compare-at price is higher than its price, and a compare-at price of $0.00 is not the same as an empty one (checked November 14, 2026).
- Fixed prices in a Shopify market catalog override the prices and compare-at prices set in the admin, so a sale reverted in the admin can stay live in another market.
- Shopify Flow's Scheduled time trigger runs a workflow at most once every 10 minutes, in the store's time zone unless you pick another.
- Amazon shows a List Price only if customers bought the product on Amazon, or other retailers offered it, at or above that price in the past 90 days (180 for seasonal products).
- The FTC's guide on former price comparisons asks that a "was" price be one the product was openly and actively offered at for a reasonably substantial period. This page is not legal advice.
The step map: who runs each step
Trigger: A sale record with a start and end time, Shopify Flow's Scheduled time trigger at the start and end, and a daily sweep of every variant's price and compare-at price.
| Step | What happens | Who runs it |
|---|---|---|
| 1 | Before the sale starts, rules save a snapshot of every variant in the sale: its regular price, its compare-at price, and the prices in each market catalog. | Rule |
| 2 | At the start time, rules set each sale price and set the compare-at price to the regular price from the snapshot, then read every variant back to confirm the write. | Rule |
| 3 | At the end time, rules compare each variant with the snapshot: where the price is still the sale price the workflow wrote, they put the regular price back and clear the compare-at price. | Rule |
| 4 | Variants whose price someone changed during the sale are held, not overwritten, and go to a person with the snapshot, the sale price and the current price side by side. | Rule |
| 5 | A daily sweep reads every variant and flags compare-at prices that are $0.00, equal to or lower than the price, inconsistent across a product's variants, or set for longer than your limit. | Rule |
| 6 | A person decides each held or flagged item: the new regular price, whether a long-running "was" price stays, and any market catalog or marketplace listing to change. | Person |
| 7 | Rules carry out the decision, read the prices back, log who decided and why, and send a short end-of-sale report. | Rule |
6 rule steps · 0 AI model steps · 1 person step. A rule follows fixed logic you approve; an AI model handles messy input and its output is checked by rules; a person makes the calls listed below.
What stays with a person
- Any variant whose price someone changed while the sale was running
- Whether a compare-at price that has been up for weeks is still a real former price (not legal advice)
- Sale prices below your margin floor, before the sale starts
- Market catalog prices and marketplace list prices the sale touched
- The sale record itself: which variants, which markets, start and end time
Which tools can run it
- Shopify Flow: the Scheduled time trigger, Get product variant data for the daily sweep, and Send Admin API request for the price writes
- A workflow tool such as Zapier, Make or n8n, reading the sale record from a sheet and posting held items to Slack or email
- Custom code on your accounts, for the price snapshot, market catalogs, Amazon and Walmart listings and the read-back after every write
Systems it connects
- Shopify products and variants
- Shopify Markets catalogs
- Shopify Flow
- A sale calendar or sheet
- Amazon and Walmart listings, where the sale runs there too
- Product cost per item
- Slack or email for held items
Build size: Medium
The price writes are simple; the work is the snapshot before the sale, the hold on prices people changed mid-sale, and every market and channel the sale reached.
A sale is easy to start and easy to forget. Someone sets 30 variants to a lower price on Friday, types the old price into the compare-at field, and plans to undo it on Monday. On Monday half are put back, a few keep their compare-at price, one had its regular price raised in the middle of the sale, and the Canadian market still shows the sale price. Weeks later the store is showing "was" prices that nobody chose. This recipe keeps a record of every price before the sale, puts it back on time, and sends anything unusual to a person. It is one page of our automation library.
What should the rule do with each price situation?
Start with the decision table, because "clean up the sale" means a different action for each variant. Shopify's sale pricing page sets the basics: you show a sale by putting the original price in the compare-at price field, the compare-at price must be higher than the price, a variant that should show no sale needs the field cleared entirely, and "a compare-at price of $0.00 USD isn't the same as an empty compare-at price" (checked November 14, 2026). Every row below uses the price snapshot taken before the sale. The limits, such as 30 days, are illustrative; set your own.
| Price situation | What the rule does | Who decides |
|---|---|---|
| Sale ended; the price is still the sale price the workflow wrote | Puts the regular price back from the snapshot, clears the compare-at price, reads both back | The rule, within the sale record a person approved |
| Sale ended; the price is back to regular but the compare-at price is still set | Clears the compare-at price | The rule |
| Compare-at price equal to the price | Clears the field: it shows no sale and only adds noise | The rule |
| Compare-at price lower than the price | Holds the variant and alerts: usually a price raised during a sale | A person confirms the new regular price |
| Compare-at price of $0.00 | Clears it to empty | The rule |
| Someone changed the price while the sale was running | Holds it; never overwrites a price the workflow did not write | A person picks the regular price and updates the snapshot |
| Variants of one product with different compare-at prices | Flags the product, since Shopify's page warns the collection page may then not show it as on sale | A person decides which variants are in the sale |
| A market catalog with its own fixed sale price | Flags the catalog line; the admin revert does not reach it | A person approves the catalog change |
| A compare-at price that has been up longer than your limit, for example 30 days | Lists it in the daily sweep with the date it was set and the prices since | A person decides whether it is still a real former price (not legal advice) |
| A sale price below your margin floor, before the sale starts | Blocks that variant from the start-time write | A person approves the price or drops the variant |
| The same sale on Amazon or Walmart | Lists each listing the sale touched, with its sale end date | A person checks each marketplace's own reference price |
The rule that matters most is the hold. A workflow that blindly writes the snapshot back will undo a price increase someone made on purpose in the middle of the sale. So the end-time step only reverts a variant when its current price is exactly the sale price the workflow wrote. Anything else goes to a person, with the three prices side by side.
What does a sale ending look like?
Every number here is illustrative, not from any store or client. Picture a home fragrance brand running a weekend sale from Friday 12:00 am to Sunday 11:59 pm, store time. The workflow saved its snapshot on Thursday. At 12:00 am Monday the end-time run finds seven situations.
| Variant (illustrative) | Snapshot before the sale | Found at 12:00 am Monday | What happens | Who decides |
|---|---|---|---|---|
| Candle 8 oz | $32, no compare-at | $24, compare-at $32 | Price back to $32, compare-at cleared, read back | The rule |
| Candle 16 oz | $48, no compare-at | $48, compare-at $48 (put back by hand, field left) | Compare-at cleared | The rule |
| Gift set | $55, no compare-at | $60, compare-at $55 (raised Saturday after a cost change) | Held: compare-at lower than price, and not the workflow's price | A person confirms $60 and the snapshot is updated |
| Travel tin | $14, compare-at $0.00 | $10.50, compare-at $14 | Price back to $14, compare-at cleared to empty, not $0.00 | The rule |
| Candle 8 oz, Canada catalog | Fixed price CA$42 | Fixed price CA$32 | Flagged: the admin revert does not change it | A person approves the catalog fix |
| Wax melts (not in this sale) | $15, compare-at $20 set 74 days ago | Same | Listed in the sweep as a long-running "was" price | A person decides whether it stays |
| Candle 8 oz on Amazon | List Price $40 entered at launch | Sale price ended Sunday | Listed for review against Amazon's own List Price rule | A person |
Seven lines, four closed by rules and three with a person, each carrying its evidence. The report goes out at 12:10 am. Without the snapshot, the gift set would have dropped back to $55 and lost the price increase, and the travel tin would have kept a $0.00 compare-at price that is not the same as empty.
Why do market catalogs and channels need their own step?
A sale often reaches further than the product page. Shopify's page on catalogs for markets says a price set by hand in a catalog "overrides the overall price adjustment", and that a catalog's price adjustment applies to compare-at prices by default unless you turn off Include compare-at price (checked November 14, 2026). The sale pricing page adds that fixed prices override the prices and compare-at prices set in your admin. So a store that set a fixed sale price for one market must undo it in that catalog, not only on the product.
Marketplaces work differently again. On Amazon you do not set the strike-through price the way you set a Shopify compare-at price. Amazon's strike-through pricing page says it shows a List Price only if the product was bought on Amazon or offered by other retailers at or above that price in the past 90 days, or 180 days for seasonal products, and that the Typical Price is the 90-day median price paid on Amazon, leaving out limited-time promotions (checked November 14, 2026). Walmart's comparison price rules define its comparison price as an item's List Price or Was Price, with the Was Price taken from a 90-day median that leaves out promotions such as rollbacks and clearance (checked November 14, 2026). The workflow lists those listings for a person; it does not try to set a reference price the marketplace calculates.
How long can a "was" price stay up?
The FTC's guide on former price comparisons, 16 CFR 233.1, says a former price is a fair basis for a comparison when it is the price at which the article was offered to the public on a regular basis "for a reasonably substantial period of time" (checked November 14, 2026). It also says the product should have been "openly and actively offered for sale" at that price in the recent, regular course of business, and it names as fictitious a price that was "not maintained for a reasonable length of time, but was immediately reduced", or one used only in the remote past without saying so.
The guide sets no number of days, and this workflow does not pick one for you. What it does is keep the evidence a person needs: when each compare-at price was set, the prices the variant actually carried since, and how long the "sale" has run. A compare-at price that has been up for months, with the product never sold at that price in that time, is the line a person should look at first.
Can Shopify Flow schedule and revert sale prices?
Partly. Flow's Scheduled time trigger starts a workflow at a date and time you choose, in your store's time zone unless you pick another, and is "limited to run a maximum of one time every 10 minutes" (checked November 14, 2026). The Get product variant data action returns up to 100 variants per run, filtered by a query, which suits a daily sweep of a small catalog. Our guide to what Shopify Flow automates covers the basics.
| Job | In Shopify Flow | Where a workflow tool or custom code takes over |
|---|---|---|
| Start and end the sale on time | Scheduled time trigger, once for the start and once for the end | Many overlapping sales from one calendar |
| Write prices and compare-at prices | Send Admin API request, if the mutation is in its list | Read-back and retries after every write |
| Save the snapshot before the sale | Not practical for more than a few variants | A stored copy of every variant and catalog price |
| Hold prices people changed mid-sale | Not practical | Comparing each price with the one the workflow wrote |
| Daily sweep of compare-at prices | Get product variant data, 100 variants per run | Large catalogs and the age of each compare-at price |
| Market catalogs, Amazon and Walmart | Not covered | Catalog imports and marketplace listing reports |
In Flow, the write can go through the Send Admin API request action, which can call "most mutations in the GraphQL Admin API" but not asynchronous ones or ones that return a Job (checked November 14, 2026). The price mutation is productVariantsBulkUpdate, which updates many variants of one product at once and needs the write_products scope; its input object carries price and compareAtPrice. By default a single error stops every variant in the call from updating, which is what you want for a sale: all of a product's variants change, or none do.
How does it fit with your other pricing workflows?
Sale prices are one of several ways a price moves without a plan. Discount codes are another: our discount code leak alerts watch codes that stack with a sale, and the margin floor should be the same number in both. If your regular prices change because a supplier's cost moved, our supplier cost change alerts set the new regular price, and the snapshot here should pick it up before the next sale.
Wholesale buyers need their prices kept apart from a retail sale; our wholesale price list sync writes B2B catalog prices on their own effective dates. Resellers dropping below your advertised price are a different problem, which our MAP price monitoring recipe watches. And on Amazon, a sale ending can change whether you hold the Buy Box; our Buy Box loss alerts say why it moved. Margin math across all of them starts from contribution margin.
Who builds sale price cleanup?
Ecomsellertool builds workflow automation on your own Shopify store, market catalogs and marketplace accounts; we have built software on Amazon's seller APIs since 2017. We write the sale record and the decision table with you, then build the snapshot, the timed writes, the hold, the daily sweep and the report. Price decisions stay with your team. Send a project brief or see how our workflow automation service works.
The building blocks are work we have shipped, though none of these was a sale cleanup build:
- Promotions and sale dates in view. For Seller-Drive we built a promotion view from Amazon's FBA promotion reports, showing the ASINs with promotional data and the sales after each promotion, next to insights on important Amazon Prime Day dates.
- Prices set by rule, with a floor. Repriser is rule-based automation we built that reprices Amazon offers by rule without holding a price below the seller's margin.
- Listing prices managed from one place. For Little Owl we built prep-center software where clients enter listing details such as price and quantity, and manage their Amazon listings from the software.
The workflow can run on Ecomsellertool Growth OS beside your catalog and margin data.
Get your sale price cleanup built
Send us your next sale and the prices it touches
Describe the workflow you want automated. We reply with a written plan: scope, timeline and a fixed price.
Frequently asked questions
How do I end a sale on Shopify and go back to the original price?
Set the price back to the regular price and clear the compare-at price field entirely. Shopify has no undo for a sale, so the safest way is to save each variant's regular price before the sale starts and restore from that list when it ends. Check market catalogs too, because fixed prices there are separate from the admin price.
Why is my Shopify product still showing a sale price after the sale ended?
Usually the compare-at price was left in place. Shopify shows a sale whenever the compare-at price is higher than the price, so a variant whose price went back up but whose compare-at price stayed higher still looks discounted. A market catalog with its own fixed sale price can also keep the sale live in that market.
What happens if the compare-at price is lower than the price on Shopify?
Shopify's help page says the compare-at price must be higher than the price for a sale to show, so a lower or equal one shows no sale. It is still a sign that something went wrong, such as a price raised during a sale, and it should be cleared or fixed rather than left in the data.
Can Shopify Flow schedule a sale and revert prices automatically?
Partly. Flow's Scheduled time trigger can start a workflow at the sale's start and end, and Get product variant data can find the variants to change. The price write itself can go through the Send Admin API request action or through an app or custom code.
How long can I show a was price or compare-at price?
The FTC's guide on former price comparisons does not set a number of days. It asks that a former price be one the product was openly and actively offered at, for a reasonably substantial period, in the recent regular course of business. Amazon and Walmart set their own rules for the reference prices they show. This is not legal advice.
Does Amazon let me set my own was price?
Amazon shows a List Price you provide only if customers bought the product on Amazon, or other retailers offered it, at or above that price in the past 90 days, or 180 days for seasonal products. Its Typical Price is the 90-day median price customers paid on Amazon, leaving out limited-time promotions. Amazon calculates it; you do not set it.
