Every catalog sheds products. A line gets discontinued, a seasonal item ends, a supplier drops out, a duplicate listing gets cleaned up. So you open the product, hit delete, and move on. What you do not see is that the URL that product lived at does not disappear with it. It stays in Google's index for weeks. It stays in the backlinks other sites built to it. It stays in your own internal links, in old newsletters, in the Google Shopping feed, in a paid ad still running. And now every one of those points at a page that returns a 404. Shopify deleted the product and left the address pointing at nothing.

This is one of the most repeated SEO complaints in the merchant communities, and it always arrives the same way: someone notices a wall of 404s in their analytics or Search Console, traces them back to products they deleted or handles they changed, and asks how to make Shopify redirect them automatically. The early answers are a version of "it does not, you have to build the redirects yourself, or get an app." The merchant is not asking for anything fancy. They want the obvious thing a store that constantly adds and removes products needs: when a URL stops working, send the visitor to the closest page that does. The platform running their entire storefront has no built-in way to do it.

The product is gone. The link to it is not, and Shopify will not point it anywhere on its own.

Why a 404 from a deleted product costs you a real sale

A 404 is not a neutral event. The page you deleted had spent months earning things you cannot see at a glance: a position in search, a set of backlinks, a place in your internal linking, a row in your product feed. That accumulated value, the link equity, only passes forward if the old URL points at a live page. When it 404s instead, the value drains. Search engines eventually drop the dead URL, the backlinks now lead to an error, and the ranking you built is simply gone rather than handed to a replacement.

The more immediate cost is the customer standing on the dead page right now. They clicked a real link, from a search result, an old email, a friend's recommendation, a paid ad you are still paying for, and they landed on "page not found." They do not patiently go hunting through your menu for the new thing. They hit back and they are gone, and if the click came from an ad, you paid for the privilege of bouncing them. On a store that churns through products, this is not one stray error. It is a steady trickle of paid and earned traffic poured onto pages that no longer exist.

What Shopify actually does, and where it stops

Shopify is happy to let you delete a product, change a handle, or unpublish a page. What it does not do is manage the wreckage those actions leave in the URL space. The gap is the whole cleanup: noticing a URL just broke, deciding where it should now go, and writing the redirect. It shows up in three places merchants hit fast.

First, deleting a product creates no redirect. The old URL just starts returning a 404. There is no prompt at delete time asking where to send the traffic, no default forward, nothing. The same silent break happens when you change a product's handle, which rewrites its URL and orphans the old one, and when you unpublish rather than delete. In every case the previous address dies quietly and you only find out later, if you go looking in Search Console.

Second, the native fix is manual, one URL at a time. Shopify does let you create a URL redirect, under Online Store, then Navigation, then URL Redirects. You type the old path, you type the new one, you save, and you repeat for the next dead URL. There is a CSV bulk import for doing many at once, but you still have to know every broken URL and decide every destination before you can build the file. Either way the thinking and the typing are entirely on you.

Third, Flow cannot create a redirect. Shopify Flow, free on every plan, can react to a product being deleted and can send you an internal note about it. But Flow has no action that creates a URL redirect, so the one step that actually fixes the 404 is the step it cannot take. You can automate the alert that a product was deleted. You cannot automate the redirect that keeps its URL alive.

◆ DATA The tempting shortcut is to point every dead URL at the homepage in one blanket rule. Search engines treat that as a soft 404, because the destination has nothing to do with the page that was requested, so you lose the link equity anyway and you dump the customer somewhere they then have to start their search over. A redirect only does its job when the target is genuinely relevant: the replacement product, the same collection, the nearest alternative. The homepage is the fallback of last resort, not the plan.

Why the usual fixes don't hold

Once the 404s start showing up, merchants reach for one of a few workarounds. Each gets you partway and each breaks in a familiar spot.

"I will redirect everything to the homepage." Fast, and it does technically stop the 404. But it is the soft-404 trap: the value of the old page evaporates because the homepage is not a real answer to what was requested, and the customer who wanted a specific product now faces your whole catalog with no idea where the thing went. You have swapped a dead end for a shrug.

"I will set up the redirect by hand whenever I delete something." A fine intention that survives until the first busy week. The discipline slips, the bulk catalog cleanup happens in one sitting with no redirects attached, and worse, the silent breaks, the handle you tidied, the product you unpublished, never even feel like deletions, so they never trigger the habit. The 404s accumulate in the background while you believe you are keeping up.

"I will install a redirect app." There are good ones, and for many stores they are a fair answer, which by itself proves the native gap is real. But most work by logging the 404 after a visitor has already hit it, then surfacing it for you to fix, or by offering a blanket rule that forwards by pattern. The detection is genuinely useful. The judgement it still leaves to you: which live page each specific broken URL should actually point at, so the redirect is relevant rather than just present.

What the automation actually has to do

The real job is not "redirect deleted products to the homepage." It is "for each URL that a delete, an unpublish, or a handle change just broke, find the closest live page based on what that product actually was, write a permanent redirect to it, and keep that redirect honest as the catalog keeps moving, while holding the high-value URLs for a human to confirm." That is watching, reading, and matching, run per URL across everything that changes. As a Dugong playbook, in plain prose, it reads like this: 

# trigger
On a product deleted, unpublished, or its handle changed

# steps
1. Catch the URL that is about to stop working before it
   goes cold
2. Read what the page was: product type, collections,
   tags, and title
3. Find the closest live page: a replacement product,
   the same collection, or the nearest alternative
4. Write a 301 redirect from the broken URL to that page
5. Recheck later that the target is still live and in
   stock, and repoint it if it is not
6. Hold URLs with real traffic, rankings, or backlinks
   for a human to confirm the destination
7. Log every broken URL, the chosen target, and why

Seven lines. The compiler fills in everything beneath: watching for the delete, the unpublish, and the quiet handle change alike, reading the product so the match is based on what the page was about, choosing a destination that a customer would actually accept as the next best thing, writing the permanent redirect, then circling back so the redirect never chains into a fresh dead end when the target itself sells out or gets removed. The merchant never sat in Search Console exporting a list of 404s on a Friday. They described how a careful shop owner would handle a discontinued product, and let the compiler do the volume and keep the map current.

◆ NOTE The redirect that does damage is the lazy one. Forwarding a deleted product to a page that does not match it looks tidy in your redirect list and still loses you the ranking and the customer, because relevance is the entire point of a redirect. And a redirect that points at another product you later delete just moves the dead end down the road. The job is not "make the 404 go away," it is "send this specific visitor and this specific link equity to the live page that best answers what they came for, and keep it that way."

Why this is a compiler problem, not an app problem

There are capable redirect apps, and for many stores they are a fine answer. But the thing that makes a redirect worth anything is not catching the 404, it is the choice of where to send it: reading what the deleted product was, knowing whether a real replacement exists or only a loose cousin, deciding when the collection is the better target than any single product, and noticing when the destination you picked last month has since gone out of stock. Those are judgements about a particular URL, weighed together, and the right answer changes with every product, which means the work is never set-and-forget. It is decided fresh each time.

A natural-language compiler fits because broken-URL cleanup was never really a redirect-button problem. It is a matching and judgement problem wearing a redirect button's clothes. Firing a blanket forward to the homepage is the easy part, which is why the shortcuts do it. The hard part is everything around it: read the page that died, find the closest living one, write the 301, keep it pointed at something real, and pull a human in on the URLs that carry real weight. You can write that brief in a paragraph. You could never hold it in your head across every product you removed this quarter. It sits in the same family as the meta description you write one at a time, where the SEO field Shopify stores but will not fill has to be written page by page, and the image alt text no one writes, where the same blank-field-times-the-whole-catalog problem quietly costs you search visibility.

The workflow worth building this week

If your catalog turns over at all, if you discontinue lines, run seasonal products, clean up duplicates, or rename handles, this is the automation to set up before your next round of deletions, because every product you remove without a redirect is a link you used to own pointing at a dead page, and the only thing between you and a clean URL space is a cleanup you keep not getting to. It pairs naturally with the merchandising work we have written about, like the sold-out products that sink your collection, where what happens to a product as its status changes decides how the storefront performs, and it feeds the same flywheel as the Shopify automations no one builds, where the work that compounds is the part merchants never get to.

Describe it the way you would brief a careful shop owner: whenever a product gets deleted, unpublished, or renamed, do not leave its old link pointing at nothing, send it to the closest thing we still sell, keep that redirect aimed at something live, and bring me the few pages that actually matter before you touch them. That is the whole brief. The compiler does the watching, the matching, the writing, and the upkeep. You keep your attention for the products that matter, and you stop quietly bouncing paying customers and hard-won rankings off pages that no longer exist.

◆ READING If this resonates, two companion pieces: our dispatch on the meta description you write one at a time, where the SEO field Shopify stores but will not write has to be filled page by page, and the field study on the Shopify automations no one builds, where the work that actually compounds is a sliver of what merchants wire up.

If you are a Shopify merchant who has wired up an auto-redirect that picks a relevant target instead of dumping everything on the homepage, or you have a story about the day you found a thousand 404s from a catalog cleanup six months earlier, the inbox is open: field-notes@dugong.live. We are collecting case studies for the next issue.