Discontinuing a product line: the run-out plan

Deciding which products to cut is a separate discipline with its own method, and it is already covered: SKU rationalisation for marketplace catalogues and assortment pruning for quick commerce. Assume that work is done and the verdict is in. What follows is the retirement itself, which is where the avoidable damage happens, because the line you are killing still has customers buying it today.

Sequence the last units, do not just stop reordering

Left alone, a discontinued line runs dry at a different moment in every channel, which produces the worst version of the outcome: out of stock everywhere at unpredictable times, angry subscribers, and orphan units in the one channel that cannot move them. Sequencing means deciding where the last units go.

Start by counting what is committed, in four places: your own warehouse, stock inside marketplace and quick commerce networks, units with distributors, and open purchase orders and packaging you have already paid for. Then stop replenishing the channels you want to close first, usually the ones where a gap does most damage to your other products, and push the remaining volume into the channels least sensitive to intermittent availability, typically your own site and offline partners. Quick commerce is normally the first to switch off, because a slot that is empty half the time hurts the rest of your range there. For the mechanics of clearing the residue, liquidating dead stock and markdown timing cover the ground.

Subscribers are owed a date, not a surprise

The customers who will be hurt most are the ones on a replenishment or subscription cycle, because they have stopped thinking about this purchase entirely. They deserve three things before anyone else hears: a dated notice, a clear statement of which shipment is their last, and a cancellation that happens automatically rather than one they must chase.

Reserve stock against the active subscriber base before you open the run-out to general sale. Getting this backwards, where general demand consumes the units your committed customers were promised, converts your most loyal cohort into your loudest complaint. Also stop the billing cleanly. A recurring charge that keeps attempting after the final shipment is a chargeback and a support ticket, and the plumbing for that sits in your dunning and billing flow.

A substitute only helps if it substitutes

Offering the nearest thing you still make is reasonable. Pushing it is not. Customers bought the discontinued item for a specific attribute, and if the replacement misses that attribute you get a return, a one star review and a lost customer instead of a clean goodbye. Shoppers are already sceptical of swaps, which is visible in substitution behaviour when an item is unavailable.

State plainly what is different, offer a trial size or a discount on the first replacement order, and make it easy to decline. A customer who leaves politely may come back for something else. A customer sent an unwanted substitute will not.

Refills, spares and warranty outlive the product

This is the part brands forget, and it generates complaints for years. If the line has consumables, refill packs, filters, cartridges or replaceable components, the obligation continues long after the base unit stops selling. Same for durable goods with a warranty: your warranty commitment runs from the date the last unit was sold, not the date you stopped making it, so the final week of sales sets the real end date.

Three decisions, made before the run-out and written down. How long you will supply refills and spares after the last sale. How many units of each you will hold to back that promise, sized on installed base and failure rate rather than on recent sales, which is the logic in criticality based spares planning. And where a customer goes to get service once the product is off your site, which matters most if you rely on a third party repair network that will otherwise deprioritise a dead line.

Announce what you can actually service

A discontinuation notice reliably produces a spike from people who want to stock up. That is good if you have units and bad if you do not, because the spike lands on a fixed quantity and everyone who misses out has now been told twice: once that it is ending, once that you could not supply them.

So match the announcement to the stock. With healthy cover, announce broadly and let the run-out clear inventory at full price. With thin cover, tell subscribers and warranty holders directly, update the listing quietly, and skip the campaign. Never announce a last-chance window you cannot fill, and never let the news reach distributors and platform category teams from a dashboard, which is the argument in telling the channel before a change lands.

After the last unit ships

Do not delete the listing. It carries reviews, backlinks and search traffic that took years to build. Set it out of stock, mark it discontinued where the platform allows, and point it at the successor if there is one. On your own site keep the URL live with a discontinued note and a link to the replacement, rather than serving a dead page to people looking for support. Then close the loop in the ledger: unsold stock and committed packaging need a formal decision, and write down versus write off is where that gets settled.

Write the end-of-life note

Finish with a short internal record against the item code, held wherever your team will actually find it. Six fields: why it was discontinued, the last production date, the last ship date, the support and spares end date, what the substitute is, and who signed it off. Without it, somebody sees an old bestseller in a report two years later, relaunches it, and rediscovers the reason it was cut at full cost. If the retirement is happening because a whole channel is going away rather than the product itself, leaving a platform cleanly is the sequence to run instead.