Your marketplace listings stopped selling: how to check
A sequence for diagnosing a drop in marketplace orders, how to separate a visibility problem from a conversion problem, and when to stop investing in a listing.
A cliff or a slope
Orders were steady and now they are not. Before touching anything establish which of two shapes you are looking at. A cliff is a drop over two or three days from a normal level to almost nothing. A slope is a fade over six or eight weeks that nobody noticed until the month closed badly.
The two have different causes and mixing them up wastes a fortnight. A cliff is nearly always something switched off or broken. Stock flagged out. A listing deactivated. A buy box lost. An advertising budget exhausted. A slope is usually competitive or seasonal and it will not be fixed by a single setting. Open the orders report and look at daily counts for ninety days before you form a theory.
What this costs you and what it cannot recover
This diagnosis costs time rather than money and most of it can be done from your own seller dashboard in an afternoon. What it cannot do is restore demand that has left the category or win back a buy box that the brand owner has taken for itself. Neither of those responds to listing work and no amount of optimisation changes them.
Account-level problems also do not clear on request. Where cancellation or return or dispatch metrics have breached a threshold the platform watches them over a measurement window so a fix today shows up in visibility some weeks later. Plan for the lag rather than expecting a switch. Where the diagnosis points at content and pricing and advertising across a catalogue rather than one listing the work has a published scope and rate as marketplace optimisation. Rates are stated before tax. Sales tax on marketing services varies by state.
When to stop investing in the listing
Some listings should be allowed to die. If you resell a brand that has since appointed an exclusive seller or started selling directly you will not win that buy box back and spending on advertising to sit second on a shared product page funds somebody else's conversion. Delist and move the capital.
The same applies to a product at the end of its life or out of its season. A ceiling fan listing in December is not broken. Ignore the drop entirely if your baseline is small. A seller moving three units a week does not have a trend and will chase noise for a month. And if the per-order contribution on the product was already thin a recovery simply returns you to selling at no profit. Check that number before you invest in the fix because the honest answer is sometimes that the listing was never worth having.
The check sequence
Work in this order and stop when something is clearly wrong. First confirm the listing is live and in stock and serviceable to the ZIP codes that gave you most of your orders. Serviceability changes without warning. Second search your own main keyword in a logged-out browser and see who occupies the slot you used to hold. Third open the product page and compare every offer on it including delivery charge and any badge the other seller carries that you do not.
Fourth open account health and read cancellation rate and return rate and dispatch breaches. Fifth check advertising. An exhausted daily budget and a bid that no longer clears the floor both look exactly like a listing nobody wants. Sixth read the catalogue record for recent edits. A rejected image or a changed attribute or an automated title revision can remove you from a filter that was sending most of your traffic. Last compare your slope against two competing listings. If everyone faded together you are looking at the category and not at yourself.
The two numbers that split the diagnosis
Separate impressions from conversion and the problem names itself. If impressions fell and the view to order rate held then you have a visibility problem and the cause sits in ranking or serviceability or advertising. If impressions held and the view to order rate fell then shoppers are still seeing you and choosing somebody else and the cause sits in price or images or reviews or delivery promise.
Total views is the flattering number. It rises when you pay for it and it tells you nothing on its own. The listing quality score flatters in the same way. It can read well on a listing that sells nothing because it measures completeness of the record and not desirability of the offer.
The common mistake
Almost every seller cuts price first. It is the quickest lever in the dashboard and it is the most expensive thing in this guide. A price cut on a shared product page invites the other sellers to match within a day so you arrive back where you started at a lower realisation. Worse the new price becomes the reference point shoppers and the platform both remember and raising it later costs you the ranking you bought.
A close second is building a fresh listing to start clean. That splits your reviews across two records and leaves the old one competing with the new one. Diagnose impressions against conversion first. A visibility problem never gets solved by a discount and a conversion problem rarely needs one.
This guide explains how something works. It is not legal, tax or financial advice, and rules change. Confirm anything regulatory with a qualified professional before you act on it.