Tracking Your Product Ads Clicks with Your Analytics Service | Main | Having trouble accessing a section in your seller account?

June 25, 2010

Preventing Stock-outs

Managing your inventory levels is an important part of selling on Amazon. A high number of pre-fulfillment order cancellations can have a negative impact on your Customer Metrics. Canceling orders due to lack of inventory causes a poor buying experience which can also result in negative feedback.

When a seller cancels an order prior to fulfillment, we have found this is usually because the seller does not have the item in stock. This article will cover the most common causes of stock-outs and how to prevent them.

I would like to start off by providing some insight into our order and feed processing systems.

  • Inventory updates submitted via Product Data Feeds are processed sequentially. When 10 inventory update files are submitted in rapid succession. The first must run through to completion before the second is processed. The second must complete before the third is processed and so forth.
  • File size, or number of SKU updates to be processed, impacts the time necessary to complete the processing. Small files can complete processing in a manner of seconds. On the other hand, extremely large files can take 15+ minutes to complete.
  • Amazon systems automatically decreases inventory counts when an order is placed.
  • There is a 30 to 90 minute window when an order is place to the time the order will show up on your orders page as ready to fulfill.
  • If the order takes longer due to payment or other clarification the order will display as pending with no shipping details.
  • Order Cancellations with a cancellation reason other than “No Inventory” will automatically increment inventory counts. Please refer to blog post ”Reason for cancellation – inside canceling an order” for more information

Now that we are armed with these facts, let’s look at a few example scenarios that can result in “stock outs”:

Scenario 1: Anthony’s Snowboard Chalet sells snowboards and winter apparel. Anthony operates a retail store in Mt. Baker, Washington, and has a successful on-line business through During high season he gets a large volume of traffic through his retail store and submits inventory update feeds every time he makes a sale. This results in an inventory update every few minutes. Below is a view of his inventory updates for today.


At 12:39:51, Anthony sold a snowboard through his retail outlet for which there was only one in inventory and submitted an inventory feed setting the inventory count to zero. However, this feed is stacked behind a number of other feeds and processing for this feed will not begin for approximately 10 minutes after submission.

At 12:47:05, a buyer places an order for the same item via his Selling On Amazon account. The order is accepted by the system because the inventory count on Amazon has not been updated to zero yet.

Best Practice: To prevent this kind of situation, submit your inventory updates in batches. Our feed processing system can process 20 SKU inventory updates submitted 5 minutes apart faster than it can process 10 feeds with only 2 SKU inventory updates submitted 5 seconds apart.

Scenario 2: Petar’s Motorcycle Emporium sells motorcycle parts and accessories out of Daytona Beach, Florida. Petar has a large inventory and submits a full inventory update every hour.

Image002 In each of the feed submissions, Petar has set the inventory count of SKU Zola-005-BLK to a value of 1. With the inventory count defined as 1 at the completion of the third feed from top at 1:39:11, an order is placed for the item 2:09:15 for the item. Because Amazon’s system automatically decreases inventory counts upon order placement, the seller’s current inventory is now set to zero. No further orders will be accepted for this item now that the inventory level has been depleted.However, at 2:40:11, the inventory count was incremented back to a value of 1 at the completion of the second feed from top. At 3:18:38, a buyer placed another order for the same item for which there is no actual inventory.


Best Practice: Inventory feed submissions should contain inventory data for only those SKUs for which the inventory count has actually changed. Submitting the same data over and over will likely cause inventory counts on Amazon systems to get out of sync with actual inventory. Also, submitting data for only inventory count that has changed greatly reduces the size of the file which improves processing time. (Note the difference in processing time for each of the examples provided in this article.)

Scenario 3: Ralph’s Sailboat Superstore sells innovative sailing gear for competitive and trans-pacific sailors in the San Diego, California, area. An order for a racing spinnaker was placed yesterday from buyer competing in the upcoming Transpac Yacht race. Because there was only one left in stock, Amazon’s system decremented the inventory count to zero when the order was placed.

However, the item was damaged during the shipping process and the order could not be fulfilled. The seller contacted the buyer promptly and cancelled the order. The “Reason for Cancellation” reason submitted to Amazon indicated “Buyer Cancelled” instead of the more accurate “No Inventory” code. Due to the “Buyer Cancelled” reason code, Amazon’s system automatically restored the inventory count for that item back to a value of 1.

Shortly after the order was cancelled, another order was placed and accepted for the same item even though the seller actually has no inventory for the item.

Best Practice: Use the appropriate “Reason for Cancellation” code when cancelling an order, and understand the impact these codes have on your inventory counts. Read our related blog article, Reason for cancellation – inside cancelling an order , for more information.



TrackBack URL for this entry:

Listed below are links to weblogs that reference Preventing Stock-outs :


The comments to this entry are closed.

June 2013

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29

© 2011,, Inc. or its affiliates. All rights reserved.
The Amazon Seller Support logos are trademarks of, Inc. or its affiliates.
About our blog | Privacy Policy | Conditions of Use | Careers