Predictive Purchasing

Understanding predictive purchasing

Predictive purchasing helps you forecast the quantity to order for a predicted amount of future sales. Projection is based on a number of factors, including the number of past sales and the available quantity. Past sales are used to determine a velocity, or the rate of sales per day over a given date range. Velocity is then multiplied to calculate the deficit supply for multiple date ranges.

In more complex cases, days in stock and on-order quantities are also factored into the prediction. See the predictive purchasing-related Client Settings in this topic.

Key terms

• Days OF Order— Days of past sales from which you are basing the prediction. This generates a daily velocity. Days with no stock availability are skipped in the Days Of Order Count.
• Velocity— Rate of sales per day when calculated by Qty Sold in Days of Order divided by Days of Order
• Deficit Supply— Predicted deficit to your supply for multiple date ranges calculated as: (Velocity x Deficit Days) – (Available quantity + On Order) See below.
• Days TO Order— Days to which you are multiplying the daily velocity to generate an Add to PO Qty. While the deficit supply will be calculated for all dates, this selection enables a quick add to PO. See below.

Example: Days OF Order 30/ Days TO Order 15

In this example, 108 units sold in the past 30 days (Qty Sold column). Since it is based on a 30-day period, per day velocity is 3.6.

• Deficit Supply 15— Multiplying 3.6 by 15 = 54. Subtract available qty of 28 to get a Deficit 15 of 26. ((108/30) * 15) – 28 = 26 
• Deficit Supply 30— Multiplying 3.6 by 30 = 108. Subtract available qty of 28 to get a Deficit 30 of 80 ((108/30) * 30) – 28 = 80 
• Deficit Supply 45—Multiplying 3.6 by 45 = 162. Subtract available qty of 28 to get a Deficit 45 of 134. ((108/30) * 45) – 28 = 134 
• Same calculation for 60 and 90; see below for explanation of 11/M. 

Since Days TO Order 15 was selected, the deficit qty of 26 was auto-entered into the Qty field to add to a PO.

Generating a purchase prediction

Purchasing predictions can be generated automatically on a schedule; enable the Client Setting Calculate Qty Sold in X days for Predicted Purchasing.

Additionally, the service Product Inventory Movement Service must be running on the server. This service runs once daily around 2 AM. Go to Inventory > Predicted Purchasing and start from Step #4, below.

To manually generate for the most current data:

1. Go to Inventory > Predicted Purchasing.
2. If the services above have been set, go to Step #5.
   If not, or if you want to calculate based on even more recent data, open the Action Menu and choose Generate Data for Sold Since Last Year > Go. This will calculate data for all products in your catalog. Alternatively, you can select specific items on the grid and choose Generate Data for Selected SKUs from the Action menu. SKUs will only appear in the grid after Steps 5 and 6 have been completed.
3. After the job completes, refresh the page.
4. Select the Days Of Order / Days To Order from their respective dropdowns.
5. Check Only show rows with quantity to be ordered to see items with deficit quantities only.
6. See an explanation of warehouse filters in the Client Settings section in this topic.
7. Click Search to populate the grid with items with the following columns:
   1. ProductID
   2. ProductName
   3. ASIN
   4. Avail Qty – Quantity in all warehouses. See the Client Setting below Predictive Restocking: Consider FC Transfer and FC  Processing as reserved for managing “Sellable” FBA warehouses that have inventory that is reserved by FBA because of a transfer, processing, etc.
      1. Note: To exclude from this column any reserved quantities (these are unavailable), you must have enabled the Client Setting Predicted Restocking: Do not include Reserved in FBA Available Qty.
   5. Qty – Enter the quantity to add to a purchase order. See below.
   6. Velocity— Rate of sales per day when calculated by  – Qty Sold in Days of Order divided by Days of Order
   7. Qty Sold— Quantity sold in the selected “Days of Order.” Based on the Client Setting above, it will calculate sales for the days inventory was available. This means that if there were days within the selected “Days of Order” when there was no available inventory, those days will be skipped in the count.
   8. Deficit Supply— The predicted deficit to your supply. This is calculated as follows: (Velocity x Days To Order) – (Available quantity + On Order)
   9. 11/M— Qty Sold 11 months back in time for a full month period. This can be helpful when trying to determine quantities to order for seasonal periods, which are often different than preceding non-season months. For example, if you are at the beginning of November and want to determine how much to order for the coming seasonal month of December, this column will look back 11 months – to December – and display quantity sold for the month of December. This can be used as an indicator if you need to order more qty than was calculated based on the regular velocity.
   10. On Order— Shows how many units are on open Purchase Orders. PO must be set as Approved to consider quantity On Order. It is critical for determining if, and how many units, need to be ordered. After determining how many units are needed for the given time period, both the available quantity and on order quantities are factored in to determine how many need to still be ordered. Clicking the OnOrder Quantity will open the related PO.
   11. Vendor— Default vendor on the product.
   12. Vendor Price— Price of the default vendor.
   13. Amazon BuyBox Price— Retail price of Amazon’s BuyBox. Many online sellers use this information to determine if they can be competitive on a given product, or if they should discontinue the product. Read how to get the Amazon BuyBox.

1. Deficit quantity based on the Days To Order selection will auto-populate in the QTY field; edit the field as necessary.
2. Select the products on the grid.
3. Open the Action Menu and choose Add to PO > Go. You will be prompted to create a new PO or add to existing.
4. Choose your option and click Continue.

Exporting a predictive purchase report

You can create a scheduled task that automatically exports a predictive purchasing Excel file.

You can export a predictive purchase report as an Excel file and save it for future reference and documentation. This report will track a variety of information, ranging from Site Cost to your Average Profit to your Total Quantity Extended.

1. Click Manage Inventory > Predicted Purchasing > Search > Select the items you want to create a report for.
2. Click the Select Action drop down menu > Click Export to Excel > Go.