When it comes to Supply Chain Analytics, it's important to know a few different things:
What are the business needs?
What data do we have available?
How do we calculate it?
This glossary is designed to give you the answers to the third question. How do you calculate your analytics? For each of these supply chain focused metrics, we'll discuss our way of calculating and some potential nuances of that metric.
Weighted Payment Terms
Definition: The Average time it takes to pay your suppliers, weighted based on spend.
Use Case: In the world of Procurement Analytics, payment terms often reigns 2nd only to cost savings. This is the number that Supply Chain teams use to impact Free Cash Flow, with the goal of creating a negative cash conversion cycle
Beware: Suppliers usually hate extending terms. And it costs you money indirectly becuase they have to pay In order to extend terms, you need to build a ton of trust and ensure that the business relationship is beneficial to both parties.
How WPT is calculated:
Download all your invoice data. Make sure it has Spend, Invoice Date, Pay Date
Calculate the days between Pay Date and Invoice Date. This is Days to Pay
Spend * Days to Pay = Weighted Spend
WPT = SUM(Weighted Spend)/Sum(Total Spend)
Bonus: How to calculate Weighted Payment Terms, grouped by Supplier:
Spend * Days to Pay = Weighted Spend
WPT for one supplier = SUM(Weighted Spend for one supplier) / SUM (Total Spend for one supplier)
Extra Bonus: If you're a supplier, I highly recommend diversifying your business so that you have a portion of revenue at Net 0 days, likely for small and new customers, and a portion of revenue at the extended terms with your larger customers.
Cycle Count Accuracy
Definition: The accuracy of your warehouse processes, comparing system to reality. The inverse is essentially "inventory loss"
Use Case: This is typically the most influential metric with a direct financial impact for Warehouse Analytics.
Beware: The scale for "pass/fail" is very different than the American school system. If you're at 93% accuracy, that's a failure. In some instances, 97% is failure. We don't personally like the 0-100 scale because it distorts expectations. The REAL range should be 90-100.
Decisions before the calculation:
Do you measure accuracy by # of SKU's that are wrong, total $ lost, or total Quantity lost?
Do your operations completely stop when you cycle count?
It should really be "Accuracy over [x] amount of time"
If you have MORE quantity in stock, how do you incorporate that?
How to calculate total $ lost without a stop in operations, losses only:
For each part number, record system quantity vs actual quantity
System Quantity - Actual Quantity = Quantity Lost (Gained)
Lost Inventory = Quantity Lost * Unit Price <-- note, if your inventory is serialized against PO transactions, you may not need this step. You can use SUM(Received Value for all the serial numbers that were lost)
SUM(Lost Inventory) / SUM (Total Inventory) = % lost
1 - % Lost = Cycle Count Accuracy
% of Revenue tied to a supplier
Definition: The % of revenue (or profit) tied to a specific supplier, via the bill of material.
Use Case: This is used to measure risk. If one supplier is tied to 100% of your revenue, you should ensure that you have a ton of contingencies. You can also modify this to be "country of origin" to measure reliance on one country.
Beware: Two suppliers could both be at 100%. Let's pretend that all your revenue has a cardboard box and a shipping label, and you have one supplier for boxes and one for labels. Both of those suppliers would be 100% of your revenue. This isn't necessarily bad because there's plenty of cardboard box suppliers. However, this should be clearly known and contingencies should be built.
How to calculate total % of revenue with a supplier:
Download your sales data, with part numbers
Download your bill of material data, with one column being "parent part number that is sold" and the other being "sub part number that is purchased".
Download purchase order data tied to part numbers
We're going to pretend that each part number has one supplier. There are ways to calculate multiple, but for the sake of this exercise, we'll keep it simple.
Create mapping of purchased part number, parent / sold part number, and vendor
Create a distinct list of parent / sold part number and vendor
Pivot Table to get sum(revenue) for each parent part number, grouped by vendor
Divide vendor revenue by total revenue
Definition: The percentage of orders that are shipped without error, compared to the total number of orders shipped.
Use Case: Shipping accuracy is a crucial metric for any company that fulfills orders. It measures how effectively a company is fulfilling customer orders, and can have a direct impact on customer satisfaction and retention.
Beware: In addition to measuring the percentage of orders shipped without error, it's important to also track the reasons for errors, such as incorrect address or missing items. This can help identify areas for improvement and reduce the likelihood of future errors.
Total number of orders shipped without error = 100 - 5 = 95