Metrics For The Rest Of Us
Metrics for the Rest of Us
You can't manage what you can't measure ... and you can't measure without metrics to measure to. But what metrics should you use? Especially when there are literally thousands to choose from? It's a hard decision to make, but since such a decision cannot be made without a set of metrics to choose from, this wiki-paper is going to present some of the more common basic metrics and discuss some of their advantages and disadvantages, so that a decision can be made.
After all, with energy and raw material prices rising rapidly, the value of certain currencies falling more rapidly than most of the world expected, and outsourcing on the rise, it's more important than ever to get one's house in order and identify savings opportunities. The first step, as always, is to create a baseline, and that, of course, requires metrics.
Cost Avoidance and Reduction
Let's start with the cost reduction metrics, since most business believe these to be the most important, even though there ain't no saving in a perfect world.
Unit Cost Reduction
This is the one of the simplest cost reduction metrics that one can define. It is simply calculated as:
(last year's unit cost - this year's unit cost) x units purchased
However, it's also the most dangerous. If the lowest cost is from the supplier furthest away, then this could cause freight costs to considerably increase. Thus, some organizations will measure:
Landed Cost Reduction
This is the next simplest cost reduction metric. Where landed cost is defined as unit cost plus freight cost per unit, this metric is calculated as:
(last year's landed cost per unit - this year's landed cost per unit) x units purchased
However, it's still quite flawed. When an organization is buying globally, it also has to consider import duties and tariffs, export duties and tariffs, temporary storage costs, and so on. Thus, some organizations will measure:
Total Landed Cost Reduction
This is the next cost reduction metric in a sequence from simplest to most complex. Where total landed cost is defined as unit cost plus freight cost per unit plus tariffs and duties per unit plus incidental transportation costs per unit, this metric is calculated as:
(last year's total landed cost per unit - this year's total landed cost per unit) x units purchased
However, this is still not perfect. There are still storage costs if JIT (Just-In-Time) delivery is not an option, utilization costs if processing is needed or if there is associated waste, and if there are quality issues and a certain number of units will have to be tested and returned, then there are associated costs that should be taken into account. Thus, some organizations will measure:
Total Cost of Ownership
This is about the most complex, and complete, cost reduction metric that can be defined. Where total cost of ownership is defined as total landed cost plus total inventory costs plus total utilization costs plus total reverse logistics costs, this metric is calculated as:
(last year's total cost of ownership per unit - this year's total cost of ownership per unit) * units purchased
However, it's not always about cost reduction. Sometimes costs are going to go up, and there's nothing that can be done about it. The key in this situation is to restrict cost increases to an amount that is less than the market average. This brings us to our first cost avoidance metric.
Basic Unit Cost Avoidance
This is straight forward to calculate.
(average cost increase per unit - actual cost increase per unit) * units purchased
where the average cost increase is based upon accepted market indexes.
However, this is not the best metric because it does not capture why the cost increased and whether the average cost increase was justified. A better metric takes into account the expected cost increases based upon the market cost increases in the labor, raw material, and production (energy) cost increases.
Advanced Unit Cost Avoidance
This involves calculating the expected increases based upon the increases in the appropriate market indexes and the percentage of unit cost they contribute to. Where last year's raw material cost per unit equals % raw material cost * total unit cost, last year's labor cost per unit equals % labor cost * total unit cost, last year's production cost per unit equals % production cost * total unit cost, and last year's other cost per unit equals % other cost * total unit cost, then a complete calculation would be along the following lines:
(last year's raw material cost per unit * (1 + average % increase in raw material costs) + last year's labor cost per unit * (1 + average % increase in labor costs) + last year's product cost per unit * (1 + average % increase in labor costs) + last year's other cost per unit) * total units purchased this year - this year's total cost.
Of course, all the cost reduction and cost avoidance in the world is not going to help if the negotiated savings are never realized. That's why an important metric is realized savings.
Realized Savings & Avoidance
This is simple to calculate.
(Last Year's Total Cost Volume Adjusted - This Year's Total Cost) /
(Last Year's Total Cost Volume Adjusted - This Year's Negotiated Cost)
last year's total cost volume adjusted = last year's cost per unit * this year's volume
Of course, the key to realized saving is contract compliance.
The goal is to have 100% of spend on contract for a contracted commodity or category. For a given contract, this is calculated as:
Commodity Spend on Contract / Total Spend on Commodity
For all contracts, this can be calculated as:
Total Spend on Contract / Total Spend on All Contracted Commodities
It's also important to get a global picture of the reach of your efforts. This is captured by the spend under management metric.
Spend Under Management
This is a straight forward calculation.
Total Spend Under Management / Total Spend
In the last section we covered some of the most common metrics, which are also often the most important metrics to the CEO and CFO. However, cost isn't the only aspect one needs to be concerned about. Although it is the most noticeable impact on the bottom line since procurement is usually the largest cost on the balance sheet, process improvements are also important as not only do they reduce operational expenses, but they also allow a department to get more spend under management without increasing labor costs. This section will cover a number of common process metrics relating to orders, delivery, and customer service.
Sourcing Cycle Time Reduction
On average, how long does it take to progress through a sourcing cycle from initial need identification to final award, and how much shorter is the cycle compared to last year's average cycle time reduction? This metric, which is quite important after the introduction of a new system or process, is calculated as:
Last Year's Average Cycle Time - This Year's Average Cycle Time
The average sourcing cycle time should eventually reach a steady state consistent with the average sourcing cycle time required by best-in-class organizations.
Average Lead Time Reduction
This metric captures, on average, how much lead time procurement requires to get a product to engineering. Although this metric is usually fairly steady over the lifetime of a contract, it is a very important metric to track and one that should be reduced, if at all possible, when a contract comes up for renewal. This can be calculated as:
Average Old Lead Time - Average New Lead Time
Payment Cycle Reduction
This metric is very important to calculate after a new payment system or process has been implemented and a significant number of payment cycles have been completed. It is calculated as:
Old Average Payment Cycle Time - New Average Payment Cycle Time
Manual Order Reduction
This metric becomes critical if you install new software that automatically re-orders when inventory drops below a certain level or a certain period of time has past. It measures the percentage reduction in the number of times a human has to (intervene and) manually cut an order. It is calculated as:
(Number of Orders Manually Cut Before System Implementation - Number of Orders Manually Cut After System Implementation) / (Number of Orders Manually Cut Before System Implementation)
From an operational perspective, there are a number of other metrics that should be considered since they will ultimately impact the sourcing and procurement organization's ability to lower costs and improve process efficiency. Some of the more relevant metrics are covered in this section.
Simply put, how accurate the forecast was compared to actual orders. Since forecasted volume usually influences prices, and failure to reach certain levels can lead to lost discounts and rebates while exceeding certain levels can lead to lost leverage opportunities during negotiations, forecast accuracy is important. This is calculated as:
(1 - |forecast volume - actual volume| / forecast volume) * 100 %
This measures what percentage of commodities that have valid, active contracts relative to the percentage of commodities that are supposed to have valid, active contracts.
(commodities supposed to have contracts - commodities with contracts) / commodities supposed to have contracts
Order Accuracy Rate
How many orders are 100% accurate? Easily calculated as:
(Accurate Orders / Total orders)
Simply put, how often are the products the company needs delivered on time? This is calculated as:
(On-Time Deliveries / Total Deliveries)
Average Number of Suppliers per Commodity
Too many suppliers, and you've lost volume leverage opportunities. Too few, and you could be introducing unnecessary supply risk. This metric is calculated by summing up the number of unique suppliers for each distinct commodity and then dividing by the number of commodities.
Percentage of Suppliers Accounting for 80% of Spend
It's important to have the right number of suppliers for each commodity, but it's even more important not to have too many suppliers supplying the top 80% of Spend, since these are the suppliers that need to be actively managed. This goes beyond rationalizing by commodity, and even category, but rationalizing by spend as well. Thus, if the same three suppliers can supply nuts and bolts, and the additional cost of having the same three suppliers supply both nuts and bolts as compared to having three suppliers supply nuts and three different suppliers supply bolts is miniscule, and nuts and bolts constitute a lot of the corporate spend, it is worth consolidating the supply base across these spend categories as this will reduce operational expenses and enable improved performance. This metric is easily calculated as:
Number of Suppliers Supplying top 80% of Spend / Total Number of Suppliers
Overall Supplier Performance Level Improvement
Assuming that the organization has a supplier scorecard or other appropriate supplier performance rating system, then this metric can be calculated as:
(current supplier rating - previous supplier rating) / previous supplier rating
There are two ways to measure this. The first is to calculate the % of diverse suppliers:
number of diverse suppliers / total number of suppliers
and the second is to calculate how much spend is allocated to diverse suppliers:
spend allocated to diverse suppliers / total spend
Overall Quality Level Improvement
Assuming that the organization has a supplier quality rating system, this metric can be easily calculated as:
(current quality level - previous quality level) / previous quality level
This section is addressing customer service from an internal perspective - how well the procurement department is servicing its internal customers.
Perfect Order Percentage
This is a straightforward calculation:
perfect orders / total orders
This is another straightforward calculation:
orders filled / total orders
Average Incident Resolution Time
This is simply the average amount of time it takes to resolve a problem reported by one of the internal customers of the procurement department.
Average Number of Incidents per Contract
This is simply the average number of issues that arise per contract.
What percentage of internal customers are reasonably satisfied with procurement?
Asset utilization metrics are important as they create a snapshot of how well operations are progressing from a financial perspective.
Average Cost per Contract
This is simply the average cost of negotiating, creating, and executing a contract. This should decrease whenever software or process improvements are implemented.
Cost per Invoice
This is the average cost of processing an invoice. This should decrease over time until the average cost is literally pennies.
Operational Cost per Employee
This is the average operational cost per employee. This should decrease over time.
This is simply:
(1 - |funds disbursed - funds allocated|/funds allocated) * 100 %
Michael Lamoureux, Ph.D. of Sourcing Innovation