A new way to calculate scale prices

Most manufacturers leave thousands in supplier savings on the table. The culprit? Suboptimal scale prices that no one has spotted.

Every year, procurement teams overpay suppliers simply because their price scales don’t reflect real ordering behavior. A small misalignment there can mean hundreds of thousands of dollars. 

Let’s take a simple example. 

My supplier gives me three price scales:

• 1,000 kg -> $10.00/kg
• 10,000 kg → $8.00/kg
• 25,000 kg → $7.00/kg

But here’s the problem: I rarely order exactly those quantities. In reality, I often buy 5,800 kg at once, because of constraints like expiry dates or warehouse space. On the official scale, that still costs me $10.00/kg.

Now, apply GenLots’ completely new Dense Price Scales approach (see visual below). Suddenly, we can “fill in” the missing price points between 1,000 and 10,000. For 5,800 kg, the realistic price would be $8.56/kg.

That’s nearly 15% savings on every order – without changing suppliers, contracts, or operations. Just by negotiating the right scale.

This translates into immediate, tangible savings. But how to sift through 1000s of materials and check out those opportunities? How to know which scales to pick and which prices to negotiate?

GenLots is at its heart an algorithm company. So we have developed an algorithmic, data-driven way of doing this! In the following we trade our simplified example with a real world one and go much more into the details in the following 3 steps. 

1. Extrapolation to dense scales
2. Simulation of ordering patterns
3. Analysis & Automation

Step 1: Extrapolation to dense scales

In the first step, we actually create what we call “Dense Price Scales”.  Meaning that we calculate each price between the minimum and maximum order quantity which can be ordered (multiple of rounding values). 

You can think of dense price scales like upgrading from a blurry photo to high-definition. Instead of seeing just 3 price points, in our example you see hundreds of possible ones (1256 in our example) – and hidden savings emerge. 

In our example from before, 5800 kg here means around 8.5607 $/kg – already almost 15% of price savings. 

The methodology used is a combination of different algorithms¹ .  Once we have this new, dense price list, we get to the next step

Step 2: Simulation of ordering pattern

This is where our OrderAI comes in.  In short it recommends orders for inbound materials, taking all operational constraints into account and optimizing for the lowest total cost of ownership. More can be found here, but for the purposes of this article, just know that we ideally trade-off all cost constraints, such as inventory, order costs, waste, out-of-stock risk and – important for this new feature – price. 

If you don’t currently have GenLots OrderAI, we have a solution for you.

We now simply run a simulation in our simulator, one with the old, limited pricelist and one with the new, dense pricelist. Result looks as follows:

As is usual with GenLots, the effect on all the costs is immediately visible:

We trade-off a bit more inventory for almost $ 60 000 in price savings, or 6.4% of spend for this material. This is with a negotiation where your supplier can probably easily say yes to as it is within their existing quantity-price curve. 

Step 3: Analysis & automation

We can see in the application the use of different price scales.

According to the table we would recommend negotiating an additional scale at around 30 000 kg and around 37 500 kg.

We automatically calculate optimal price scales every morning. Your strategic procurement can take out a report and start negotiating quick wins at any time! 

This report shows immediate 250k in price savings for packaging over the next 6 months alone for an existing client. Here a screenshot of our report, which can be customized: 

What should I do with this information?

• Are you already using GenLots? Good news: This feature is already available today – no extra setup required. It suffices to contact your strategic buyers with this valuable information.
• You don’t have GenLots: Contact us for a one-time optimization here. We can easily extract flat files from SAP and then do a one time simulation, including scale price advice.

1. Footnote for the technically interested: This is actually not as straightforward as it sounds. If you have only 2 existing price points, most algorithms draw a line across two points and you are very quickly in negative territory. If you have a set of 30 existing price points, you want to model them closely, so they don’t suggest different prices from what you already have. And you want your Price-Quantity curve to realistically become flat with very high quantities and avoid negative prices.