Creating a Cost Classification System

November 9, 2016 Michael Doyle

The nature of product development is that details about components emerge … evolve … and solidify over time.

intro pictureA component may start out as a concept sketch with rough dimensional information. Over time that sketch is rendered as a 3D solid model. Over still more time that 3D solid model specifies manufacturing properties such as tolerances that must be met to ensure the component meets all form, fit and functional requirements.

Even after all those details are nailed down by engineering and control of our component passes to sourcing, there are manufacturing process options that can additionally impact cost: will the component use plastic with a colorant? That affects our material and possibly process costs.  How will the sheet metal part be nested? That affects our material utilization rate, which in turn affects our material costs. Will there be one operator for the progressive die machine, or can that operator watch two or three machines because those machines use electronic monitoring? That affects our labor cost.

Because manufacturing cost estimates are driven by the granularity of manufacturing details available when calculating them, the nature of cost estimates in product development is that they also emerge … evolve … and solidify over time.

Should-Cost Estimates are Already (Roughly) Standardized

The most common type of cost estimate we hear about in procurement is a should-cost estimate. But what, exactly, is a should-cost estimate?

A should-cost estimate is a calculation of what it should cost to purchase efficiently manufactured goods or efficiently delivered services within a specified market (note the emphasis on efficiently). It is much easier to establish a should-cost for “off the shelf” goods and services such as cars or televisions or cleaning services. The real challenge with establishing a should-cost estimate comes with custom purchases … such as purchasing custom-designed and manufactured components from suppliers that cannot be purchased off-the-shelf.

To mitigate the purchasing risk of such components, it is fairly standard practice to collect three supplier quotes and evaluate the purchase decision from there. But such an approach does not establish a should-cost baseline because it does not directly address efficient manufacturing.

The Inherent Nature of Classification Systems

Every classification system allows for information it organizes to be more manageable … but an adopted classification system further allows us to infer additional information (metadata) based solely on an object’s classification. To understand this better, let’s quickly review a classification system many of us are already familiar with: the Kingdoms of Life.

When we look at the classifications used over time to represent different biological Kingdoms (courtesy Wikipedia), we see that the classification system itself is as evolutionary and adaptable as the life it classifies.  Over time, the classification became increasingly more detailed as incremental information was discovered, and was available to expand our understanding of the animal kingdom.

animal kingdom

If we collectively understand a classification system, we know all these attributes merely because of the animal’s classification. And even when classification systems are evolving (as is the Kingdom of Life classification system), we still adopt them because of the benefits of a standard classification system.

Cost Classification System

So let’s imagine, for a moment, how we might create a cost classification system that works for everyone. The illustration below shows the input granularity requirements, business uses, and functional owners of a proposed cost classification system.

cost classification

If SHOULD-cost estimates represent the finest granularity of input because they address all cost drivers, what would the coarsest granularity be? In engineering companies, that would typically be a hand sketch (or similar) concept drawing, possibly including rough dimensional information. Let’s call this a CONCEPTUAL-cost estimate. Perhaps the placeholder cost is established by evaluating costs for similar components. Possibly it would be owned by the Project Manager at this point.

Between our two extremes on the granularity continuum is where our cost estimates typically reside longest and change most frequently: in engineering during the creation of an initial 3D solid model up until our component reaches a “make vs buy” decision.

Projects have three common constraints: cost, scope, and schedule. How those constraints are balanced ultimately affects our product’s quality.

Because geometric cost drivers are designed into our components early and difficult to remove later, there are numerous software tools and methodologies emerging that focus on just this timeframe in engineering, collectively categorized as Design-to-Cost (DTC).

During design we typically don’t know absolute cost impact to changes in our solid model’s geometry, and we don’t know the absolute cost of alternative designs. We can, however, still understand relative cost impact at this point: did our costs go up? down? up a lot? down a lot? So let’s call this a DIRECTIONAL-cost estimate, and likely the cost estimate would be owned at this point by the Design Engineer.

And finally, once a cost estimate turns into an actual cost, in finance that is referred to as STANDARD cost … the averaged historical cost of our component that might be updated annually for financial planning.

An Adaptable System Carrying Important Metadata

Such a cost classification standard would account for all cost estimates at the macro level, from the most rudimentary conceptual placeholders to cost estimates with details that let us enter into line-by-line negotiations with suppliers if the cost-reduction effort warrants it.

Our classification system would inherently carry important metadata about the use of cost estimates based on their classification: for example, if we are working with an estimate classified as DIRECTIONAL cost, we can tradeoff feature decisions or compare alternatives, but we can’t yet use it to negotiate final price quotes with our suppliers because we haven’t nailed down all cost drivers.

Or if we are working with an estimate classified as SHOULD cost, Sourcing would have a detailed cost estimate that addresses all cost drivers and can be used for detailed cost negotiations with suppliers.

And our classification system would be adaptable. For example, we could create custom sub-classifications of cost estimates wherever our workflow required them. Perhaps we want two classification levels below DIRECTIONAL costs: Tolerances & Surfaces and Manufacturing Setup.

The Real Power: Value Tracking and Business Intelligence

The real power of a cost classification system traces to our ability to use value tracking and business intelligence tools to understand and manage costs throughout a product’s lifecycle. This disciplined ability is the backbone of a Product Cost Management (PCM) solution.

For example, if we tracked the initial DIRECTIONAL cost estimate and the final DIRECTIONAL cost estimate and plotted them against each other, we could see what costs engineering avoided

If we added an additional data-point of a cost target, we could not only track costs that engineering avoided, but how those costs track against target.

If we tracked chosen and contending supplier quotes against the final SHOULD cost estimate, we could understand how each supplier’s quote compares to the efficient manufacturing cost baseline in a given market as established by that SHOULD cost estimate.

Having a standardized cost classification system would allow standardized value tracking and business intelligence tools to be used for cost management.


Product manufacturing cost estimates are driven by the granularity of manufacturing details that are available at that particular point in time when the estimate is calculated.  Thus, the nature of cost estimates in product development is that they emerge … evolve … and solidify over time.  By designing and implementing a Cost Classification system that sets proper expectations and identifies appropriate use cases for a cost estimate as a product design evolves from concept to manufacture, companies can effectively implement a culture of cost consciousness across an entire, distributed global organization where everyone understands the inherent assumptions and value of the information they are consuming, and how to use that information most effectively.

Want to Learn More?

Read This Whitepaper about Creating a Cost Conscious, Profit Centric Culture in your product development organization.



Previous Article
6 Metrics That Will Help Improve Your Engineering Productivity
6 Metrics That Will Help Improve Your Engineering Productivity

Metrics are essential to a healthy business. For engineering teams, these metrics will empower you to measu...

Next Article
Strategies for Successful Design-to-Cost Initiatives
Strategies for Successful Design-to-Cost Initiatives

Interview with aPriori Client Engagement Manager, Rick Hyde

Conduct a Design/VAVE Workshop

Learn More