What numbers are good? Which ones are bad?

In our industry, we strive to eliminate confusion by crafting specifications tailored to the job or application at hand. One of our biggest challenges is balancing specificity with clarity by providing enough detail to be useful without overwhelming or diluting the intent of the specification.

Historically, specs were simple: weight, material composition, and dimensions. These were easy to measure, easy to communicate, and easy to compare. But as materials and applications evolve, so do product characteristics. Today, the numbers we rely on can be misleading if they’re not tied to performance.

Take the classic example of a 2” x 4” framing stud. In reality, it measures 1 ⅝” x 3 ⅝”. Structurally, that’s fine, since it still meets building code load requirements. Adjacent components, like pre-hung doors, have adapted accordingly. The actual dimensions don’t matter, as long as the wall is structurally sound. This is a perfect example of how performance beats precision in certain contexts.

We’ve seen similar shifts in our own space. Consider industrial mineral wool, which is sold by nominal, rather than actual, density. Over time, process improvements have removed much of the un-fiberized material, commonly referred to as “shot,” which reduces the product’s weight while maintaining the fiber content necessary for performance. To maintain consistency with legacy specifications, the nominal reference remains. But this raises a key question: Do we care about weight or performance?

Ultimately, customers are buying performance, not just a list of numbers. So how do we evaluate it? And more importantly, how do we write specifications that reflect it?

Building Better Specs: Key Considerations

Dimensions and Performance Requirements
Let’s revisit the wood stud example. Instead of specifying a material by its nominal size, ask, what performance is needed? If you’re considering different materials, what thickness is required for thermal or acoustic performance? What temperature range must be supported?

These questions help define performance requirements rather than locking in a product that may have changed over time. This approach allows for flexibility and innovation while still meeting the core need. For example, specifying a required R-value or decibel reduction gives manufacturers and contractors room to select the best material for the job—even if the product itself evolves.

Resources such as NIA’s Insulation Materials Specification Chart (www.insulation.org/about-insulation/system-design/techs-specs) offer valuable insights into how different materials perform under various conditions. Specs should reference these kinds of performance benchmarks rather than static product attributes.

Single-Number Ratings versus Spectrum Data
Single-number ratings simplify material selection. Flame spread and smoke development ratings (25/50), R-values for insulation, and pass/fail combustibility tests are easy to understand and compare. But they don’t always tell the full story.

Some performance metrics—like thermal conductivity or acoustic absorption—are better represented as curves across a range. For example, thermal conductivity changes with temperature, and acoustic absorption varies by frequency. These nuances are often lost in a single-number summary.

In acoustics, the noise reduction coefficient (NRC) is a single-number rating derived from spectrum data. Higher NRC values mean better sound absorption, making it easier to compare materials. But if you’re designing for a specific frequency range—say, low-frequency industrial noise—you’ll need to look beyond the NRC and examine the full absorption curve.

For thermal applications, tools like the North American Insulation Manufacturers Association’s 3E Plus® (www.insulation.org/training-tools/3e-plus) or NIA’s Simple Insulation Calculators (www.insulation.org/training-tools/designguide/simple-calculators) help determine the required insulation thickness based on a material’s thermal conductivity data. These tools rely on spectrum data to provide accurate, application-specific recommendations.

System versus Component Ratings
Specifications often focus on individual components, but real-world performance depends on systems. A system typically includes multiple components such as insulation, coverings, fasteners, and sometimes structural elements, each contributing to the final performance.

For example, insulation may reduce heat flow, but a covering might add durability, aesthetics, or resistance to environmental conditions. When it comes to fire resistance or acoustic performance, the entire system matters, not just the insulation.

Customers often ask about the hourly fire rating or sound transmission class (STC) of a specific insulation product. But insulation alone doesn’t carry these ratings: They’re system-level metrics. A 2-hour fire rating or an STC of 55 is achieved through a combination of materials, installation methods, and design details.

And remember, performance isn’t always additive. Doubling insulation thickness doesn’t guarantee double the fire resistance, or half the sound transmission. Specs should define system-level goals, such as a 2-hour fire rating or a target STC/insertion loss, rather than relying on assumptions about individual components.

Good Specs Focus on Outcomes

Specifications should guide users on what to use and what selections are supposed to do. Good specs define the goal, allowing for multiple solutions. This approach encourages innovation, accommodates evolving materials, and ensures that the end result meets the intended performance.

For example, specifying a 2-hour fire rating for a 12” schedule 40 pipe exposed to hydrocarbon fire is more useful than naming a specific insulation product without context. The same applies to acoustic ratings, which vary by pipe size, insulation thickness, and wall thickness.

It’s easy to get lost in the numbers. Sometimes small numbers are good, like thermal conductivity, flame spread, and smoke development. Other times, bigger is better, like hourly fire ratings or acoustic sound transmission results. But when you focus on what the insulation system needs to do, the specs start to make sense.

Conclusion: Move from Numbers to Needs

Bad specs happen when we confuse numbers with needs. Good specs start with a clear understanding of the problem to be solved, whether it’s thermal control, sound attenuation, fire resistance, or durability. From there, we can define performance requirements that guide product selection without locking ourselves into outdated or overly rigid criteria.

In a world of evolving materials and applications, performance-based specifications are not just helpful, they’re essential. They ensure that we’re building systems that work, not just checking boxes. And that’s a spec worth writing.