DDBA 8511 Week 8 Innovation Metric Critique Example

A Measure That Cannot Register Failure: Critiquing an Industrial Adhesives Maker's Revenue-From-Recent-Launches Metric, Asking What Result Would Show Its Innovation Program Failing and Proposing a Replacement That Could

Student Name

Walden University

DDBA 8511: Seminar in Innovation Management

Instructor Name

Month Day, Year

A Measure That Cannot Register Failure: Critiquing an Industrial Adhesives Maker's Revenue-From-Recent-Launches Metric, Asking What Result Would Show Its Innovation Program Failing and Proposing a Replacement That Could

The Measure in Use

Brenmore Adhesives, a composite maker of industrial adhesives and sealants for packaging, furniture and construction customers, with about 450 employees in Pennsylvania, judges its innovation program by a single number: the share of each year's revenue earned by products introduced within the prior thirty-six months. The company's target is 25 percent, and it has met or exceeded the target in each of the past four years, reaching 31 percent in 2025. Bonuses for the research and product management staff depend partly on it. A "launched product" is any product with a new product code entered into the company's catalog.

What Result Would Show the Program Failing?

A good measure of an innovation program should be able to show that the program is failing. If the measure rises or holds steady whatever the program does, it is measuring something else. Popper (1959) argued that a claim earns scientific standing only if some possible observation could refute it; the same test applies to a performance measure. So the question is: what would Brenmore's number look like if its innovation program were producing nothing new?

The answer is that it could still look fine. Review of the 2025 launch list shows 47 new product codes. Of these, 29 are reformulations of existing adhesives with a different viscosity or cure time for one customer, 9 are existing products in new package sizes, 6 are renamed products moved into a new branded line and 3 are adhesives based on chemistry the company had not used before. The 3 new-chemistry products account for about 2 percent of revenue. The other 44 account for 29 percent. If Brenmore stopped all research into new chemistry tomorrow, its product managers could continue to meet the 25 percent target by issuing new codes for variants. The measure cannot register failure.

Why the Measure Behaves This Way

Kerr (1975) showed how organizations end up paying for conduct they did not want while hoping for conduct they do not pay for, and Brenmore fits the pattern. Brenmore hopes for new technology that will keep it competitive. It rewards new product codes. Staff, reasonably, produce new codes. Hagedoorn and Cloodt (2003) studied measures of innovative performance and found that common indicators, such as research spending, patents, patent citations and new product announcements, overlap considerably but each captures something different; relying on any one leaves parts of performance unmeasured. Brenmore relies on one, and the one it chose is the easiest to inflate.

A Replacement That Could Return a Failure

I propose replacing the single number with three measures, each defined so that it could show failure.

First, revenue share from new-chemistry products, defined as products using a polymer system, curing method or raw material base the company had not commercialized before, approved by a review panel that includes someone from outside research. Target: 5 percent within three years. If the program produces nothing new, this measure falls to zero.

Second, the number of new-chemistry projects that reach a customer trial each year, with a target of four. This leading indicator would show a stalled program long before revenue does.

Third, a kill rate: the share of exploratory projects stopped at each review gate. A program that never stops anything is probably not taking risks; a rate between 30 and 60 percent would suggest real exploration with real selection.

What the Replacement Costs

The new measures are harder to collect, need a review panel and will almost certainly show lower numbers than the current one, which may upset staff whose bonuses depend on them. The bonus link should be phased over two years, and the old measure can be kept for reporting customer responsiveness, which is what it actually measures.

How the Classification Was Made

The 47 launches were classified from product specifications and formulation records by two people working separately: a senior chemist and the author. They agreed on 44 of 47 and settled the other three by discussion. All three disputed cases involved reformulations with a new additive; they were counted as reformulations rather than new chemistry because the base polymer system was unchanged. The classification is reproduced in an appendix so others can check it. Anyone who disagrees with a particular call can see exactly how changing it would move the totals.

Limits

Any measure can be gamed. A review panel can be lenient about what counts as new chemistry. The proposal reduces the problem by making the definition explicit and including an outside reviewer, but it does not remove it.

References

Hagedoorn, J., & Cloodt, M. (2003). Measuring innovative performance: Is there an advantage in using multiple indicators? Research Policy, 32(8), 1365-1379. https://doi.org/10.1016/S0048-7333(02)00137-3

Kerr, S. (1975). On the folly of rewarding A, while hoping for B. Academy of Management Journal, 18(4), 769-783. https://doi.org/10.2307/255378

Popper, K. R. (1959). The logic of scientific discovery. Hutchinson.