R&D Productivity: How to Target It. How to Measure It. Why It Matters.

By Gerald Dundon

Based on work from the frontline of high-tech business, this book describes a new approach to targeting and measuring research and development (R&D) productivity. Using logic and basic math, life cycle revenue and profit multiples of R&D project spending can be calculated that are intrinsically allied with corporate goals for annual R&D spending, revenue growth and profitability. The book describes how to measure and track R&D productivity performance versus target and how to interpret and report on variance.

• Language: English
• Publisher: BookBaby
• Release date: May 8, 2015
• ISBN: 9780986152511

Book preview

patience.

Introduction

PRODUCTIVITY MEASUREMENT IS A well-established science. It measures the output of an operation per unit of input. Production processes are well suited to the application of productivity measures when the output of a production process is discrete and measurable and occurs relatively close in time to the deployment of the input resource. Production cycle time can vary depending on the type of product being produced—a PC can be assembled in hours, a jet engine takes considerably longer—but the inputs and outputs are quantifiable and relatively easy to attribute. We know how many welding rods were used in the assembly of the jet engine, so we can associate that input spending to the output.

Measuring the productivity of research and development (R&D) endeavors is considerably trickier. The first difficulty is one of time separation of inputs and outputs. The time elapsed, from the deployment of input resources to a particular product development initiative to the time the last dollar of revenue is received from the products resulting from the development effort (the economic output), can often stretch over decades. This long delay in measuring the final achieved productivity of any R&D initiative tends to undermine the usefulness of the metric itself, unless appropriate economic milestones can be established during the lifecycle of the project. The time separation problem even calls into question the units of measure used to express economic value. Since the primary measure of economic output—cumulative dollars of revenue or profit—is distanced in time from the input, we revert to expressing the output as net present value, discounting the future cash flows to express their current value.

The second problem is one of attribution, of associating inputs to outputs. Unlike production processes, where inputs and outputs can readily be associated, when input resources are spent on research there is no guarantee that what is learned or discovered will be of any practical use for developing a saleable product. So, what output should we associate this input with? We can’t ignore the fact that we spent R&D input, but we don’t have an identifiable output vehicle that we can readily attach that particular spending to. Even in the case where research yields a new competency that can be used for the development of new products, attribution problems still arise. For example, assume we use this new competency to develop three new products. Two of the products fail in the market and produce little or no revenue. One of the products is successful. Should we attribute the input research costs only to the successful product, or equally across the three products that used that competency? What about future products that rely on this new competency? Associating the input R&D spending to applicable outputs that have a measurable economic return (product sales, royalties, licensing revenue from intellectual property, etc.) is not a trivial task.

In part because of these difficulties, it is not uncommon for companies to employ several different measures of partial productivity to track the efficiency of their R&D efforts. Partial productivity is the term used when only some of the inputs or outputs are used in the calculation, and when those inputs or outputs are not expressed as their economic value. A typical example of a partial productivity measure is total labor hours used to develop a new product, normalized in units of complexity. This yields a measure of design complexity per man-month. Another example is patents produced per total R&D costs. In partial productivity measures, the economic value of the output is not being measured, but instead the completion of a task, like a patent filing, or the release of a product, is used as the output. Such measures are useful because they help overcome some of the issues of time and attribution.

But for any business, economic value dollars of output per dollars of input is the ultimate measure of total R&D productivity. Appropriately targeting and measuring total dollars of cumulative revenue and/or profit returned for a product or portfolio of products, per dollar of total R&D costs to develop those products, is the holy grail of R&D productivity measurement.

Product-oriented companies set business goals for revenue growth and profitability and give guidance on the amount of annual revenue they are prepared to spend on R&D to fund that growth ambition. In doing so, they are implicitly setting total productivity goals for R&D. To meet that revenue growth and profitability ambition, the new products produced must return specific levels of cumulative revenue during their lifecycles, per dollar spent on R&D. Yet, as we will discover, very few companies are satisfied with the metrics systems they have deployed to explicitly target and measure total R&D productivity. Many companies concede that they do not even have such metrics systems in place, and that raises some interesting questions. If we cannot target and measure R&D productivity correctly, how do we hope to observe, evaluate, or improve it?

Time and attribution are the two main factors that make total productivity metrics difficult to apply to R&D endeavors. But they are not impossible obstacles to overcome. This book describes one company’s journey to overcome these difficulties and develop a useful measurement system for R&D total productivity.

The book is divided into two parts. Part One is aimed at a general and executive audience who want to understand more about the topic of R&D productivity measurement, why it matters, how a business can benefit from measuring and reporting R&D productivity, and the challenges and solutions to setting appropriate productivity targets that are allied to corporate business goals.

Part Two is aimed more towards financial analysts or operations management—those charged with deploying such a metric system. It discusses the shortcomings of the increasingly prevalent R&D performance metric: percent of revenue coming from new products, where new products are defined as those introduced in the last n years. It explores the mathematical underpinnings of the newly derived cumulative revenue return multiple model in more detail. It discusses limitations of the model, and shows product sales data to support the generalized S-curve of cumulative revenue.

[1]

The Challenge

INNOVATION, PERFORMANCE, AND EXCELLENCE are the cultural pillars on which Analog Devices Inc. (ADI) has built one of the most enduring companies within the technology sector. Acknowledged industry-wide as the world leader in data conversion and signal conditioning technology, Analog Devices serves over 100,000 customers, representing virtually all types of electronic equipment. With annual revenue of approximately $3 billion, the company has an impressive history of growth since its founding in 1965.

ADI believed that improving R&D productivity was vital to the company’s future success, and that correctly targeting, measuring, and reporting new product productivity performance was fundamental. After an extensive search through published literature and engagement with consultants on how other companies measure R&D productivity, we were surprised and disappointed with what we found. There was a marked absence of published literature on this subject, and very few companies reported having a satisfactory method of targeting, measuring, and reporting their R&D productivity performance. My tenure as VP of New Product Productivity at ADI therefore began with an imperative to find a meaningful way to measure R&D productivity.

Product-oriented companies survive and grow by introducing new products that have compelling value propositions. Without that stream of innovative new products, the growth of any product company will slow and the business will ultimately fail. New products, and the new technologies on which they are based, are realized as a result of investment in R&D. R&D spending is the engine that drives the growth of product companies, and R&D