Taking Your Stage Gate to the Next Level

    Amy Muller

    The stage gate process has been a cornerstone of product development since the 1960’s bringing efficiency and control to the development activities. As applied to innovation, the stage-gate process often yields incremental innovation. The same controls, metrics, gate reviews and screening mechanisms that bring control to the process also tend to weed out the true outliers – those game changing innovations that cannot be easily managed with traditional criteria and controls.

    The “fix” for this lack of truly new innovations has been to bolt new idea generation mechanisms on the front end of the process. Recent trends in open innovation, online idea management, and corporate venturing have increased the quantity (and sometimes the quality) of ideas at the “fuzzy front end” – but the back end (stage-gate) process used to bring those ideas to market is still broken. The very controls meant to improve efficiency and control (and to protect management from the embarrassment of blown budgets and missed deadlines) have driven learning and adaptability out of the process.

    Despite its benefits, the stage-gate approach has led to a lot of unintended consequences that inhibit innovation. Stage-gate processes assume that a lot of the critical information about an opportunity is knowable at the outset. For most truly innovative projects, this just isn’t true. The result is that employees are “encouraged” to make up and defend fantasy figures. Second, most stage-gate processes are built around a “typical” project … and we all know that most game changing innovations are not typical. The effect of this forced project “uniformity” is to subtly or non-so-subtly encourage all projects to fit the mold – thus squeezing out the truly innovative aspects of the opportunity until it becomes another incremental improvement. And finally, there is the Go/Kill decision, which, when applied literally leaves little room for re-scoping a project or altering a project based on new learnings.

    So how can we adjust the stage-gate process to be more innovation friendly? Here are three principles that can help bring learning and adaptability to the stage-gate process, while still preserving efficiency and control.

    1. Let assumptions drive the process

    Re-conceive the process as an assumption driven process centered on learning rather than simply a sequence of activities marching toward an outcome. Ask yourself, “ What do I need to learn about the opportunity at this stage of development, in order to proceed to the next stage at an acceptable level of risk?” Then build your tasks, deliverables, and expected outcomes to focus on the key assumptions under test. With a flexible, assumption-driven approach, in contrast to a rigid task-driven approach, the project team is focused on learning about the opportunity and progressively reducing its uncertainty.

    2. Diverge, Converge, Repeat

    So now that we have the critical assumptions identified, how do we test these assumptions? The first step (of each stage) is to hypothesize possible insights, solutions or outcomes (diverge). Next, design and conduct experiments to test these hypotheses. Finally synthesize (converge) the outcomes of the experiments and use the knowledge to refine the offering or underlying business model. The diverge-converge approach can be applied in any stage of development, but the “experiments” performed will differ widely. For example, during early concept development, design teams often generate numerous renderings of a product and then test (experiment) these designs with consumers or other stakeholders, and then use the results of these tests to choose the designs to move forward. In later stages, the product can be launched into select markets and extensive observations (both qualitative and quantitative) are taken. Based on the results of these experiments, changes or refinements to the opportunity are made as assumptions prove valid – or not.

    3. Gate Reviews: Focus on Metrics AND Learning Objectives

    Now that we have identified the key assumptions and validated them through experimentation in the Stages, it is time to turn to the Gate Reviews. Gate reviews that focus on metrics at the expense of learning can inhibit the natural (and often non-linear) evolution of an opportunity. But by focusing on learning objectives in addition to metrics, the downstream potential of an opportunity can be harnessed in at least two ways. First, by adjusting and refining the business model and the value proposition in line with market learning, the opportunity has a better chance of succeeding with customers. Second, a learning based approach gives you more creative options than a simple “Go/Kill” decision. With greater understanding of the issues facing an opportunity, a team can decide to adjust and recycle an idea, salvage an idea (harvest the useful technology or IP for other projects), spin off or license an idea, or possibly find a more appropriate home for the idea elsewhere in the company.

    Thomas Edison once said, “The real measure of success is the number of experiments that can be crowded into twenty four hours.”1 He understood that learning is essential to the development of raw ideas into viable commercial applications. To the extent that companies are fighting against their stage-gate processes to learn about opportunities, the stage-gate process has fallen short. We have traded learning for process for process control and in so doing have discarded adaptability as well.

    1Andre Millard, Edison and the Business of Innovation, John Hopkins University Press (1993).