The two most common product development metrics are R&D spending as a percent of sales and revenues from new products over time. These metrics, although necessary to track, do not predict a result. There’s no linear relationship between R&D spending and new product success. For example, a company might allocate ten percent of its R&D budget toward products in brand new technologies and markets. This allocation should reflect the company’s appetite for risk, the balance in its product portfolio between incremental and entirely new products. While it orients the team in the right direction, the R&D spending number does not gauge the effectiveness of that investment.
Similarly, the revenues from new products over time metric reports on past successes. It does not help forecast the future. It’s typically measured at a corporate or business level, perhaps organizational layers beyond the project team, down in the trenches where decisions are made that add or subtract from the project’s timeline every day.
It’s essential to measure these corporate metrics, but they’re not motivational at the team level and they do not tend to predict results. Predictive metrics help guide the team’s day-to-day decision making. They measure the team’s ability to execute toward defined goals. For example, a tool we have used called the deliverable hit rate chart tracks completed tasks against their target dates. It shows whether programs are on schedule by tracking the rate of task completion. This metric is dynamic, i.e. it is updated throughout the program and indicates the progress of a process over time, rather than measuring a result, after the fact.
A company that wants to discover quality predictive metrics for its teams should first consider the desired outcome. What behaviors would the company like to see in its teams, and what can be measured that will motivate the team to produce the desired outcome? Take these desired organizational changes and divide them into small, incremental steps. Think in terms of a sequence of steps that create larger-scale improvements in team effectiveness. Then choose the most effective levers that will motivate the team to make a small step toward those outcomes and have a target metric for each improvement. When the metric indicates that an incremental step-improvement has taken place, then move on to the next step.
Step-by-step these small improvements should follow in a logical sequence that builds into a larger improvement initiative that unfolds over time. As these improvements show results, they then spread throughout the organization, taking root in other teams.
The most effective predictive metrics for teams are often behavioral. For example, if the team is using a new milestone process, we measure the percentage of teams that use the milestone’s new names in their reports. These simple but precise metrics help companies track how teams are behaving today, rather than the success they enjoyed yesterday.
One caveat: remember to test your metrics. Pilot them with a team before propagating them elsewhere. The adage is true: “You get what you measure.” And what “you get” is sometimes an unintended and undesired consequence. Experiment, test, and pilot before any undesired consequences get out of hand.
The very best metrics for product development are strategic in that they support specific behaviors in the wider system that serve broader goals. Most companies are tracking far too many metrics. They become onerous and hinder more than help. The best product development metrics systems incorporate the critical few predictive metrics that drive decision making at the team level.