Agile can’t be measured. We hear that a lot from fans of waterfall development.
We respectfully disagree. It’s true that waterfall provides excellent metrics for benchmarks like spending against budget and progress along a timeline. But more often than not, waterfall development is too rigid to thrive amid today’s perpetual changes in markets and technologies.
That’s why DMI usually recommends an Agile approach to software development, which provides speed-to-value in ways that waterfall typically cannot deliver. We have to admit, however, that we have shared the frustrations of developers who craved useful data on Agile performance.
That’s why we developed APIX (Agile Performance IndeX), which offers the ability to quantify and score multiple benchmarks to help managers track the success of Agile projects. The numbers are a bit softer than the ones you get with waterfall projects. But they can be extremely effective when you apply them to a prioritized list of the things you want to measure the most.
We prioritized what we measure with the Agile index like this: value > quality > progress > productivity
Understanding these priorities — and putting them to work — requires a fundamental shift in perspective. Here’s a quick look at our thinking on these priorities:
Why value ranks first: Agile methodologies were developed to enable speed-to-value. You build a minimum viable product (MVP) as quickly as possible to start producing value — whether it’s revenue, market share, other business priorities or a combination of those things. Then you use short development iterations and frequent customer feedback to keep making the product better. Thus, any attempt to measure the performance of Agile projects must place the highest priority on building value.
Why quality ranks second: Quality represents your ability to deliver an MVP with the most critical functionality and the fewest bugs. Quality has to rank below value because you have to deliver an MVP within a reasonable time frame at a sensible cost. You also have to reduce the risks of failure or delays. High costs, long time frames and overlooked risks all erode value and the total cost of delay grows exponentially over time. Thus, when you quantify value first and measure quality in the context of your value metrics, these numbers help drive success in the next two categories — progress and productivity.
Why progress ranks third: Progress is among the easiest Agile metrics to quantify. Burndown charts, for example, provide an excellent illustration of how far along an Agile team is. Because Agile projects tend to be difficult to quantify otherwise, Agile team leaders tend to emphasize progress. You measure what you can see, after all. But troubles can arise if teams make excellent progress on projects that are not delivering value and full of bugs and hence unlikely to deliver speed-to-value. Thus, progress must support value and quality.
Why productivity ranks fourth: Productivity is among the most crucial measures of performance in any activity. The challenge in Agile projects is that measuring productivity can have unintended consequences. For instance, if you try to quantify the productivity of individual team members, you may miss their ability to help the entire team succeed. Productivity and progress can be interchangeable, depending on the needs of your organization and the skills of your team members. Regardless of your approach to these two metrics, you need to make sure they reflect their contributions to quality and value.
When DMI developed APIX, we acknowledged that metrics drive behaviors. People automatically adjust their priorities to meet numerical goals. Thus, we established our Agile index to prioritize the behaviors required for Agile success. We also built flexibility into our APIX parameters — we tweak them according to each client’s business requirements.
As we see it, progress and productivity are empty data points until they support quality and speed-to-value. In our experience, folding these numbers together in order of priority can deliver metrics that meet or exceed anything a waterfall project can deliver.
— Brian Andrzejewski, director of business transformation services