Category: Continuous Delivery

If you asked an airline executive what their biggest expense was just a few decades ago, they would’ve easily answered, “Jet fuel.” These days, you’d be hard-pressed to find an executive in any industry that wouldn’t answer with, “Software.” But here’s the difference: With jet fuel, that executive would have told you that they can fly x number of seat miles per gallon/liter of fuel (or pounds of fuel, if we’re being technical). With software development, the expense is becoming more material–and far more variable–but the return on that investment is rarely clear. Over the past decade or more, software development has readily assumed its position as the fastest-growing (and often largest) expense line item in virtually every industry. All companies are adding software experiences to or around their products, and they need the people and tools to build, support and continuously optimize those experiences. And that costs money–a lot of money.

The problem is that the language, the metrics and the world in which software development exists do not translate into operational or financial terms that business leaders can readily understand.

“We invested X million dollars into software development last year,” a CFO might say. “And what do we have to show for it?”

Meanwhile, a technology leader might respond with, “We’re doing X deploys a day!” or, “We wrote X thousand lines of code!”

They might as well be speaking different languages because, in practical terms, they are.

And the problem is, that you can’t manage what you can’t measure. To continue our airline theme, software development continues to be a ‘black box’—this hidden-in-plain-sight source of information—that financial executives can’t seem to peek into.

Why the Black Box?

It’s not likely that technology leaders are being purposefully elusive toward business leaders–it’s more likely that they simply are not aligned on what, fundamentally, they need to do to speak the same language.

The question at hand is, actually, “Are these investments we’re making turning into our desired business outcomes?” But the question technology leaders feel compelled to answer is, “What do your people do all day?”

These are questions that Agile and DevOps don’t solve–translating Agile or DevOps metrics (also known as DORA metrics) into actual, actionable metrics for business leaders to understand. Although these metrics are tremendously useful for teams, using them to steer the business is a mistake.

Misaligned in purpose and using different language, business leaders and technology leaders talk past each other as software development costs continue to rise.

But the cost of software development is too high for business leaders to continue to tolerate this ‘black box’ of ever-rising costs. It’s the fiduciary responsibility of financial leaders to understand the value of software investments and figure out how they are affecting business outcomes–or their jobs could be on the line.

In case this seems like a distant threat, it isn’t: I saw a headline recently about a prominent automobile executive who lost their job because they weren’t able to gain the necessary insight into, and business outcomes out of, their significant software development costs and that was affecting the competitiveness of their entire company.

It’s happening. And it’s on both business and technology leaders to break that black box open and start speaking the same language.

The next question, of course, is, “How?”

How to Bridge the Communication Gap Between Technology and Business Leaders

For far too long, business leaders have made do with the activity metrics provided by technology leaders–they’ve tolerated this lack of visibility and found other ‘strategic’ ways to justify the rising software development costs. Perhaps they’ve been questioning how A leads to B–how an increase in funding for software development might result in greater customer retention or app usage, for example–but they have not had a tangible way to prove to themselves, much less to investors, whether this is actually happening.

In order to understand and quantify software development costs, we have to start demanding more from our technology leaders: Working with them to establish realistic objectives and key results (OKRs), and holding them accountable to business metrics that actually relate to outcomes. That means insisting on business results–and not accepting activity metrics in their stead.

Stable-Fund Products, Not Projects

Here’s a hard pill for finance to swallow: When it comes to delivering software-enabled customer experiences, it’s not about optimizing costs. It’s much more nuanced than X dollars in equals X dollars out. Funding by projects might help make the numbers neat, but it doesn’t reflect the reality of the business.

Project-based funding also tends to keep project leaders in a cost-center mentality: It makes it seem as though all resources are fungible; as if the work and the people performing it are interchangeable.

One of the goals of Agile is to maximize value for the customer. But when we fund by project, we often stop funding a piece of work–declaring it done–before it actually starts generating value. We have this false end date that doesn’t reflect the actual ending of the value that we funded to create.

Instead, finance organizations should move to the stable funding of products: Funding an outcome and giving teams the ability to stop, change or extend their work to achieve the business outcomes of the product.

This is a shift that one prominent UK bank has been going through in recent years–from traditional portfolio management to lean portfolio management, including the shift to stable-funding projects. You can learn more about how NatWest Group (formerly Royal Bank of Scotland) has made this shift in this webinar.

Visualize the Flow of Value with Flow Metrics

Finally, we have to stop treating the collection of technology metrics and business metrics as two separate, unrelated efforts. Agile and DevOps are great at accelerating the throughput of your software development … but vanity metrics around how quickly teams are delivering are not enough to justify the growing and material investment in software development.

Finance leaders have the opportunity to be the Babelfish between technology leaders and business leaders. As mentioned above, one way to encourage clearer communication between teams is to use OKRs to naturally tie software development objectives to tangible business results. The more we practice drawing connections between technology and business metrics–the closer we get to finally breaking open that elusive black box.

A key part of this is communicating operational metrics differently: Developing a core set of metrics for tracking the flow of business value in software delivery, and providing a mechanism that correlates the investment in flow for each product’s value stream with the business results for that value stream.

Visualizing the flow of value in a software development process may not be as easy as watching value flow on a manufacturing floor–but that doesn’t mean it’s impossible. Thanks to the vast number of data and visualization tools at our disposal, it is possible to make software products and their value streams just as visible as production lines.

The Flow Framework, created by Dr. Mik Kersten, addresses the pivotal challenge of breaking open the black box of software development and bridging the gap between business and technology to optimize the flow of business value to customers.

Break Open the Black Box with Value Stream Management

The black box of software development is, at its core, a communication issue. It’s on both business and technology leaders to bridge this communication gap. Business leaders need to demand more from technology leaders and insist on translating activity into business metrics. Technology leaders need to practice tying their efforts to tangible business outcomes to help justify their growing expenses.

Of course, it’s helpful to have tools purpose-built to facilitate this visibility and communication.