Financial data, as traditionally defined, is a good tracker of outcomes but a poor guide to how a business operates and achieves results. Everything that happens in an organization ultimately falls to the bottom line, which makes finance about much more than just dollars, says Siqi Chen, co-founder and CEO of Runway Financial. That’s a wider territory than many finance teams care to explore.
For current and future CFOs, this expanded view needs to be reflected in models and forecasts. Data fluency (aligning messy data across departments) and cross-functional insight (understanding how various departmental models impact the bottom line) will be essential competencies for finance executives. And software solutions need to progress, too.
We asked Chen, a former VP of growth at Postmates, head of product at Zynga and CEO of Sandbox VR, to explain how and why organizations need to tie together data to get a more accurate picture of a business.
What’s AI going to do for finance in the foreseeable future with regard to an organization’s data?
One role for AI, besides automation, is to make sense of the complexity of your data, including the various types of data generated and how they all fit together. What kind of question comes up often for a CFO? A number changes for an account or a line item from one month to the next. Why did it change? Finance probably needs to trace the formula back to Excel and drill in. However, AI is starting to comprehend how a model works in Excel—not only what has changed in your model and the data but also the underlying context. That’s a pretty dramatic [development] in AI models, and it’s occurred over the past couple of months.
What kind of “data fluency” will a CFO need?
It’s an understanding of all the things happening in your company and all the decisions being made. For finance, that is usually captured in the HR system, the GL (spend and revenue) and probably the CRM. But there’s a lot more data inside a company: marketing data, product usage data (usually stored in a data warehouse) and engineering cost data (which may be in AWS), for example.
There is also contextual unstructured data, like a product roadmap or a strategic plan, and that’s probably in a [workspace like] Notion or a Google Doc. The concept of data fluency involves understanding what is reflected in the data and ensuring it’s clean and captured in a helpful manner. And then being able to put it all in one place to make sense of it. You also need to understand the quality of your data, how it flows through the system and how to improve it.
To truly understand what’s happening inside the company and translate that into a financial model, finance needs to collaborate with other departments within the company.
At the finance level, most models are concerned about budgets and spending. At an operational level, though, it could be product development—where are we going to allocate our resources?
One of the key outputs of a product development organization, for example, is a product roadmap. What kinds of levers does a product roadmap pull? Maybe a new feature or version will increase contract values or improve sales conversion rates. We want to live in a world—and I think this world is within reach—where a product or engineering manager should be able to say to finance, “If we do X instead of Y, this is going to be the impact on margin and growth two years from now.” A shared understanding and alignment of the business is more important than ever.
What’s preventing that from happening?
Many companies can’t imagine having those conversations today. But people want people to know the implications of what they do and [tie that to] financial outcomes. If you think about the default tools that are available for finance, however, how is finance going to share that information? I think it’s fundamentally a product and tool limitation.
This interview first appeared in the June 27 issue of CFO Leadership’s Finance & Accounting Technology Briefing.
