Innovation Growth Boards at Moody’s Analytics: The Why, What, and How

Danielle Ferry has a succinct answer to what innovation is not: wearing jeans to work and playing beer pong at lunch. Her alternative is more demanding. Teams need to understand a customer problem, test their assumptions and show what they have learned before asking for the next commitment of resources. Senior leaders need to ask questions that fit the uncertainty of the work.

In this episode of Innovation Rockstars, Ferry joins Dr. Christian Mühlroth to discuss the growth board introduced at Moody’s Corporation in 2021 and the accelerator experience behind it. The conversation combines governance with practical product discovery. It also includes a useful contrast: a machine-learning product that moved into an operating business, and a rapidly built supplier tool that earned positive feedback but attracted too few users. Speed, the examples suggest, is only part of the test.

Change the funding conversation

Ferry describes an innovation growth board as an enterprise equivalent of venture capital. Its role is to bring accountability and funding in stages to work that may otherwise receive a large, long-term allocation before key assumptions have been tested.

The approach grew from experiments within the Moody’s Analytics Accelerator. A track record there helped build support for trying the model at the wider corporation level.

The transition was not immediate. Budgets for 2021 had already been agreed when the growth board launched. Ferry explains that teams therefore presented as though the next funding allocation still needed approval, even where money was technically available. The aim was to practise the new decision process before applying it more fully.

That detail matters. The interview describes a system being introduced and refined, rather than a finished model whose every funding decision already operated under new rules.

Ask for evidence appropriate to the stage

In early discovery, the important questions concern the problem. Who has the team interviewed? What did it learn? What evidence supports moving to the next step?

A design sprint might then help the team develop an early prototype and put it in front of users. At that point, feedback and learning are more useful than a precise forecast of returns from a product that barely exists.

Later, as an initiative approaches scaling, customer traction and commercial expectations become more relevant. Ferry’s point is to match the questions to the stage, rather than remove accountability.

Both project teams and growth board members needed training. People accustomed to asking about ROI had to learn what credible progress looked like earlier in the process. The team requesting resources and the executives assessing it needed to understand the same evidence.

Give the board enough authority to help

The growth board brought together senior leaders across functions and operating businesses. Ferry also describes an external adviser who could challenge the discussion without the same internal sensitivities.

Its responsibilities extended beyond funding. Members could help balance the portfolio, remove obstacles and enable collaboration. Hearing several domains present together revealed opportunities and shared problems that might otherwise remain separate.

Hiring provided a concrete example. Teams needed to recruit faster, and a senior HR leader on the board could take that problem back to the relevant people to develop a response. The interview identifies that follow-up as a useful outcome; it does not claim a measured improvement in hiring speed.

“This thing has to have teeth if it's going to work.”

Ferry also warns that setting up and running the board takes substantial effort. Senior backing and ongoing capacity are necessary. The governance model itself must be tested and adapted to the organisation.

Demonstrate the methods before teaching everyone else

The accelerator initially expected to teach other teams how to apply lean startup techniques. That approach met resistance. People did not welcome the implication that they needed an outside group to explain how to innovate.

The team changed course. It used the methods within the accelerator, built examples and let the results create interest. Over time, colleagues began asking for help applying the tools themselves.

Ferry describes several ways the learning could then spread: through the growth board, through an enabling role envisaged for the next version of the accelerator, and through former accelerator members moving into operating teams. Those plans and experiences were at different stages when the interview took place.

The sequence is instructive. A mandate to introduce a method can help, but practical credibility gives colleagues a stronger reason to use it.

Let adoption challenge the success story

One accelerator project Ferry highlights is Quick Spread, an automated spreading tool using machine learning. The team initially worked with an external partner because it lacked the necessary internal capability. She describes the product eventually moving into an operating unit and the work helping build a wider internal machine-learning practice.

The Know Your Supplier Portal tells a different story. During the pandemic, the team built a free tool intended to help healthcare providers assess unfamiliar PPE suppliers. It came together in two weeks and impressed senior management. Positive feedback and promotion followed, but adoption remained limited.

“We have abandoned those aspirations.”

Ferry is referring to hopes of turning that particular effort into a commercial opportunity. She describes healthcare as a difficult entry point for the team, while leaving open other work in the supplier space.

The contrast makes the purpose of staged learning tangible. A fast build can demonstrate capability without proving demand. A promising response from users does not necessarily become sustained use. A growth board earns its value when it helps teams confront those distinctions and decide whether further work is justified. That is a more useful discipline than celebrating every prototype or demanding a confident return forecast before the team understands the problem.