A featured contribution from Leadership Perspectives, a curated forum for enterprise technology leaders, nominated by our subscribers and vetted by the CIOApplications Editorial Board.

RBC Capital Markets

Innovation through Scientific Experimentation

In his book, ‘Think Again’, Adam Grant showcases the power of thinking scientifically, where we accept how little we know and consistently rethink our understanding of the world based on new knowledge from hypotheses and experiments. While there are times where the three other mindsets of preaching, prosecuting, and politicking might be relevant, performing digital transformation as a scientist is critical.

Some of the largest multinationals have massive R&D budgets to test novel concepts and products, while financial institutions tend to be more cost-conscious when it comes to innovation, particularly around selecting the right strategic platforms in the short, medium, and long-term. Firms like RBC, Fidelity, and Blackrock are industry-leading for their successful investment strategies, but due to their successes in running low-cost experiments before spending millions of dollars on technology investments.

So, what does running an experiment look like at a bank? Do we need white lab coats and glass beakers? Proof-of-concepts are the most common types of experiments, with your organization likely already completing several this year. PoC’s could be a good tool if you want to efficiently test how a “new technology” such as AI or distributed ledgers fit within your existing architecture, or if you need to filter past the salesmanship and marketing materials of prospective vendors by running a bake-off based on your success criteria.

Alternatively, you can also run a proof-of-value or PoV where you have little to no doubt on the solution capabilities but need to validate the prospective benefits and ensure you have an accurate business case. Thirdly, scenario analysis or simulations offer a relatively low-risk option to test the feasibility of new solutions or changes. Regulatory bodies, such as the Canadian Securities Administrators (CSA) have regulatory sandboxes that offer fintech’s seeking to offer innovative products and services an opportunity to test their applications throughout the Canadian market while obtaining exemptive relief from securities laws requirements.

Before providing a methodology on how to innovate through experimentation, let us first explore the benefits.

Firstly, testing new technologies and concepts through experiments is cheaper, faster, and easier than the sluggish, expensive, and risky alternative of implementing a solution based on intuition and marketing materials. Next, your organization can foster a culture of innovation and sense of ownership by creating a shared understanding of the hypothesis to prove in a PoC. Lastly, knowledge is a powerful raw material for any organization. Even if the experiment was deemed unsuccessful (i.e., new solution is not a viable fit for the organization’s goals), there is an influx of insights around customers, markets, operational processes, and technology ecosystems.

I have been fortunate enough to work for some fantastic organizations in the wealth management and capital markets industries, with a shared vision of prioritizing clients’ financial goals while providing an effortless experience. Finding the right balance between experimentation and accelerated value delivery is critical; here are a few steps to aid anyone looking to kick-start the scientific mindset in your organization:

Start by mapping out the capabilities within the function you support, so you have a good sense of what areas to prioritize. Once the vision and strategy are defined, it is up to your innovation and digital transformation practitioners to turn that dream into a reality.

PoC’s could be a good tool if you want to efficiently test how a “new technology” such as AI or distributed ledgers fit within your existing architecture


Intake: Managing innovation starts here, so rely on process and product owners in providing change opportunities as a starting point, albeit not necessarily the only approach.

Discovery: As always, it is pivotal to ensure workstreams are clearly defined, expectations with key stakeholders are aligned, and commitment is gained from all partners, including oversight teams (e.g., Compliance, Risk, Legal, etc.). Rushing this step means you may be the only unbiased party in this experiment, so ensure there is an open-mindedness to explore new solutions.

Design: The essence of any experiment – what are the hypotheses you are trying to prove/disprove? Ideally, I would encourage you to develop cross-functional and target-operating-model-agnostic success criteria. For example, if your current setup involves your IT department creating and maintaining all executive reports/dashboards, you might want to assess if that approach is possible but would also want to test if your business users can be empowered to create their own reports for a speedier process. This step is also important to start creating a high-level business case (e.g., Solution can reduce time-to-market by X%).

Execution: Once the logistics are finalized, it is time to oversee the experiment. Try to keep the focus tightly knit and avoid scope-creep or delays, by reminding your users that your experiment is only proving a concept/value/capability, and so does not require a comprehensive view of every possible scenario.

Prepare for Go-Live: Hopefully you have started to observe benefits with a future implementation of this solution and have kicked off discussions with Procurement (and other relevant teams) to onboard the solution. A good next step would be to plan a minimum viable product (MVP) to continue learning with each phase. Additionally, monitor any delivered benefits to your KPI’s and pivot your roadmap, if needed.

To recap, rapid experimentation of new technologies and vendors is critical to delivering quick wins (and losses), while setting up a foundation for an iterative development of digital transformation. You might still need to ‘preach’ the benefits of your selected solution but can now offer much-added confidence to your leaders that an exhaustive assessment methodology was undertaken, and you are implementing the right tools to move your organization forward. When done right, a digital transformation program is akin to a caterpillar turning into a butterfly. Alternatively, not selecting the right tools and maintaining a biased thought process could turn your program into a faster caterpillar.

The articles from these contributors are based on their personal expertise and viewpoints, and do not necessarily reflect the opinions of their employers or affiliated organizations.