Amazon is one of the most relentlessly innovative companies in the world. Their advances in cloud computing enable Duco to deliver reliable speed, scale and efficiency to financial services firms across the globe.
We hosted our latest Innovation Day event at the Amazon offices in Bangalore, which gave us a perfect opportunity to uncover what makes them such an innovative company.
After all, innovation is a word financial services uses a lot. But doing it well and delivering transformation that actually sticks remains stubbornly difficult. 70% of transformation projects fail to meet their objectives. In capital markets specifically, EY reports that two-thirds of leaders have experienced at least one underperforming transformation in the past five years.
Madhu Ganguly is an AWS Industry Leader for the automotive, manufacturing, healthcare, life sciences, media and entertainment verticals. She very kindly agreed to lift the lid on Amazon’s innovation practices.
Amazon’s business is very different to that of a financial services firm. But it was striking how many of the key principles behind successful innovation map directly onto the transformation challenges we see our clients grapple with every day.
Here are the standout lessons, and why we think they matter for leaders in financial services.
Start with the mission, not the technology
Ganguly was clear about what drives innovation at Amazon: "Innovation really starts [with] the mission to be the Earth's most customer-centric company."
This idea of customer-centric transformation is already at odds with how many firms attempt to innovate. And despite the difference in industries, what she said about this is sure to resonate:
"There are lots of ways in which companies innovate. Some people do technology-based innovation: 'Hey, we have a technology, now what can we do with it?' Some people might be focused on competition. 'Our competitors are doing this, so maybe we should also innovate something around it'. At Amazon, we’ve chosen to centre our innovation approach around obsessive customer focus.”
Research by EY found that the most frequently cited cause of unsuccessful transformation is an unclear vision. In fact, less than half of employees they surveyed said they understood and believed in their organisation's transformation strategy. This is often because, in our experience, financial services innovate for the reasons Ganguly outlined above.
Tying innovation to a company-level mission, as Amazon does, makes it far easier to get the cross-functional buy-in that transformation demands.
Your firm likely has strategic priorities around efficiency, regulatory compliance or client experience. Innovation should be working to advance those goals. And when it does, the whole process becomes much easier.
Amazon uses a remarkable method (called Working Backwards) for codifying the end state of their innovation: they write an announcement for the hypothetical product.
"We take a leap into the future and we write a press release for a product which we haven't even started writing the first line of code," Ganguly explained. These mock press releases must address who the customer is, what problem is being solved, what the biggest benefit will be, how the solution changes the customer experience, and how it can be tested.
You don't need to draft a mock press release. But the principle is sound. Financial firms are so nuanced that there is no one-size-fits-all operating model. The desired end state needs to be concrete and specific if it's going to survive contact with reality.
Build the environment for innovation
Amazon bases its approach to innovation around four pillars: culture, mechanism, architecture and team organisation.
"These four pillars put together provide the environment for each of us to innovate at every level in the organisation," Ganguly said; an idea they refer to as “taking innovation to edge”.
The idea that innovation isn't siloed to a particular role or function is critical. Transformation in financial services often takes a narrow approach. The focus is on small initiatives aimed at solving a problem for a single business unit, rather than unlocking value for the whole organisation.
This is compounded by how banks allocate funding: typically on a function-by-function basis. That supports departmental changes, but undermines bigger innovation initiatives such as operating model change.
For example, one team needs engagement outside of their jurisdiction to progress, but the other team has conflicting priorities. By not creating alignment around common goals, functional agendas dominate and crowd out potentially more valuable outcomes.
Amazon's model is a direct counter to this. Innovation is everyone's responsibility and the structures exist to support it. Firms looking to innovate should ensure that their own processes and culture are set up to foster it.
One solution used by Amazon is to create single-threaded cross-functional teams that ensure buy-in from various stakeholders across the organisation:
"Make the team such that there is enough representation. Have a designer, a database specialist, a product manager, somebody from Marketing, from Legal, from Finance as needed. Make the team such that there is full ownership and information to take decisions."
Embrace failure
Perhaps the most provocative part of Ganguly’s talk was about failure:
"If you are innovating, if you're experimenting, then you really don't know for sure what is going to be the outcome. If you knew what the outcome was, it's not experimentation."
This is where financial services have a well-documented problem. Fear of change - and specifically fear of failure - is understandably a blocker, given how often transformation underperforms.
A lack of psychological safety makes people reluctant to absorb risk. This has only intensified with the rise of personal accountability regimes like the UK's Senior Managers and Certification Regime, where the consequences of getting it wrong can include dismissal or remuneration clawback.
The result is that many firms reason they should keep transformation small in scope to minimise risk. Paradoxically, this is often why it fails. The smaller scope makes it impossible to get the cross-functional buy-in necessary to drive meaningful change.
The lesson from Amazon isn't that failure doesn't matter. It's that you build systems to fail cheaply and learn quickly. In financial firms, that means adopting a phased approach: start small, prove value early, and use those wins to justify the next stage of the project.
Let data drive the decisions
Amazon's leadership principle of "have backbone, disagree, and commit" caught our attention:
"Have backbone is really about being able to state your point of view, your conviction, irrespective of who is in the room. It could be the CEO, it could be your boss, it could be anybody. Idea is to respectfully challenge decisions you disagree with - backed with data - even when uncomfortable, but fully support the final decision once made."
The standout point for us is, unsurprisingly, the need to back up your case using data. This is directly relevant to how financial firms approach transformation, which often focuses on the most visible problems rather than the most impactful ones.
It’s often due to a deficit of data - a lack of real visibility across the tech stack, processes and workflows. This tends to happen because of a variety of point solutions and manual processes. Controls are opaque, often duplicated, and no one trusts the data they’re working with.
Without that baseline of trusted data, transformation projects routinely uncover blockers that weren't in the original scope, extending timelines and blowing budgets. This is why we always advise that new clients start by understanding their current state. This means gathering data, mapping processes and figuring out the gaps in their automation. Only then, with all the information at hand, can they innovate successfully.
The takeaway
Ganguly's talk was a reminder that the fundamentals of successful innovation are universal, even if the industries in question look nothing alike.
Start with a clear, mission-driven vision. Organise so that innovation isn't trapped in a silo. Create the conditions for smart failure. Use data to make decisions. Define a concrete end state. And involve the people who will be affected.
This enables you to avoid common transformation pitfalls and focus on delivering innovation like the best of them.
