Flybits’ Hossein Rahnama: ’90 percent of digital transformation budgets are spent on managing IT complexity’

Banks talk about deriving insights from their customer data. They want to provide actionable automated advice to their clients but they struggle just wrangling the data together.

Flybits wants to help banks by taking their proprietary customer data — like the activities a customer has done in her account — and adding in machine learning and external contextual data — like her risk profile or her consumption preferences. The goal is to deliver relevant and timely recommendations to FI cusotmers.

The technology and vision behind Flybits was born in a university lab. Founder and CEO Hossein Rahnama’s PhD was about understanding context in large data sets. He spun out his tech to launch the company. He’s also a professor of computer science at Toronto’s Ryerson University and a visiting prof at MIT’s Media Lab in Cambridge.

Subscribe: Apple Podcasts I SoundCloud I Spotify I Google Podcasts
The following excerpts were edited for clarity.

Research spinoff

The foundation for Flybits came from a research lab and my PhD research, which was about understanding context in large data sets. I spun that off about five years ago and now, Flybits is a venture backed company headquartered in Toronto with offices in New York, San Francisco, London and Dubai.

One of the key drawbacks of AI is the lack of domain knowledge and understanding context. For example, it's very hard for an AI engine to understand the needs of a parent when she's at home versus that parent's needs traveling across borders. You can train AI algorithms to understand and predict patterns but understanding that context and domain knowledge has always been an impediment.

My research was about bringing context and the world of ontologies into the AI world and really apply that to digital channels. Your context basically determines the behavior of the experience that you'll see on a digital channel.

How Flybits helps

If you look at consumer, you see the advent of AI-based concierge services like Siri, Google Assistant, and Alexa. The question for banks is how to create concierge capabilities on their own channels. Most of them want to keep their data but come up with predictive services on their own digital channels, personalizing services while keeping their customers' data secure. They want to be a lifestyle concierge engine for their customers.

Flybits simplifies data unification, aggregation and synthesis. Instead of expecting a bank to over-rely on their IT department, we give them tools so that product and marketing can focus on creating use cases. A common story we hear from digital executives is that 90 percent of their digital transformation budgets is spent on IT complexities, creating a challenge to bring these use cases to market. Flybits essentially empowers that digital executive to bring those creative use cases to market without having to worry about the underlying complexity.

What TD Bank has done with AI concierge

We have 16 very large banks around the world on our platform. TD Bank is one of our best customers. The bank is known for its customer service at the branch level. They spoke to us a few years ago about serving Millennials with high-touch customer service in their digital channels.

We created something for them on their mobile app called TD for Me. Think about it as an embedded Siri on their app that can understand the needs of their customers and proactively provide relevant services. For example, if my mortgage is up for renewal in the next three months and I'm walking by a branch that has an available mortgage specialist, I receive a message on my device with an invitation to visit the branch to renew my mortgage.

If a customer commits to receiving services from TD, the bank can figure out if a customer is looking to purchase a new home. Then, at the right time and on the right channel, TD actually provides a pre-qualified mortgage offer based on the customer's credit score and the type of house they're looking to buy. TD partners with a lot of data providers in the community to provide great service to the customer. In a way, they're becoming a Smart City provider and not just a financial services provider.

The customer buying cycle

Our classic economic buyer is typically a chief digital officer or chief marketing officer. These executives have customer mandates and want to grow engagements and focus on experience design. Recently, we're seeing interest form chief data and chief information officers who see Flybits can help to de-silo their data organization. They use our data orchestrator instead of spending millions of dollars on creating data lakes.

We want our customers to see results right away and we define pilot stages in one to three months. We demonstrate our capabilities right away and then we empower their team from the second phase on to work with our tools. So, we've positioned ourselves as an experience design tool rather than just a data platform or CRM.

Conversational interface

We don't impose any restrictions on the presentational layer of the channel. So, it can be a conversational or chatbot interface or a classic UI or card layout. It could be push notifications. We even have a client that's using Flybits in a humanoid robot. We're one layer down from the presentation layer. There are so many PFMs but they're grammar and passive analytics mapping and pattern matching.

You can use Flybits to empower these capabilities to also become contextual. We've worked with a chatbot provider so that a wealth customer receives a different conversational bubble compared to a student applying for his first credit card. That's what I meant that context is missing from a lot of PFMs. We can help them become more relevant and impactful.