Connect with us

Technology

The Unbundling & Rebundling of Banks

Published

on

Share this:

This story originally appeared on iNovia conversations.

“Silicon Valley is good at getting rid of pain points. Banks are good at creating them.” — Jamie Dimon, CEO JPMorgan Chase

FinTech has made massive waves across the world in recent years, with more than 5,000 companies founded, raising nearly $6 billion in venture capital financing.

Let’s take a step back and reflect on why we have seen so many unbelievable entrepreneurs choose banks as the next establishment to disrupt. Of course, the simple answer is that banking is a large sector, with lots of room for improvement; and millennials desire digital experiences in their financial lives. While those are the most often quoted reasons we see in pitch decks today, we believe the real tailwinds behind the growth of this sector lie even deeper.

First off, an attractive quality of a market ripe for disruption is one where the critical infrastructure is already in place for innovation to be built upon. Matt Heiman at Greylock references that critical infrastructure in this post, citing data APIs like PlaidYodlee, and Flinks making it easier to work with financial data; payment APIs like Stripe making it easier to accept payments; and financial market APIs like Xignite, making it possible to pull in live stock prices. Having this foundation in place makes this space all the more attractive for entrepreneurs.

Second is the harsh reality that one’s financial picture today is much different than it was a decade ago.

  • Real annual wages have been stagnant since 2000
  • Education costs are rising exponentially
  • Credit card balances are higher than ever
  • Credit scores are lower than ever, limiting access to capital
  • Home ownership is at its lowest level since the Census starting tracking it
  • High deductible health insurance plans are now the norm
  • Populations are aging and retiring older

As Sarah Tavel points out in this post, called “Saving People Money”, in response to these changing macroeconomic factors, people need new ways to save money, manage money, and invest money. Cue the need for tons of innovation.

Lastly, banks have turned into modern-day conglomerates. The large banks today offer every product you can imagine, from insurance to loans, to mortgages, to cross-border money transfers, etc.

This is a prime example of an institution that does a whole lot of things, but does none of those things really well. This is the perfect landscape for a startup to focus its efforts on a select few of those items and execute to perfection.

The ability for a startup to begin “unbundling” some of those banking services is predicated on the notion that a physical presence in banking is no longer at the core of the customer experience. Alex Rampell, Partner at a16z, makes a great comparison between the financial sector and the retail one, citing Amazon as the retailer that caught these same tailwinds and removed the physical presence from the equation, allowing it to displace WalMart as the biggest retailer in the world. He leaves us with a crucial thought in this video, asking “when will banking have its Amazon moment?”.

These factors have opened up the possibility for thousands of startups to transform the way modern banking is done. Every transformation happens in multiple phases, and the first step in this equation is unbundling. The opportunity at hand is for startups to be exclusively focused on only one banking product (say savings or lending). Consider several examples of this being successful:

Alternative lending platforms such as SoFiKabbage, and Clearbanc have all taken advantage of the inefficiencies in the lending departments of large financial institutions. These lenders all have similar formulas in the way they disrupt traditional banks:

  • High Touch: Start by focusing on a specific user and understand that user really well. This allows the company to underwrite that specific user much more effectively than a bank would. Underwriting more users means more originated loans, and lower loss ratios. For example, SoFi has focused on students, Kabbage on SME’s and Clearbanc on entrepreneurs.
  • Reduce the margins banks earn on loans: A traditional bank accepts money (deposits) and pays a minimum amount of interest (<1%) and then loans that money (primarily on credit cards) for closer to 19% interest. There is a lot of margin there to cut in to for a startup. Given the lenders mentioned above don’t have brick and mortar operations to pay for, they can beat the banks loan rates in most cases, and offer credit to more borrowers.

Robo-Advisors such as WealthfrontBetterment, and Wealthsimple have automated a routine job that was typically done by financial advisors. Large mutual funds and ETFs were already mostly being managed by algorithms, however clients had to consult financial advisors prior to investing. By removing the advisor from the equation and promising individuals a balanced, diversified portfolio that fits their lifestyle, robo-advisors are able to offer equal returns with much less management fees (typically 0.5% vs 1.5% at traditional banks).

  • Savings platforms such as Acorns have found new, innovative ways to encourage people to save money. They are able to offer a fully digital experience, and have ‘gamified’ saving, by incorporating goal setting, rewards points, and a social element. Alternative savings platforms can earn a higher return for clients on their savings all with free accounts that they can contribute to or withdraw from anytime.

These examples continue across every facet of a bank. This image below gives a taste of the FinTech landscape today, and highlights every element of unbundling currently under way. This is the home page of Wells Fargo and it outlines the top startups picking apart every piece of the bank.

CB Insights/iNovia

Of course, unbundling is not the holy grail, it is merely phase one. Entrepreneurs’ ambitions and world domination plans are much larger than simply mastering one banking product. The first product, or the unbundling, is just the “hook” to acquire customers and begin building trust and brand name. Startups exploit the banks on one simple product as a hook to win the consumers’ business in the hopes of then being able to target that same consumer with other financial products in the future, hence phase two: rebundling. The more startups that begin offering additional financial products, the more those startups will begin to resemble traditional banks. Here are some examples of rebundling happening in action:

  • Acorns has now differentiated beyond simply a savings platform by launching Acorns Spend, a debit card product.
  • Square began as an easy-to-use terminal for on-demand workers to receive payment. They have now begun issuing loans to their merchants.
  • Paypal has launched a prepaid debit card that includes bank transfers, deposits, and cashing checks.
  • SoFi has moved beyond just student loans and into mortgages, wealth management, and life insurance.
  • Robinhood has extended its trading services to cryptocurrency
  • Stash has launched core banking and custodial services
  • Credit Karma knows everything (far beyond just credit cards).

What has emerged in the FinTech space is a race to own the end client relationship. Each startup took a different approach, chose a different vertical, and unbundled a different element of the bank. But as all those startups look to layer on, and rebundle the core services of a bank, they will all be vying for mindshare from the same customers. Suddenly, a group of thousands of companies solving various, unrelated problems, will become competitive and will race against each other to be the “go-to” digital bank (or the ‘Amazon Moment’). Despite the numerous examples of rebundling above, we are not quite there yet. As the graph below depicts, we are still at the tail end of the unbundling phase, with startups trying to achieve critical mass in their verticals, prior to commencing the rebundling process.

Once the rebundling phase begins on a macro level, the threat to traditional banks will increase exponentially. Today, consumers excited by digital offerings startups are delivering are faced with the pain of having to piece together all of their financial needs like a puzzle (since every startup only unbundled one product). Getting all of your financial needs serviced, requires interacting with many startups. This pain still generates enough friction for consumers that they maintain their relationship with their traditional bank, and experiment with one or a few new innovative products on the side. Most customers with an Acorns account, also have a traditional savings account at their bank (likewise with investments and loans).

While we don’t expect startups to attempt to put together products that cover everything Wells Fargo offers today, we expect them to bundle a subset of elements that have high synergies. In the past, a HENRY (High Earner, Not Rich Yet) would have one relationship — a big bank; in the future they won’t have 50 relationships (one for each service) but they may have 3–8 relationships with digital rebundlers. Customers will have the opportunity to transfer more and more of their banking relationships to their most trusted digital providers, and will be able to move further away from their traditional banking relationships.

We have yet to see the true threat to banks. But it is around the corner.

Let’s conclude by summarizing what all of this means for inovia in terms of how we allocate capital and make investment decisions in the FinTech space. Here are some of the key elements we look for:

Having a unique and differentiated customer acquisition machine: the ability to acquire customers cost effectively is of utmost importance in the FinTech space. As mentioned above, owning the client relationship is the holy grail, and having a customer not only counts as revenue for the current product, but also allows the startup to target that customer with additional banking products in the future. Here are some examples of unique customer acquisition strategies that have proven successful for FinTech startups;

  • SoFi began by targeting students with its loan products. This led them to be able to use universities as distribution channels and acquire students cheaply (this customer profile was being ignored by traditional lenders). They coined the term HENRY to describe their target customer. This profile was not of interest to banks since they were not wealthy enough (yet) to drive a significant amount of business.
  • Clearbanc offers revenue-based financing to entrepreneurs. They quickly realized that many small businesses use Facebook as their primary advertising channel and that one of the barriers for small businesses is access to capital. Clearbanc partnered with Facebook to help provide capital to these small businesses (much of it to be re-invested in Facebook ads for customer acquisition). This allows Clearbanc to acquire users cheaply through the Facebook merchant network.
  • Affirm allows consumers to pay for large retail purchases in installments. Rather than target consumers directly, Affirm used merchants as their distribution channel. Once a customer reaches the cash, the merchant would ask the customer if they wanted to pay using installment payments (powered by Affirm). This turned the business into a B2B model of selling to merchants rather than a B2C model of competing on customer acquisition.

A well-thought out rebundling strategy that involves owning the end consumer or merchant: Entrepreneurs need to think about pitching the big vision from day one. Building a massive business in the FinTech space will not happen with a series of accidental product additions along the way that we “hope” consumers will enjoy. Owning the end customer should be the objective from day one, it is the core of the business and the reason for existence. Then it is up to the entrepreneur to experiment with various “hooks” to lure in their first batch of customers cheaply. These hooks are more flexible and far less important than the actual master business plan. Here is some advice on choosing the right hook:

  • Test and iterate quickly on initial customer segments you are targeting and the product offering you’re selling. Try something and kill it within a few weeks if you are not luring a unique kind of individual. It is crucial to find a differentiated customer base to initially target, rather than going after the same customers as everyone else.
  • Pick something that resonates with millennials. For example, Ellevestcreates mutual funds tied to the unique career path of women, OpenInvestallows clients to add social impact stocks to their portfolio and Quantopianallows anyone to create financial trading algorithms. The overall vision of all of these companies is to be the trusted financial partner for their target client base, however they have all approached the market with hooks that resonate deeply with that market they are targeting.
  • Once you’ve found a differentiated customer base and a product that resonates with that base you will begin attracting attention to yourself. The idea is that you can use your initial base as a springboard to layer on your rebundling strategies in a more cost effective way. Start with engaged users, build brand awareness among them, garner attention, and then begin rebundling.

Innovate in a new area of banking: Over 40% of all investment dollars into FinTech startups to date have been poured into the alternative lending space, leaving massive industries (such as mortgages and insurance) with few well-funded companies. Additionally, there potentially many innovative ways to improve one’s financial lives that don’t even exist yet and are not even done by banks. Finding a new way to add value financially is a compelling way to disrupt the antiquated banking industry. Examples of radically new financial products are;

  • Mortgage companies like RibbonPoint and Properly that allow consumers the ability to sell their homes more efficiently and even offer the possibility of unlocking some of the equity in their home (things banks don’t do today).
  • Contextualized insurance companies like Lemonade, and Slice. Today, an individual may act as a business one day (renting our their home on Airbnb, or driving their car for Uber) and as a regular citizen the next. Insurance needs to adapt to understand the context in which your assets are being used.

Create your own infrastructure and be self-reliant: Many FinTech companies simply add a new layer or application on top of existing banking infrastructure. This is a great way to validate the problem, but in the long-term the majority of the gains will still accrue to the financial institution serving as the infrastructure layer. FinTechs that are self reliant can be more disruptive and rebundle other apps even easier than those that rely on others. This is one example of a well-planned rebundling strategy from the start.

At inovia we look to partner with audacious founders building enduring technology companies. It is clear that the ability to have an impact on one’s financial experience has the potential to disrupt everything we know about our banking systems. Those are the types of ‘big bets’ we thrive in undertaking.

Share this:

Business

5 tips for brainstorming with ChatGPT

How to avoid inaccuracy and leverage the full creative reign of ChatGPT

Published

on

Share this:

ChatGPT recruited a staggering 100 million users by January 2023. As software with one of the fastest-growing user bases, we imagine even higher numbers this year. 

It’s not hard to see why. 

Amazon sellers use it to optimize product listings that bring in more sales. Programmers use it to write code. Writers use it to get their creative juices flowing. 

And occasionally, a lawyer might use it to prepare a court filing, only to fail miserably when the judge notices numerous fake cases and citations. 

Which brings us to the fact that ChatGPT was never infallible. It’s best used as a brainstorming tool with a skeptical lens on every output. 

Here are five tips for how businesses can avoid inaccuracy and leverage the full creative reign of generative AI when brainstorming.

  1. Use it as a base

Hootsuite’s marketing VP Billy Jones talked about using ChatGPT as a jumping-off point for his marketing strategy. He shares an example of how he used it to create audience personas for his advertising tactics. 

Would he ask ChatGPT to create audience personas for Hootsuite’s products? Nope, that would present too many gaps where the platform could plug in false assumptions. Instead, Jones asks for demographic data on social media managers in the US — a request easy enough for ChatGPT to gather data on. From there he pairs the output with his own research to create audience personas. 

  1. Ask open-ended questions

You don’t need ChatGPT to tell you yes or no — even if you learn something new, that doesn’t really get your creative juices flowing. Consider the difference: 

  • Does history repeat itself? 
  • What are some examples of history repeating itself in politics in the last decade?

Open-ended questions give you much more opportunity to get inspired and ask questions you may not have thought of. 

  1. Edit your questions as you go

ChatGPT has a wealth of data at its virtual fingertips to examine and interpret before spitting out an answer. Meaning you can narrow down the data for a more focused response with multiple prompts that further tweak its answers. 

For example, you might ask ChatGPT about book recommendations for your book club. Once you get an answer, you could narrow it down by adding another requirement, like specific years of release, topic categories, or mentions by reputable reviewers. Adding context to what you’re looking for will give more nuanced answers.

  1. Gain inspiration from past success

Have an idea you’re unsure about? Ask ChatGPT about successes with a particular strategy or within a particular industry. 

The platform can scour through endless news releases, reports, statistics, and content to find you relatable cases all over the world. Adding the word “adapt” into a prompt can help utilize strategies that have worked in the past and apply them to your question. 

As an example, the prompt, “Adapt sales techniques to effectively navigate virtual selling environments,” can generate new solutions by pulling from how old problems were solved. 

  1. Trust, but verify

You wouldn’t publish the drawing board of a brainstorm session. Similarly, don’t take anything ChatGPT says as truth until you verify it with your own research. 

The University of Waterloo notes that blending curiosity and critical thinking with ChatGPT can help to think through ideas and new angles. But, once the brainstorming is done, it’s time to turn to real research for confirmation.

Share this:
Continue Reading

Technology

Multimodal AI promises transformative changes to health care delivery

The health-care industry is slowly adopting first-generation AI technologies, but with multimodal AI on the horizon, it’s time to get ready for what’s next

Published

on

Share this:

Multimodal AI is the next big thing in health care. Except the previous next big thing — regular ‘garden variety’ unimodal AI — hasn’t even fully arrived yet.

“In health care, AI is a promising technology that has been deployed in very small pockets here and there, but it has not broadly impacted health care in a meaningful way,” said Dr. Amol Verma, a physician, scientist, Assistant Professor in General Internal Medicine at St. Michael’s Hospital and the University of Toronto and the 2023 Temerty Professor of AI Research and Education in Medicine at the University of Toronto. 

“The vast majority of clinicians don’t really use AI in their clinical practice.” 

That fact doesn’t mean we’re not at the outset of a multimodal AI revolution in health care, with transformative changes including diagnostics, enabling remote care, and driving efficiencies in an industry not exactly known for them.

But before we get too far ahead of ourselves here — what exactly is multimodal AI?

Multimodal AI systems are capable of processing and synthesizing multiple forms of data in order to provide outputs that include decisions, recommendations, and predictions. 

Just like people.

“We are multimodal,” said Elham Dolatabadi, a data scientist and Assistant Professor at York University focused on machine learning in health care. “We can smell, see, and hear. These are all different modalities. Our brains integrate them to come to one decision.”

So, where a unimodal AI system in health care might only be able to assess MRI scans, a multimodal AI system could process the information from that MRI scan, along with clinical notes, lab tests, genomic data and real-time patient health tracking information, amongst other inputs. In this way, a multimodal AI system would replicate the way actual doctors process patient health information — by considering and analyzing multiple sources of data.

What multimodal AI could mean for health care

Multimodal AI systems could bring almost unfathomable computational power to health care, especially when it comes to diagnostics and predictive capabilities. This could mean a much more personalized and precise approach to health care for individual patients. Think earlier, more accurate diagnoses and better outcomes — all at scale. 

For example, a computer could process radiology images and incorporate that data into its predictions to make them more accurate. It could also incorporate actual conversations between a patient and physician and even assess clinical deterioration from the sound of a patient’s voice.

Multimodal AI could also unlock the potential of telemedicine. Right now, health care providers are limited to assessments based on their conversations and observations over video calls. But if a patient had sensory technology in their home capturing their personal data and feeding it into a multimodal AI system, that could change health care dynamics for millions of patients. 

“It could dramatically improve access to care for people who are far away from clinical care or who, for whatever reason, have a hard time getting out of their house,” said Dr. Verma.

Wait times could also be improved. For example, patients regularly face wait times for ultrasound imaging. That’s partly because operating an ultrasound machine is what’s called ‘operator dependent.’ A clinician needs to be properly trained to use it. But AI could actually help patients capture their own images if it was embedded in a device they could use themselves.

“It could change who can deliver health care, not just who receives it,” said Dr. Verma.

Ethical considerations remain central to AI deployment in health care

For many, the widespread use of AI in health care still seems risky. What about privacy? What about the risks of entrusting critical patient care decisions to AI? 

Shaina Raza is an Applied Machine Learning Scientist for Responsible AI at the Vector Institute, an independent, not-for-profit focused on AI research. Her work focuses on the ethics of responsible AI in public health.

“With critical decisions about the life or death of a person, the doctors should make those decisions,” said Raza. “It’s not appropriate for generative AI models to do that. But if AI is used to facilitate research or help the doctors’ decisions, that’s different. We can save hundreds or thousands of hours that way.”

Raza notes that ethical AI in health care is ultimately about humans creating the frameworks for the AI before it’s ever fully deployed. That’s how best to address issues like patient privacy and systemic biases. 

“Patient privacy is very sensitive. We need to de-identify or mask patient data before feeding it into the AI models.” said Raza. “We can also clean the data for biases before we feed it into the models, with what we call prompt engineering, the instructions we provide to the models.”

Once these sticky issues are addressed, multimodal AI will likely have revolutionary, positive impacts on our health-care system. But since the health-care industry is typically slow to adopt new technologies, that revolution may have to wait a few years.

This might actually be a good thing, according to Dr. Verma.

“Technology is deployed much more widely in the general society than it is in medicine,” said Dr. Verma. “We’re talking about a profession that still frequently uses fax machines. I can do more sophisticated things on my mobile device than I can for applications that are medical-specific. That’s a good thing because the stakes in medicine are very high — literally life and death. We don’t want to be deploying unproven technology so rapidly in that context.”

So while multimodal AI isn’t yet ready for broad deployment, the industry is still preparing for what’s next.

“Our aim is to include as many modalities in the models as possible, including images, text, electronic medical records, wearables, signals like ECG and EEG, and genomics,” said Dolatabadi, who is currently conducting research in multimodal learning with generative AI. 

“Then the models could be used for different applications. And once the models are built, hospitals or health-care organizations can fine tune them for their own applications and patient populations.”

How health-care organizations and providers can prepare for the age of multimodal AI

With multimodal AI a matter of when and not if, it’s incumbent upon everyone in the health-care industry to be prepared for the changes ahead. 

Dr. Verma outlines four key key areas of focus over the next few years: 

  1. Design the right systems to safely deploy AI technologies. “We need to be designing the right institutions, relationships and incentives. That means creating, effectively, centres of excellence with the infrastructure, data and skilled personnel that can monitor AI technologies. These centres would then connect to primary providers, who you can’t expect to assess whether an AI solution is working. That’s just not feasible or practical for them.”
  1. Create a plan for collecting critical data to avoid exacerbating system biases. “We know systems are prone to bias, so the second thing that must happen at a system level is creating a plan to collect data about patient race, language and gender, which we currently do not collect. If we don’t collect that information, we won’t know whether solutions are biased and will therefore perform poorly. We could end up exacerbating biases in our health-care system.”
  1. Building up AI expertise, skills and organizational capacity. “Organizations need to identify AI champions, either through recruitment or upskilling. They need people that have technical skills and skills related to the legality and ethics of AI. And they need people that have skills related to the change management aspects of AI implementation. There’s also the question of scale. Not every organization can do this. Big organizations should scale up and smaller organizations that can’t should partner with big organizations that can.”
  1. Get educated on AI technologies. “I think at the individual practitioner level, people basically just need to become more aware and educated through professional development. Some basic understanding of these technologies would be good.”

Despite all the complexity with AI, there are options for more people to live longer, healthier lives if multimodal AI is deployed across the health-care system. The fact that health-care organizations are still working to embed unimodal much less multimodal AI doesn’t change the trajectory. In the coming years, almost all of us will find our health care experiences enabled by AI in ways both obvious and hidden. It’s the inevitable next step.

Share this:
Continue Reading

Technology

New IT infrastructure for Gilbane includes LLMs, payment automation, and unified tech departments

A look at 153-year old real estate titan Gilbane’s digital transformation

Published

on

Share this:

Canada’s construction industry has been slower to get on the digital transformation train. Yet virtual design and automation presents serious opportunities for the industry to evolve with digital transformation. 

Research shows that more than 80% of construction companies have room to improve their data capabilities, and the most commonly cited benefits of digital transformation were increased productivity, customer experience and staff safety.

One company that made it happen is Gilbane — a U.S.-based real estate development company worth more than $6.5 billion, with family roots that persevere to this day. In fact, they even have employees in the family’s sixth generation. Despite the “legacy” mindset in the company’s fabric (it’s 153 years old) and business industry, Gilbane boosted risk management and productivity under a brand new modular IT infrastructure. 

Here’s how they did it (and how you can, too):

Unify all digital initiatives under one leader

The company created a position that was responsible for overseeing the progress on multiple tech initiatives, including AI, analytics, IT, and cybersecurity. This helped them avoid the risk of wasted budgets through silos and miscommunication. 

“We believe this structure is the most effective to bring together our data and technology resources to drive transformation and get a real return on invested capital.”Karen Higgins-Carter, Chief Digital Information Officer (CDIO)

Invest in analytics and AI for risk management

Safety is a primary concern in the construction industry. Despite improvements in safety measures, equipment, and training, the construction industry still experiences high rates of death and injury. In fact, in 2022 the National Safety Council ranked the construction industry in the top four most dangerous, noting that it experienced the most workplace deaths.

Gilbane’s team is investing in analytics and AI with large language mode experiments to help them identify similar trends that indicate potential unsafe characteristics on a worksite, Higgins-Carter told CIO. “In construction, our teams are managing the construction of hundreds of projects happening at any one time,” she said. “Our analytics capabilities identify potentially unsafe conditions so we can manage projects more safely and mitigate risks.”

“To help us manage risk, I need to understand the leading indicators of risk on a job, like attrition or high volumes of change order.”Karen Higgins-Carter, Chief Digital Information Officer (CDIO)

Automate payment processing with operation-specific triggers

Higgins-Carter told CIO the company recently piloted an automated payment program for Gilbane to pay subcontractors more efficiently. Powered by videos and photos of work completed as triggers, payments are automatically dispensed to the necessary parties. 

Educate the entire team and inform new processes with their experience

Hold meetings and training sessions to ensure executives and employees  understand the benefits and functions of any new tech or business processes. 

“We can’t deliver technology if we don’t understand our employees’ experience. If I go out to a job site once a month, then my team will too.”Karen Higgins-Carter, Chief Digital Information Officer (CDIO)

Read the full article on CIO here.

Share this:
Continue Reading

Featured