Banks now have access to more data and the means to profit from that data than ever before.
The problem, according to new McKinsey research, is that they are failing to meet the challenge. As a result, they risk disruption from new and non-traditional competitors.
Meanwhile, the gap between leaders and laggards just goes on getting wider.
McKinsey cite these cross-industry challenges:
- Incorporating data-driven insights into day-to-day business processes.
- Attracting and keeping enough qualified and experienced data scientists and change agents.
- The losers are falling further behind as winners spin-up new business models that are able to leverage network effects to quickly dominate new markets.
- Data is a valuable asset in its own right. It’s coming from an ever wider range of sources including the web, billions of phones, sensors, payment systems, cameras, and a huge array of other sources. The trick though is to combine this data in new ways and add value through relevant analytics.
- Data and analytics done right is enabling new business models that offer personalised products and services via personalised customer journeys.
- As if all that isn’t enough, machine learning and deep learning are about to profoundly up-the-ante..
Meanwhile, here are the key opportunities and threats for retail banks:
- There is enormous potential for retail banks via massive data integration. Benefits include “better cross-selling, the development of personalized products, dynamic pricing, better risk assessment, and more effective marketing“. McKinsey estimate a potential economic impact of $110 billion to $170 billion in the retail banking industry in developed markets and around $60 billion to $90 billion in emerging markets.
- Non-bank financial service players can benefit from so-called ‘orthogonal data’—such as non-financial data that provides a more comprehensive and detailed view of the customer. “These players may have large customer bases and advanced analytics capabilities created for their core businesses, and they can use these advantages to make rapid moves across sector boundaries. Alibaba’s creation of Alipay and Apple’s unveiling of Apple Pay are prime examples of this trend.“
- A large retail bank in the United Kingdom used machine learning algorithms to identify fraudulent transactions with more than 90 percent accuracy. In another example, a large payment processor deployed machine learning on its extensive transaction data to identify “mule accounts” involved in money laundering.
- Other examples include Apple using “its unique data, infrastructure edge, and product platform to push into the world of finance with Apple Pay. Similarly, Chinese e-commerce giants Alibaba, Tencent, and JD.com have leveraged their data volumes to offer micro-loans to the merchants that operate on their platforms. By using real-time data on the merchants’ transactions to build its own credit scoring system, Alibaba’s finance arm was able to achieve better non-performing loan ratios than traditional banks. Furthermore, banks and telecom companies are sharing data to drive new products and to improve core operations such as credit underwriting, customer segmentation, and risk and fraud management.“
- Retail banks have the opportunity to unlock significant value from the massive volumes of data that is currently fragmented across silos. McKinsey suggest combining this data into ‘data lakes’ from both structured and unstructured sources.
- Three types of retail banking players are likely to emerge: 1) Solution and analytics innovators who focus on a market niche such as Betterment for wealth management, 2) traditional data owners such as the banks themselves but only if they continue to own the relationship with their customers, 3) Non-finance companies that combine non-finance data to provide a more comprehensive and personalised view of the customer.
How should incumbent banks respond:
- “In an environment of constant churn, organisations should have one eye on high-risk, high-reward moves of their own. They have to keep thinking ahead, whether that means entering new markets or changing their business models.“
- “At the same time, they have to maintain a focus on using data and analytics to improve their core business. This may involve identifying specific use cases and applications on the revenue and the cost side, using analytics insights to streamline internal processes, and building mechanisms for constant learning and feedback to improve. Pursuing this two-part strategy should position any organisation to take advantage of opportunities and thwart potential disruptors.“
Banks Don’t Know Their Customers
In the developed world, many of us use Netflix and Amazon and many other services where we’re treated as individuals. Increasingly, we expect our banks to treat us the same way. The trouble is banks and credit unions are failing to employ data and analytics to meet our expectations. Oracle and Efma conducted a survey of senior executives from 60 financial institutions globally. They found that: 1) Other than account-level data the scope of data and insights is falling short, 2) More than half are not using real-time analytics at all, and the rest is patchy at best, 3) Only 6% are using social media data for customer insight and engagement, 4) Just 8% are using big data insights on a daily basis, 5) 79% are using a product-centric approach to pricing rather than a risk-based or engagement-based approach. In sum: To remain competitive, banks need to organise data and draw timely insights from it so as to provide a more compelling customer experience. See the above McKinsey report for more answers. DAMNING
‘First-Mover Advantage’ is a Myth
Steve Blank explains why so-called First-Mover Advantage is a myth. In essence: 1) It’s unlikely that you’ll understand your customers’ wants and needs as a first-mover, 2) As a first-mover you’re either going to end-up merely burning through your cash, or merely praying that the hype will let you flip you company, 3) “None of the market leaders in technology were the first movers.” EXCELLENT
What is your SaaS startup likely to sell for? According to a study by Tomasz Tunguz of software company acquisitions over the last 6 years, the median premium Enterprise Value / Revenue multiple right now is 6x to 7x. USEFUL
False Signs of Traction
Rick Zullo, VC at Lightbank, cautions against falling for these fake signs of traction: 1) Selling to Friends and/or Previous Customers, 2) Selling to Non-Core Customers, 3) Non-Repeatable CX, 4) Pilots and Renewal Cycles, 5) Phantom Customer Acquisition / Retention Economics. BEWARE
Series C is Different
Mattermark explain why Series C investment is different to Seed, A and B series. For a Series A check, you need a solid product. For Series B, you need to demonstrate real traction. For a C round, a solid business is required. Growth capital is for company-building. USEFUL
If you’re looking to build an important multi-generational company then going public probably makes sense. Advantages include: 1) M&A currency to help acquire other companies, 2) more stability for acquiring otherwise elusive customers, 3) more resilient capital if private capital becomes problematic, 4) Being IPO-ready can actually give founders more options and more control over their destiny. Nicole Irvin from Andreessen Horowitz walks us through a 16-point checklist for preparing to IPO (or a robust and enduring business ready to IPO when you are). INSIGHTFUL
Mobile Eats the World
Ben Evans explains in a great video presentation: “As we pass 2.5bn smartphones on earth and head towards 5bn, and mobile moves from creation to deployment, the questions change. What’s the state of the smartphone, machine learning and ‘GAFA’, and what can we build as we stand on the shoulders of giants?” TAKE HEED
Artificial Intelligence & Machine Learning Explained
Sam DeBrule has put together a handy list of non-technical articles and other resources to explain what Artificial Intelligence and Machine Learning is. EXCELLENT
Deep Learning <> Machine Learning
Carlos Perez is a software developer and is writing a book about Deep Learning. He’s fed up of people mixing up Machine Learning and Deep Learning, so he set out to set the record straight. MANDATORY
Please consider forwarding the BeyondFintech newsletter to a friend or colleague … thank you!