- Ham radio is still a disaster lifeline, even in the iPhone era - here's why
- You can now talk to ChatGPT on the phone - no Wi-Fi needed
- Accelerating AI for financial services: Innovation at scale with NVIDIA and Microsoft
- You can finally buy LG's transparent OLED TV - if you're willing to pay $60,000
- Docker Desktop 4.37: AI Catalog and Command-Line Efficiency | Docker
Banks Are Running Out of Time
By David Andrzejek, head of financial services, DataStax
There’s an important and growing trend occurring in financial services business processes and practices that’s upending how banks build applications and manage data. Bank product leaders are demanding instant, responsive, and personalized services, and bank technology leaders need to quickly execute a “real-time data” strategy.
Why? Because everywhere you look, time is being wrung out of financial processes.
For example, equity and other investment trades used to be processed and settled in three days (T+3 processing, in security trading parlance), but in September 2017 settlement was condensed to two days (T+2). The industry is currently working on T+1 settlement.
Credit card transactions seem fast at the terminal, but they’re actually much slower than they appear. Card transactions are only “authorized” in seconds; the actual payment settlement happens a day or two (or sometimes four!) later. Now, instant payments like Single Euro Payments Area (SEPA) in Europe; the RTP Network from The Clearing House in the United States and the Federal Reserve’s FedNow service, are fully settled with funds available in seconds.
If fully irrevocable payments are settled in seconds, it follows that fraud detection and anti-money laundering checks will need to happen in sub-second time.
As competition in financial services intensifies, banks are shifting from product-centric to customer-focused businesses. This translates into personalizing the bank’s digital experience to the corporate treasurer or consumer in real time, and supplying up-to-the-second information to support financial decisions in the moment.
For example, consumers want to know if they have exceeded their budget for the week or month before they place an order, and they need an immediate credit decision at checkout to buy now and pay later. Also, fintechs and some banks can now open accounts in a minute or less, which means “know your customer” (KYC) processes need to occur in seconds.
Instant payments, instant settlement, instant fraud checks and credit decisions, in-the-moment KYC—there is no time for time in bank processes any more.
The legacy ball and chain
Contrast today’s market reality with the legacy of bank data processes, and you’ll understand the massive challenge bank technology teams face. Bank data processing began in the era of the classic “banker’s hours,” when all banking was done in branches from nine o’clock to four o’clock. Scarce and expensive compute resources were dedicated to branch operations during open hours.
Overnight, while the bank was closed, transactions were resolved and settled into the ledger and reports were generated to serve the business. Thus the historical bedrock of bank processes was built on batch data processing and ETL. This “legacy heritage” is the proverbial ball and chain tied to the legs of bank engineering staff as they race to compete with fintechs, tech giants, and upstart banks.
Richer, smarter, real-time digital services
Today the bank is an app on your phone and is expected to be always available. As processing windows continue to shrink, financial services product leaders are under pressure to deliver customer journeys that can be completed from start to finish online and fully self-service.
This has product leaders thinking about all the data both the bank and the bank’s customers require in order to complete a task (say, to open an account, approve a transaction, or make a spending decision in the moment). Bank product leaders are thinking about how to incorporate data sources from outside the bank into their services so that decisions can be made instantly.
Product leaders designing these immediate services are causing bank technology teams to re-engineer their infrastructure for “instant” processing and decision making. These technology teams are modernizing bank infrastructure in part by building a “real-time data layer” over core transactional systems. The real-time data layer combines the abilities to capture, move, transform, and make decisions on data in sub-second time, and can accommodate both the increased volume and velocity of data, while ensuring bank services are always available.
Take Macquarie Group. The Australian financial services company understood that a key to differentiation in this brave new world is speed, so they prioritized building a data architecture, based on Apache Cassandra® that could take advantage of real-time data and build real-time engagement with their customers.
This real-time data strategy underpins the bank’s overall modernization and digital strategy, enabling bank technology teams to deliver richer, smarter digital services by capturing and making more data, including third-party data, readily available to developers via modern APIs. Developers are building data-driven services for these “instant” applications, and powering algorithms that deliver instant decisions from fraud detection, credit approvals, payment processing, and more.
Competition and heightened customer expectations of instant, always-on service are driving bank leaders to think beyond just “big data” and execute real-time data strategies. Given the pace of change in the industry, banks need real-time data capabilities, right now.
Learn more about us here.
About David Andrzejek
David Andrzejek, head of financial services, DataStax
DataStax
David Andrzejek has spent 25 years helping companies adopt technology to achieve outsized business transformation results.
