Published on December 1, 2020 by Rajul Sood
Credit teams have a number of matters on their minds today – from supply-chain disruptions, changes in industry and country risk metrics, and volatility in capital markets to a potential recession. The COVID-19 pandemic has also certainly challenged banks’ operating machinery. Banks have had to support customers as they themselves struggle to survive. Government announcements of stimulus measures have created even more work, intensified by the resulting loan forgiveness. Additionally, activity has increased in areas such as portfolio maintenance, repricing emerging risks such as covenant waivers and managing payment defaults.
Artificial intelligence (AI) and machine learning (ML) have also come to the fore and significant investments have been made by these financial institutions in a host of new technologies. “Digitalisation” has clearly become a buzzword in the corporate and commercial banking world. But what’s the reality of how to achieve digitalisation in lending, to have the most beneficial impact?
It begins with data analytics
In the current environment, key strategic priorities are not limited to adopting a unified loan management system but extend to building analytics capabilities across the lending value chain. According to a leading research firm, 88% of financial institutions believe “improving the customer experience” is the most important digital banking transformation strategy of 2020 and 2021, while 77% of them believe it is “improving the use of data, analytics and AI”. Every bank has digital ambition, but implementation requires staff with domain knowledge. Therefore, most initiatives are dependent on front-office or credit-underwriting teams – who also have day-to-day business to complete.
How can banks use data in a truly impactful way? Here are a few examples:
For business customers – create marketing insight by utilising transactional data, industry data and data on purchasing habits that can help banks offer the right banking solutions to customers
For operating efficiencies – understand and optimise workflows and channel selection to increase speed to market, eliminate redundant processes and lower costs
For portfolio risk – create agile and robust portfolio monitoring capabilities that share information across platforms to quickly identify emerging risks, appropriately price products and improve credit decisions
AI and ML transforming the way banks assess their credits
The proliferation of data sets exists of course because we now live within an ecosystem of IT systems. Just a few years ago, ML was considered to be science fiction. Those companies that know how to analyse and interpret the data the world produces every second of every day, and can successfully integrate this into real-time decision making at the point of business, will become industry leaders. However, using model technologies requires that the underlying data be accurate; it is otherwise a case of “garbage in, garbage out.”
As sufficiently large and comprehensive datasets are required to train AI models, it is important to ensure data standardisation, accuracy and integration of these datasets across multiple lending platforms and systems. Banks should look at these tools as a means of augmenting the role of credit risk teams and bringing in efficiencies over time. With the information-gathering process mostly automated, credit teams would then be able to focus on more-value-added and judgmental analysis.
Banks overcoming the traditional paradox
There has historically been a paradox in using strategic partners in the credit analysis process due to loss of customer knowledge and sharing credit DNA. However, in recent years, banks have actively worked with strategic partners that can help them with their digitalisation programmes. They understand that leveraging cost-effective centralised teams will not only free up front-office time but also help realise their digital ambition.
Automation succeeds when each process is further divided into micro-processes and technology is applied to each of these micro-processes. For example, in the case of credit risk assessment, we apply automation to micro-processes such as financial spreading, covenant extraction and validating the factual portions of credit reviews. Once AI models and automation tools are trained on these processes, there are larger aggregate efficiencies.
However, there are barriers that pose a challenge to these initiatives: legacy systems; non-standard datasets; insufficient cooperation between businesses and the risk, IT, and operations functions; limited bandwidth in front-office or credit teams and, most importantly, no single owner of the credit process that could drive a change at scale.
….and transforming from a ‘legacy banks’ to a ‘digital banks’
Despite all the challenges, there are banks that have achieved success in their digitalisation journey. Here are a few examples of how they’ve done that:
Standardize credit processes: In order to ensure accuracy and regulate the way in which credits are analysed. Technology can fail if things are done differently across different teams and departments.
Leverage partnerships: A strategic partner can help you execute change and bring innovation, domain expertise and best practices, while keeping the lending engine running by providing additional bandwidth. Of course, choosing the right and cost-effective partner is critical to success.
Cultural change: It is important to effect a cultural shift towards adopting change and for different functions to accept the change. Start small through proof of concept, breaking credit processes into micro-processes and applying technology to each micro-process. While efficiencies would materialise over time, the benefits of standardisation, increased front-office bandwidth and improved user interface (UI) would be felt immediately.
Digitalisation is about breaking traditional barriers and achieving success in transformation initiatives. Success means much faster credit decisions; customers getting cash up to 70-80% sooner; lower costs, with centralised teams in cost-effective locations; 30-50% less time spent on credit decisions; and improved risk governance, all of which translate into greater profitability down the road. By improving processes and implementing technology, this is achievable for banks both large and small.
What's your view?
About the Author
Rajul heads the commercial lending practice at Acuity Knowledge Partners and has been with the firm for over 15 years. She is responsible for strategic planning, delivery oversight and management, quality assurance and supporting the innovation and technology initiatives in Lending. Rajul has extensive experience in investment banking analytics and commercial lending research services. Apart from banks, the teams she oversees also have in-depth experience in working across different lending products, processes and systems for Fortune 100 companies, SMEs and real-estate businesses. She holds a Master of Finance and Control, and a Bachelor of Commerce from Delhi University.
Munis expected to shine soon....
Overview of municipal bonds and types Municipal bonds (or munis) are debt instruments is....Read More
Metaverse – the financial services sector’s ....
The metaverse appears to be among the most exciting investing themes for the next decade. ....Read More
Why effective covenant monitoring is essential f....
In 2020, the global economy suffered from the effects of the pandemic such as supply-chain....Read More
AI and ML in banking – the next significant
Artificial intelligence (AI) and machine learning (ML) are shaping how banks work and perf....Read More
Impact of 5G on the fintech industry
Fifth-generation (5G) cellular technology is gaining popularity due to the advantages it o....Read More
Strong partnership with client teams: An impe
Service provider relationships are evolving multifold as businesses transcend boundaries a....Read More
Like the way we think?
Next time we post something new, we'll send it to your inbox