Open Site Navigation

Guest Column

From hindsight to foresight: Data - driven financing for SMEs

2021 නොවැම්බර් 19

From hindsight to foresight: Data - driven financing for SMEs
00:00 / 01:04

While SMEs faced chronic funding gaps pre-pandemic as well, difficulties in accessing finance have been greatly exacerbated by...

While SMEs faced chronic funding gaps pre-pandemic as well, difficulties in accessing finance have been greatly exacerbated by the economic pressures of COVID-19. With rising Non-Performing Loan (NPL) ratios, traditional financial institutions (FIs) are even less inclined to finance SMEs, but data-driven lenders are providing viable alternatives.

SMEs are vital to Sri Lanka’s economy, accounting for approximately 75% of all businesses and contributing to ~45% of total employment in the country. Despite the central role they play in driving growth, SMEs have long struggled to obtain capital from commercial lenders, primarily due to high financing costs and the lack of accepted collateral.

COVID-19 making access to finance even more difficult for SMEs

Pandemic-induced stresses have made it even more difficult for SMEs to stay afloat. An IFC study reported that in 2020, eight out of ten Sri Lankan SMEs found it difficult to meet their operating expenses and debt/other financial obligations due to the adverse effects of COVID-19. Despite the freezing of loan classifications to non-performing categories under the Government’s moratorium, the Sri Lankan banking sector’s gross NPL ratio had increased to 4.9% by end-2020 (just three years before, the ratio was 2.5%).

Global studies show that SMEs are being viewed as higher risk after the pandemic and that banks are seeking out safer borrowers. Even before COVID however, traditional FIs were largely risk averse and relied principally on historic data and tangible assets to assess creditworthiness. Technology-first financial players, on the other hand, have been circumventing some of the barriers to SME lending through the use of alternative data sources, predictive analytics, and machine learning―relying on foresight instead of hindsight.

Data analytics aiding the lending process

Technology giants like Amazon and Alibaba, as well as smaller fintechs like Tala and Fundbox, are applying data analytics to the lending process―reducing the time, costs, and formal paperwork needed to underwrite a credit application. They are completely redefining business models by not simply digitizing existing financial products but by using rich data and predictive analytics to customize products, to lend against intangible assets, and to seamlessly embed lending into other services.

These analytics-based lenders use varied alternative data sources to assess an individual’s or SME’s credit worthiness and to predict ability to honor loan payments. Some of the most creative data-driven lending models come not from mature financial markets but emerging ones, as the dearth of reliable information is often particularly pronounced in the latter.

Fintech Tala creating credit scores based on phone data

For micro loans, a good example is the fintech Tala, which started out in Kenya where most MSME founders could not provide evidence of a formal credit history. Instead, Tala uses data from the mobile phones that are ubiquitous in the country to create its own credit score and predictions―using indicators such as timeliness of paying bills and the consistency of relationships or applicant’s support system as demonstrated via call activity.

The advantages of technology-based underwriting is most evident in the speed of decision making and disbursement―a reported ~85% of Tala’s customers receive cash in their wallet within 2 minutes of completing their application. While the value of loans provided by Tala is very small, this can still be sufficient to allow MSMEs to smoothen cashflow in difficult circumstances. Overall, Tala’s model has been very successful thus far in Kenya – with a repayment rate of higher than 90% – and has been replicated in the Philippines, Mexico, and India.

MYbank underwriting loans based on SME activity on Alibaba & Alipay

Another success story in terms of more efficient commercial lending has been China’s first completely online bank―MYbank, which is affiliated with Ant Group and leverages AI and risk management technologies to offer collateral-free loans. To do this, MYbank relies heavily on data gathered via other Ant Group companies – primarily SME activity on Alibaba and Alipay – to feed its risk assessment algorithms.

MYbank pioneered the ‘3-1-0 model’ of SME financing―taking three minutes to apply on mobile, one second to approve, and zero human intervention. Here too, the data-enabled underwriting model has proved to be extremely reliable, with ~98% of SMEs funded by the 3-1-0 model repaying their loans on time, even through the pandemic.

With rising digital activity, similar models could become more viable in Sri Lanka

Both Tala and MYbank were operational before COVID-19 forced the world to consider extensive virtual options. With more small businesses now embracing digital channels, we can expect more real-time data to be generated and harnessed, reinforcing analytics-based lending models, and helping to drive further adoption globally.

In Sri Lanka, we know that the pandemic induced many SMEs to fast-track their adoption of online trading and digitization of processes. One of the country’s leading ecommerce platforms, Alibaba-backed Daraz, has reportedly seen thousands of SME sellers signing up each month since the pandemic broke out. Studies estimate a 5x growth over the next 5 years in Sri Lanka’s online retail market, and with this accelerated activity, there is great potential for richer data to be harnessed by the financial sector. For SMEs in Sri Lanka to flourish, we need data-driven lenders that can challenge incumbents and a regulatory framework that supports players with foresight.


“Gendered Impacts of COVID-19 on Small and Medium-Sized Enterprises in Sri Lanka,” IFC, 2020; Annual Report, Central Bank of Sri Lanka, 2020; “Redefining E-Commerce in Sri Lanka: Prospects Post-COVID-19,” Asia Securities, 2020; “Catalyzing Positive Change: business lending in the post-pandemic era,” Trade Ledger, 2020; “Daraz announces seller promises,” Daily News, April 2021; “Money, technology and banking: what lessons can China teach the rest of the world?” BIS, 2021; “Jack Ma’s Online Bank Is Leading a Quiet Revolution in Chinese Lending,” Bloomberg, July 2019; “MYbank unveils 5-year plan to reach more SMEs across China via supply chain finance and rural lending,” Ant Group press release, June 2020; Amazon Lending website; “Prashant Fuloria, CEO of Fundbox — Promoting the Post-Pandemic SMB Recovery,” Wharton Fintech, October 2021; “Tala – How data boosts access to credit for low-income individuals in emerging economies,” Harvard Business School, March 2021; “Tala grabs $145M to offer more financial services in emerging markets,” Tech Crunch, October 2021.

Related Articles

Learn More

Financial Insights

The art of Business Valuations

Thought piece

Tapping external funding sources

Our Partnerships

NDB Joins Hands with Women's Chamber of Industry and Commerce