AI has become the core that maintains and enhances process performance – expert

Born in Ufa city (Bashkortostan, Russia), Oleg Ganeev has built his career in the banking industry (Uralsib, Rus-Bank, Sberbank). In November 2014, at 42, he was appointed as Deputy Chairman of the Executive Board of Sberbank. He is in charge, in particular, of the sales network, distressed assets department, Islamic banking and related partnerships. In this exchange with East-West Digital News, Mr. Ganeev presents a few of the latest fintech products developed by the financial giant and its corporate innovation strategy.

Sberbank offers a wide range of digital solutions for its B2C and B2B customers. Please tell us a bit about some of the latest ones, Bankro.TECH and DA (Distressed Аssets). is an online platform where bankruptcy trustees and lenders working with entities and individuals can interact with each other. We created this product in partnership with [legal online portal], seeking to streamline our own operations. 

As a matter of fact, we currently handle over 5K corporate bankruptcy cases and over 15K individual bankruptcy cases, accounting for 10% and 35% of all such cases in Russia, respectively.

In the traditional method, man-hours are gargantuan and those are mostly routine operations, so we’ve decided to automate them. But with the Bankro.TECH platform, stakeholders can aggregate data about the bankruptcy procedure across all possible public sources, have it structured and analysis-ready to send it to all users who need it. 

Based on the data collected, the system automatically generates registries of creditors, builds schedules for court hearings, and generates the documents a bankruptcy procedure requires. It also allows process participants to communicate. For example, to hold creditor committee meetings remotely.

What about Portal DA? 

It is an online venue featuring Sberbank’s pledged and non-core assets that investors can acquire if they seek to expand the boundaries of their current business and diversify it. Portal DA is not merely a showcase of assets with a unique classifier, the ability to search applying multiple parameters and an integrated recommendation system. It is also a convenient tool for online interaction between potential investors and employees responsible for the sale of assets. 

When launched in 2018, this service was originally designed to search for buyers for Sberbank’s pledged and non-core assets. Later, however, having developed more service features and grown a client pool of core investors who are ready to acquire large assets for business, we realized there was potential for further development. We could attract other owners and mortgage lenders to place the assets they sell, so we have embarked on the path of transforming the portal into a marketplace. It can be used by any company seeking to acquire or sell business assets – from shopping malls and office centers to factories of various industry sectors and agricultural complexes – or get related services such as apply for a loan to buy a troubled asset, or apply for counseling services in a bankruptcy procedure. Portal DA already features assets of dozens other commercial banks and organizations, all of which are our partners.

Did you get inspired by existing foreign technologies, or are these products very specific? 

The business processes we work with and our market in general have their own peculiarities. On the one hand, every country has its debt collection issues and problems. On the other hand, the requirements and regulations are different.

Naturally, we keep an eye on the trends on our market and the adjacent ones, while also figuring out whether different solutions could be used in our operations. But one cannot say that we follow the market. We work hard on our own solutions and sometimes we even find ourselves one step ahead of others as trendsetters.

But at the same time, we do not dismiss the best global practices and embed them as separate technological elements. For example, Portal DA utilizes the cutting-edge Elasticsearch search engine.

Take, for instance, voice assistants. They’ve been out there for a long time as a market trend. But we have gone further, designing a unique Russian-voice robot with a debt handling functionality. We were the first to market this solution two years ago and today our robot processes about 90% of the initial interactions with customers on overdue debts. Moreover, the technology is not static, it is evolving: it would be more correct to call it a “consultant”, rather than a “debt collector”. The robot engages in a meaningful dialogue with a client, clarifies the reasons behind the delay, fixes the date for the promised payment, and advises on a range of products. Step by step we are expanding the list of segments this technology can be applied in – not just individual clients, but small and micro-enterprises, too.

You use AI and machine learning to deal with distressed debts and the role of robotic technology in debt collecting. Please explain

Technologies, AI, and machine learning are not just supporting resources that help businesses. They have become the core that maintains and enhances business performance.

Technologies and models are being used on each collection stage. A client might not let expire a loan repayment yet, while predictive models can already foresee such a possibility. The system can make an appointment with a client to prevent a loan repayment expiration, remind a client about an upcoming payment, send a telephone number of a bank employee to discuss any issues if there are some.

If the loan repayment has expired, models come to the rescue as well – they classify clients across a large number of parameters and actually create individual patterns to recover debt. With some customers, a call or a text is enough, with others you need to visit them in person or offer them an optimal settlement tool. 

Technology is being used for settlement purposes, too. It has enabled us to cut the decision-making time on deals. Some years ago, clients would wait for a debt restructuring decision for a few days, while today 98% of these decisions are made in one day.

Have Bankro.Tech and DA been developed in-house, or have they come to Sberbank as a result of an acquisition?

Bankro.Tech is a joint product created from scratch by Sberbank and our partner, and it has no alternatives in Russia. Portal DA is Sberbank Group’s proprietary solution; the service was built from scratch using modern technology. 

Generally speaking, when it comes to tech innovation, how strongly does Sberbank rely on in-house development vs. cooperation with tech providers vs. startup acquisition?

Taking into account the seriousness and specific nature of our business, we intend to build up our own development expertise.

Does Sberbank cooperate in these fields with foreign players? Please provide some examples of such cooperation and specify in which countries you have such partners. Also, let us know whether Sberbank intends to develop such international partnerships, in which fields and which countries as a priority.

Among other things, Portal DA features assets in foreign countries that are being sold as part of the settlement routine for the troubled debt of Sberbank Group’s foreign clients. Now we are also in talks with several European banks and corporations that are set to place their foreign assets on Portal DA to look for buyers.

Previous ArticleNext Article