Double Collection allows for automated processing of big data in search for additional
information about the Bank’s debtors on the Web.
Through mining of the open data sources, this product helps rebuild communication with previously
uncooperative debtors by the method
that is most convenient to them.
facilitates steady growth of debtor’s cooperativeness;
increases the number of promises to repay;
boosts the repayment of past due debts considerably
Ultimately, this tool significantly increases the collection of past due amounts and reduces the reserves for contingency loss, while dramatically reducing the staff required for the search for any additional information and interaction with the debtors.
This solution considerably improves the key performance indicators both for early (up to 90 days) and for the late stages of delinquency (over 360, 720 and 1,000 days).
The Company has implemented over 50 commercially successful projects with the major Banks, Collection Agencies and Microfinancial Institutions.
The software is designed to reduce the default rate, fraud, prime cost and man-hours with
respect to verification of the borrower’s identity.
guarantees the remote verification of the borrower’s identity;
automatically verifies the employment records, income and phone numbers of the borrower from the open sources;
collates more information about the borrower on the stage of the loan provision.
The solution requires integration on the following stages of approval of the requests for loan:
in front office CRM or online request form to cut off fraudulent requests on the early stages;
in automatic approval systems to automate verification of data provided in the application form and to increase the percentage of requests verified without expanding the verifier staff;
as a Web-interface used in standard manual underwriting and verification process to improve quality and sensitivity.
The solution helps to find: the borrower’s accounts in social networks; phone numbers; e-mail; places of employment and positions; graduate degrees; schools; salary/bonus assessment; personal pictures; marital status; children (if any; relatives, colleagues, classmates, fellow students; cities/regions of presence; phones of close third parties, etc.
Double Scoring helps obtain information about the borrowers from independent sources, take a
fresh look on the submitted applications and add to the Bank’s credit scoring system other
factors that are based on the outside, unbiased data.
This software product detects and gathers hundreds of additional scoring factors from social networks and other open sources on the Web for 60% of borrowers and the people in their circle. Both the obtained data and the Company solution itself may be used for improvement of predictive capacity of Application Scoring, Behavioral Scoring and Collection Scoring models.
With its extensive experience in testing and commercial operation of the product, the Company is able to advise its customers on integration of any specific factors in their individual models prior to commencement of the tests, as well as to continue the development and creation of the new strong factors, such as those based on the analysis of:
the borrower’s interests;
the close third parties.
Double Scoring has been designed on the basis of the advanced technologies to minimize the response time and to promote reliable operation under extreme load. Integrability of solution in the Bank’s loan pipeline is confirmed by its commercial operation by Top-10 Russian retail Banks.
We would specifically point out that this solution may be used for pre-scoring of new customers (the customers almost totally unknown to the Bank) to cut off the customers who are obviously fail the main scoring on the early stages or to assess the potential customers in the context of marketing budgets optimization.
Double Antifraud helps reduce the default rate for each specific portfolio and detect fraud within the Bank.
This software solution:
consolidates information about the borrower and its environment obtained from the open sources with information available at the Bank;
identifies patterns and characteristics of fraudulent nature that result in the losses;
may be integrated in the loan pipeline as one of the automatic decision support systems.
The company has a unique experience in identifying fraud patterns with the help of social networks and uses the Big Data analytics to restore social ties with or to frauds. On the basis of the patterns and the collected expertise, we develop the rules for each customer that are applied in the course of the loan provision and help prevent fraudulent practices.
This tool is being increasingly popular as the frauds frequently recruit “newbies” and keep in touch with each other online.
Our projects helped identify fraud clusters on the basis of the odd bits of information obtained from the open sources. We have identified bank officers who consciously participated in provision of notoriously unrecoverable loans, including the so-called “black brokers” and their accomplices. We have also developed procedures and a specific solution that would help prevent such fraudulent practices on the stage of loan provision.
We believe that open data sources and Big Data technologies will help tackle a number of urgent business tasks. That is why the Company invests generously in joint R&D projects involving its partners and customers. We are currently engaged in a number of projects aiming to enhance targeting, promote extra sales and develop text communication channels for financial institutions. We see a great potential for solution of the problems faced by cellular operators, insurance companies, travel operators, e-commerce companies, etc.
We like complicated Big Data tasks with virtually unpredictable outcome. There are times when, acting together with our customer, we obtain fantastic results and one has only to keep pressing on. Then again, there are times when no such results can be obtained and we respond by testing other hypotheses, tap into other data sources and develop our technologies.
We are grateful to our customers who helped us design these products. Thanks to the joint tests, they were the first to get the final products and solve their pressing problems.
We will be happy to make a new product together with you and help you solve your most pressing problem!