Resources and Latest News
Receive updates on automated decisioning directly to your inbox

Your Email

“Nothing endures but change”. – said Heraclitus. If credit policy fails to evolve it soon becomes a burden, cutting deep into lending profitability.

With the convenience of decision automation tools, risk managers can evaluate efficiency of any given credit policy, thus identifying the most efficient loan application processing work flow.

Applying Champion/Challenger or A/B Testing for Credit Risk Management

You can compare the effectiveness of alternative credit risk management strategies using champion/challenger approach .  KPIs of the different credit policies can be examined to gain insights into possible changes. For example, you can enhance decision logic, adjust an interest rate, or set a different score cut-off parameter. Once the challenger decision flow is specified, you can compare KPIs for alternative credit policies by running them in live or virtual environments.

Selecting Key Performance Indicators

You may select any KPI in line with your current business challenges. For instance, you can take the following operational or business characteristics as KPI:

Operational KPIs:

  • Delinquency ratio, LGD ratio;
  • Average loan application processing time, average time spent on a certain user workplace, average time of server data requests.

Business indicators:

  • Proportion of approved and turned down loan applications;
  • Key account characteristics.

Recommendation on Setting up Champion/Challenger Testing

In the beginning, we suggest processing the majority of loan applications using the current credit policy (the Champion). The rest of the loan applications (for instance, 10-15%) should be evaluated using the Challenger approach.

Champion/Challenger Approach for Lending

Click on the image to learn more about Champion/Challenger approach in our online learning center

Once you compare the KPIs, you can consider further improving the credit policy, or you can decide to employ the new credit policy.

Scorto’s tools allow you to set up and customize rules for automatic migration of the applications that are currently processed. Once the most optimal credit policy is identified, these applications automatically shift into the new processing mode. A wizard based interface will prompt you to specify necessary details of the migration, should this prove to be necessary.

For instance: you will be able to set certain parameters to be calculated automatically, re-calculated, or set by default.


In order to create an environment that supports great decision making and execution, institutions should embed analytics into every business decision, make informed decisions quickly and execute them smartly. Champion/challenger approach helps risk managers go beyond automating decisions according to corporate strategy, and to constantly maintain the high level of effectiveness of lending strategies and policy rules.

Author: Anna Zelenskaya | Google+

The online lending market is seeing impressive growth. The investment community is keen to meet forward thinking online lenders armed with an advanced technological platform to support borrower pre-screening, underwriting and debt collection.

But even if you’ve come up with most innovative and inspiring concept for an online financing solution, you still have to pay close attention to the following factors in order to effectively approach potential investors.

Identify Your Secret Weapon

Find your way to differentiate your online financing services against competition. For instance, you may employ a sophisticated and highly intellectual loan calculator, innovatively use third party data for borrower evaluation, present enticing features for premium users or provide best user experience through simple and intuitive loan application forms. Numerous startup success stories have proven that the best way to come up with a unique business idea is to start with a problem, preferably with problems that are experienced by yourself or people close to you.

Estimate Your Budget with Maximum Accuracy

Thoroughly calculate costs of marketing as well as tax and legislative expenses. Maximize your technological investments by selecting the most robust loan application processing platform and by using in-house scorecards.

Team Power

The team factor is one of the strongest success factors considered by the investors. According to the experience of founders, the essence of a successful startup is not an idea but the ability to bring it to reality.

Investors will evaluate to which extent people in your team are capable of bringing your idea to life. Make sure to accurately select your team, establish a corporate culture and develop an efficient working environment.

PR Plan

You might be convinced that Forbes, Techcrunch, Inc, GigaOM will be excited to spread a word about your business, but reality might prove different.

It is wiser to test and identify the best PR strategy ahead of launch. Set up a “launching soon” webpage, solicit interest and promote your project among niche bloggers. The launching soon website will further prove your determination to make your business a reality in the eyes of the investment community.

What Are Investors Expecting from Founders

Emerging technologies and new business models spur novel forms of online lending businesses. The largest UK peer-to-peer communities Zopa, RateSetter and Funding Circle have already lent GBP 218 million, GBP 29 million and GBP 41 million respectively.

As you have seen, it is much more easy to solicit investments for your project if you understand investors’ expectation towards founders. In his article “What We Look for in Founders” Paul Graham outlines the portrait of an ideal candidate for venture capital. The most important qualities in the eyes of investment community are determination, flexibility, imagination, naughtiness and strong relationship between the founders. Be genuine, rational and accurate, and success will come at your feet!

Author: Anna Zelenskaya | Google+

We are taking our first steps into 2013, and now it is the best time to re-think our business strategies.

This article considers front-burner issues related to lending, and shows leaders how they can adopt a different approach to new technologies and disciplines and fuel lending growth in the coming year.

1. Maximize the proportion of automated decisions.

Manual data input and application handling is prone to human subjectivity and mistakes, and is not optimal for standard decisions.

Exclude experts from the majority of decisions and focus them on making new and creative decisions. Apply decision automation where it works best and use a more manual approach to capture new opportunities. Standard operational decisions should be formalized and automated.

Automated operational decisions with formalized rules are available for constant analysis and improvement.

This way you can minimize the amount of skipped decisions thus minimizing missed opportunities and revenues.

Click to enlarge: Maximizing lending revenues with automated decisioning

2. Encourage reasonable risk-taking and prudent growth with custom scorecards.

Predictive models are rapidly becoming an essential part of customer acquisition and loan underwriting. With application scorecards, you can make underwriting decisions more quickly and accurately.

Jennifer Tescher, Center for Financial Services Innovation, stresses the importance of acquiring more customers among the middle class population in the Banking Predictions for 2013: “2013 is the year banks dust off their reputations by embracing the middle class. The mass market isn’t what it used to be. Bankers need to leverage technology to reinvent products, delivery channels and the overall experience to win back the hearts – and wallets – of the average consumer. By focusing on the financial well-being of their customers, banks might just improve their own financial outlook.”

3. Implement sophisticated decisioning logic using visual decision flow editing interface.

To capitalize on changes in lending environment organizations must be able to make complex process changes in a flawless and transparent way.  Our experience reveals that agile and efficient lending is impossible without a visual decision flow editing tool.

4. Harness untraditional data to maximize borrower assessment and pre-screening efficiency.

John Dineen, GE Healthcare shares in his recent Forbes interview: “Harnessing big data means gaining the ability to go out and collect data from multiple sources in order to make better decisions.”

About 88 million consumers  (35% of the adult population) in the United States have little to no credit history, which makes them unscoreable, says Steve Ely, the CEO of eCredable in Alpharetta, Ga.

To acquire best-bet customers, ensure that your loan origination software has seamless connection with alternative data sources while avoiding dependence on a single data source.

5. Ensure efficient collaboration of everyone involved in the decisioning process

In order to embed analytics into everyday operations you need a lending system that streamlines decision flow. But it is equally important to culturally orient and motivate your team toward using analytics and automated decisioning.

L. T. (Tom) Hall, CEO & Transformational Change Leader at Resurgent Performance, Inc. shares a lot of valuable recommendations on improving an operational culture: establish a cultural framework and corporate ideals by rewarding sound values and behaviors within the team.


Approximately 45 percent of the population have committed to New Year’s resolutions this year, says a recent study by the University of Scranton’s Journal of Clinical Psychology.

Aspects of life that we struggle to improve are very often nothing more than a consequence or certain behavioral patterns. It is wiser and more effective to deal with the behavioral patterns.

This principle is true for risk management as well. Instead of adding minor policy and process improvements, make a serious commitment for 2013 and introduce a new, forward thinking approach to lending.

Author: Anna Zelenskaya | Google+

December 21st, 2012

Happy Holidays from Scorto!

No Comments, Uncategorized, by Anna Zelenskaya.

Season's Greetings!

Scorto wishes you a safe and prosperous time.

May all your decisions be fruitful, and may your New Year be filled with peace, joy and harmony.

Enjoy the Holiday season!

Sun Tzu has said: Speed is the essence of war.

The ability to make wise decisions swiftly will place you well ahead of your competitors. With smart automation of the decision flow you will eclipse competitors.

Decisioning tools and processes range from excel spreadsheets and intuition to sophisticated enterprise solutions. But no one disputes anymore that speed and accuracy of today’s decisions determines tomorrow’s success.

To help you set up a robust decisioning framework we’ve identified the top three challenges to be addressed by decision automation solutions in the 2013.

1. Enhancing business decisions with data from non-traditional sources

“In markets and market sectors that are still developing, the challenge is to incorporate non-traditional data sources into traditional models, to still leverage the analytical power of decision automation systems without imposing the same level of data rigidity that is standard in the developed markets. “ – says Brendan le Grange, a credit risk professional with ten years of experience who explains the concepts of risk management in Credit Risk Strategy Blog.

We totally confirm the necessity of embedding alternative data sources into daily operations. We’d like to draw your particular attention to the importance of social media sentiment analysis. Sourcing and analyzing data from social media channels allows you to see the real-time picture of the operational environment while overcoming headaches of merging data from multiple influencers. For instance, you can use data on social media activity to uncover suspicious actions and detect fraudsters; to enhance customer pre-screening and cross-selling, and more.

It is very important to join insights gained through non-traditional data sources with your own analytical models. For example, you can boost customer retention efforts. Using metrics on social media authority of each client you can target your “Superstar” customers during your promotional campaigns and thus maximize the outcome of your promotions.

2.      Presenting the information  in the best way

“Balancing accuracy which frequently requires advanced, black-box methods, and transparency, which requires methods to be understandable and explainable” is one of the strongest challenges to be solved by decision automation systems, according to Gregory Piatetsky,  a founder of KDD (Knowledge Discovery and Data mining conferences) and the President of KDnuggets, which provides analytics and data mining consulting.

Decision automation systems need to instantly identify which data insights are expected by any given business role. This radically streamlines users’ access to large quantities of data, thus empowering businesses to resolve potential problems in real time or even before they arise.

3.      Acting on Big Data in Real Time

Decision automation systems must follow through with the results of their analysis and actually transform them into profitable decisions.

Today’s solutions learn to predict the best next action. Previously they have been operating on an if-then basis, and now artificial intelligence allows implementing a sophisticated self-learning mechanism.

An example of such advances is the inclusion of champion/challenger testing approach into a decisioning strategy. In this case, a decision automation application is supplied with different strategies, and selects the next best action based on how every strategy has performed in the past.

Magic Formula: Insights-Comprehension-Action

We’d like to encourage you to use decision automation applications in a way, which can be encapsulated into the Insights-Comprehension-Action formula:

  • Insights: use decision automation systems that are embracing advanced analytics on every stage of decisioning process, and that can enhance existing facts with data from non-traditional sources;
  • Comprehension: ensure that all the insights are presented in a transparent way, and that user interfaces are tailored according to their business roles;
  • Action: make sure that the system follows through the end of decisioning process and transforms the knowledge into actions. Take advantage of the self-learning capabilities of your system, instead of simply following rigid algorithms.

We hope that these recommendations will help you set up an efficient and robust decisioning process. Please comment and share your thoughts on top challenges to be addressed by decision automation systems in 2013.

May all your decisions be fruitful, and may your New Year be filled with peace, joy and harmony.

Enjoy the Holiday season!

Author: Anna Zelenskaya | Google+

This article examines the perfect combination of automated decisioning with advanced analytics and the human interaction.

Let’s start with a simple brainteaser. Take a look at this picture:

There is a set of four cards lying on a table. Your task is to verify the rule: “If there’s a vowel written on one side of the card, then an even number is on the other side.” Identify which card(s) you need to turn to check the validity of this rule.

Most of respondents give an answer straight away: it is enough to check the other side of the card “A”. Another popular answer is: you need to turn both “A” and “2” cards.

Sure, we need to flip the “A” card, since this card has a vowel on it, and we have no data on what’s on the other side of this card. Is it really necessary to turn the “2” card? Our rule says nothing about the even numbers – therefore, we have no interest in checking this card. But this doesn’t mean that checking the “A” card is sufficient. We must flip the “7” card as well to see if the other side of this card has a vowel written on it. If it did, that would refute the rule.

This task is called “Wason selection task” and was created by Peter Wason, a leading cognitive psychologist. According to his experiments, four out of five respondents fail to solve this puzzle correctly. Cognitive psychologists have found that people are wary of speculating on factors with high level of ambiguity; they prefer to base their decisions only on known facts. In other words, they tend to lose out of sight uncertain information.

But sound and insightful decision making is impossible without the information that is “hidden” in the data. Paul Rogers and Jenny Davis-Peccoud at Bain & Company have comprised a list of 10 Decision Diseases That Plague Companies, and this rating is topped by a lack of relevant insights or, as they call it, Blurred vision.

Below we’d like to provide just a few examples of how predictive analytics and decision automation paltforms can enhance decisioning in various areas.

Embedding scorecards and advanced analytical models into loan origination systems allows lenders to acquire most reliable and profitable accounts, render the optimal decisions for loan pricing, and capture cross-sell opportunities. This way, forward thinking lenders can reduce costs of customer acquisition campaigns, improve their loan portfolio and overall profitability.

Marketers can leverage decision automation platforms to design, test and implement customer lifecycle management activities and marketing campaigns. With sophisticated data analytics and behavioral scorecards, they can identify white spaces in the market and gain information advantage over their competitors. Furthermore, marketing automation solutions enable them to monitor performance of the product and services across different target groups, and keep their focus on changing customer demands.

Decision automation technologies streamline business flow and allow companies of all sizes to embed intelligence into their daily operations. Smart market players who can leverage predictive analytics to optimize their decisioning and risk management will step beyond accessing relevant insights: they will be able to automatically transform the information into profitable actions.

Author: Anna Zelenskaya | Google+

Loan origination systems with integrated scoring models help businesses improve underwriting decisions and provide faster customer service. In this post we will show, how lenders can introduce more effective underwriting software by improving accuracy of the predictive models, or scorecards.

Consider every stage of scorecard development

To prevent potential errors and ensure best performance of the scoring models, we need to consider all stages of scorecard development process.

Stages of scorecard development:

  1. Data sampling
  2. Processing, statistical evaluations of data characteristics
  3. Dividing data into training and validation datasets
  4. Scorecard development: performing regression procedure
  5. Scorecard validation

To ensure that the scorecard can help us evaluate risks and build a comprehensive customer profile, we must focus on the stages 1-3. It is sensible to focus on improving scorecard performance before its development is finished.

In order to build most robust loan application processing system we need to focus on improving on correct preparation of the applications data and on the processing of the loan portfolio data.

The most common scorecard development errors.

Errors during the data sampling stage

During data sampling, loan applications data is sourced from the data warehouse.

Pay close attention to maintaining main requirements for the working sample: randomness and representativeness.

The factor of randomness implies that all data from loan applications needs to be added into the working dataset independently.

To follow the rule of representativeness, you need to ensure maximum accuracy of the data characteristics in the dataset related to the actual characteristics of the potential borrowers in the loan applications portfolio. Taking into account the need of scorecard to provide insights on the population specifics, this requirement is totally sensible.

Performance and efficiency of the scoring model will be significantly impacted if requirements for representativeness and randomness are not met.

Working dataset can be randomly formed by sourcing data from the loan applications database according to a certain time frame.

This way, time frame selected for the dataset is working as a decisive factor.Let us illustrate this by showing mechanism of estimating according to the time factor:

Mechanism of forecasting based on the time factor

During Performance Period we are sourcing information on the loan performance of a certain borrower.

The Observation point is the representing the moment of the forecast. Loan performance is evaluated and determined at the Outcome point.

For instance: if the observation period lasts for 12 months and the time horizon equals to 6 months, then we can estimate the state of the borrower in 6 months by analyzing performance of tha loan case during 12 months.

Please keep in mind that the observation period cannot be interrupted, and the historical data used has to be close to the data in the loan portfolio.

Interrupting the observation period will have negative impact on the scorecard performance.

Data samples obtained during at least last three years is required for achieving best forecasting results.

To prevent data sampling errors, take twofold approach. First, you should take under direct control the procedures of sourcing applications’ data.   Secondly, make sure to evaluate statistically characteristics of the potential borrowers. 

Continue reading on improving scorecard accuracy in loan origination software…


With fierce competition among financial services providers it is getting increasingly difficult to win the battle for customer attention. This post studies latest marketing trends and shows how some of the leading players in the space are using marketing automation and demonstrating fresh and intelligent approach to raising brand awareness. Based on their examples, we will explain how you can leverage Big Data on your existing customers and scorecard development services for targeted marketing.

The most remarkable marketing campaigns in the financial industry

Invesco: “Accidental investing: it’s not always that obvious”

A recent advertising campaign of an independent investment management company Invesco is geared towards stressing the importance of prudent investing.

Invesco: Pothole (c) Advertising Agency: Leo Burnett, Chicago, USA

Invesco: Pothole (c) Advertising Agency: Leo Burnett, Chicago, USA

Barclays: “Bank without emotions”

The British bank Barclays has developed a series of print advertisements to illustrate its emotion-free approach:

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

Bank without emotions. Barclays (c) Advertising Agency: SAA, Moscow, Russia

First Bank: “For the banking insomniac”

First Bank of Colorado in Lakewood turned to innovative electroluminescent billboards that are glowing in the dark to emphasized its non-stop interaction with customers through the around-the-clock customer service.

For the banking insomniac. 1ST Bank (c) Advertising Agency: TDA_Boulder

For the banking insomniac. 1ST Bank (c) Advertising Agency: TDA_Boulder

For the banking insomniac. 1ST Bank (c) Advertising Agency: TDA_Boulder

For the banking insomniac. 1ST Bank (c) Advertising Agency: TDA_Boulder

Nordea: “Give your piggy bank a new life”

One of the major banks in Sweden, Nordea Bank, has a lot of ideas to help its customers find better ways to interact with their favorite piggy banks. Personal funds of Nordea’s customers are certainly in safe hands!

A new life for piggy bank. (c) Advertising Agency: DDB, Helsinki, Finland

A new life for piggy bank. (c) Advertising Agency: DDB, Helsinki, Finland

A new life for piggy bank. (c)  Advertising Agency: DDB, Helsinki, Finland

A new life for piggy bank. (c) Advertising Agency: DDB, Helsinki, Finland

Efficient marketing with Big Data

Accessibility of Big Data enables financial institutions to capitalize on their knowledge of the customer profile and behavior. Integrating these precious insights into marketing activities guarantees competitive differentiation and great response from the campaigns.

In his Forbes interview, James Taylor, a leading expert and independent consultant in Decision Management, has said:

“At the end of the day your customer relationships are driven by their reaction to the decisions you make about them. Developing systems to manage these decisions that are agile enough to change when that is necessary, that embed analytics to improve these decisions, and that are adaptive so they can improve over time is a critical need for better customer relationships. ”

You can read more on this subject in James’ book Decision Management Systems. A Practical Guide to Using Business Rules and Predictive Analytics.

By fully exploiting data and analytics on their customers, the forward thinking companies can answer such important questions:

What kind of customer should be attracted by a particular marketing campaign? What is their psychological and demographic portrait? Where and how does he/she spend their time? What kind of next steps suggested by the marketing activity will be most effective for this target group? What is the expected ROI of the marketing campaign?

Our clients who have implemented marketing automation and are supporting their marketing with advanced analytics, are able to accurately answer these questions and thus significantly increase their customer acquisition campaigns.

Author: Anna Zelenskaya | Google+

Loan origination systems are being reshaped by Big Data and predictive analytics, forcing banks and alternative lenders to compete on a new level. To outpace competition and maintain a leadership role, lenders should leverage decision automation systems in the most robust way.

Experience of our clients and our research of most strong challenges in decision automation systems reveals a list of  the seven necessary elements of a robust decision automation system for lending.

1. Visual editor for interacting with decision flow

Loan underwriting processes are growing in complexity. To successfully navigate the challenging journey of  building you first custom decision flow for loan origination process, you need a genuinely smart, simple and helpful visual editor.

2. Robust integration of multiple decisioning strategies

The world around is changing and so does the business flow and lending policy. Non-stop improvements in the decision flow has shifted from the frontiers of innovative lending to a mainstream necessity. A truly flexible and robust decision automation platform should provide you with the capability to manage multiple decisioning strategies and running them simultaneously,  and designing the optimal decision flow using champion/challenger approach.

3. Seamless integration with the third party data

Big data provides huge advantages to lenders if they can transform their knowledge into profitable customer acquisition decisions. When evaluating the capabilities of a decision automation system make sure it can not only source third party data, but also convert raw data into accurate, business-driven analytical models.

4. Enhancing decisioning with analytical and scoring models

In his recent interview on the future of data mining Gregory Piatetsky-Shapiro, editor and chief scientist of the leading data mining digital publication, KDnuggets, said: ” I expect that techniques and algorithms will become better adapted to real-time embedded analyses out of the box.”

Custom prediction models and scorecards enable informed, fact based decision making on every stage of the business flow. An efficient decision automation system must go beyond allowing you to upload scoring models in PMML format. Ideally, users should also be able to monitor performance of their statistical models, adjust, calibrate and update them as needed.

5. Framework for managing in-house scoring models

As banks step into a Basel III domain, supporting Internal Ratings Based systems becomes one of the top features of a loan origination system. Designing and improving internal scoring models allow executing more detailed risk assessments of potential borrowers and greatly enhance risk and capital management capabilities.

6. Balance between human intelligence and automated actions

Decision automated platform should optimize staff efficiency by presenting users with a data-driven understanding of the consequences of every possible action.

The perfect loan origination system will simplify data interpretation, extract insights and visualize data and thus deliver better decision-making to every person involved into the loan origination process.

7. A mechanism for flexible connection of those involved into the decision making process

Streamlining decision flow means maximizing collaboration efficiency across business units and between front office, analytical team and the executives.  As volume and variety of automated actions increases, it is important to maintain transparency in the way that users interact with each other and with the system.

In order to develop an agile, analytic and adaptive business, lenders should ensure rigor and balance in decision making. These seven features are building a simple and clear framework to assess different decision automation platforms. What metrics have you been using to evaluate a decision automation platforms?

The middle class in Africa has tripled between 1972 and now, reaching 34% of the content’s population, as found by African Development Bank in its report “The Middle of the Pyramid: Dynamics of the Middle Class in Africa”. In face of this trend, banks are experiencing significant growth. This inevitably brings up certain challenges that keep executives on their toes.

There are five most strongest risks perceived by the executives of the Eastern African banks: too much dependence on technology, fraud concerns, fierce competition from newly established financial institutions, credit risks and risk management framework and technology, – states PricewaterhouseCoopers research “Spotlight on financial services 2011 Risk Survey”.

This data shows that the risk of getting too much dependent on technology is now brought to the forefront of banks thinking. Potential shortage of specialists and technological resources as well as on proper credit risk management training is a significant burden for introducing innovations in decision automation technologies. But as the pace of new developments in technology is unrelenting, those banks who can ensure the proper resources and technology will easily eclipse their competitors.

Leveraging our experience in Sub-Saharan Africa and our research, we can confirm that these are issues lying over the horizon for the executives of financial institutions.

Through our Sub-Saharan office in Nairobi, we have sensed the competitive atmosphere between banks, as having encountered multiple requests from banks who strive to enhance their customer service and strengthen their margins. This tendency is certainly not going to decrease, forcing some of the banks even to relax their loan approval and underwriting procedures.

It is always best to address the causes of any given problem – rather than the symptoms. With this in mind, we would like to share a few recommendations that will help banks to mitigate and overcome their risks:

  • Put risk management on a new level by embedding analytics throughout the organization.
  • Gain information advantage over the competitors using advanced analytical models such as in-house scorecards.
  • Instead of focusing on branch network growth, win the hearts of the unbanked and under-banked population with superior mobile and online banking services.Take advantage of Big Data created by customer interactions to identify white spaces in the market and better tailor you offerings.