How it Works

Statistics SA collects data regarding various aspects of the country and is the official data
provider for the South Africa government.
Our Solution allows organizations to use data for various provinces and metropolitan areas to
enrich training datasets while also incorporating the history of these metrics.
Information includes:

• Prices of products in the CPI basket
• Residential property prices
• Population data by age and gender
• Electricity generation and distribution
• Municipality Building Figures

Key Deliverables

• Availability of data sources.
• Create relationships based on region, time and additional features.
• Provides signal to increase accuracy of Machine Learning models.

About the Partner

Knowledge Integration Dynamics (KID) is a cutting-edge data management consultancy firm
that excels in providing innovative solutions to complex data challenges. Founded on the
principles of expertise, integrity, and innovation, KID offers a comprehensive suite of services
tailored to meet the diverse needs of its clients. With a team of seasoned professionals who are
experts in their respective fields, KID leverages the latest technologies and best practices to
help organizations unlock the full potential of their data assets. From data integration and
governance to advanced analytics and machine learning, KID empowers businesses to make
informed decisions and drive growth in today’s data-driven world.

Solution Diagram

Success Story

• A propensity model was built with our collated data on past customer behavior patterns and certain geographical and demographic attributes. Sales efforts could then be optimized to increase conversion rates.
• Embracing external data sources, we identified ideal locations for new brick-and-mortar stores using optimization. This allowed greater customer reach and facilitated effective construction planning.

How It Works

In this solution, we leverage computer vision to identify product defects in hot-rolled steel plates, widely used in construction and agriculture. Using an object detection model powered by machine learning, we achieve precise and efficient detection and classification of prevalent defects like scratches.

Traditionally, the visual inspection of steel plates is time-consuming and potentially unreliable. Our approach automates the detection process, enhancing accuracy and reducing human effort and error.

This solution is split into two phases, with model training contained in Phase 1 and model deployment with DataRobot detailed in Phase 2.

Phase 1 uses a dataset containing images of steel plates with different defects to fine-tune a pre-trained Faster R-CNN model, an incredibly powerful object detection machine learning architecture, to classify and locate scratches. 

Phase 2 uses the DataRobot platform to host a non-DataRobot, pre-trained object detection model to leverage a host of ML Production features such as model monitoring with custom metrics. The specific code applied here assumes the input/output schema of the resulting FasterR-CNN model from Phase 1 but any object detection model can work with the proper data handling updates.

The output: a highly accurate and robust predictive model capable of detecting and classifying any sized scratch present in steel plates, with a DataRobot deployment monitoring model health and logging metrics. 

Key Deliverables

• AI-based defect detection solution
• Demonstration of data acquirement and preprocessing 
• Custom training and validation loops for fine-tuning the model
• Significantly automated defect detection process
• Model observability in production with custom metrics to evaluate performance
• Practical application insights to other manufacturing industries

About the Partner

KUNGFU.AI is a management consulting and engineering firm focused exclusively on artificial intelligence. With a deep understanding of key contributors to AI outcomes, including data, designers, users, and leaders, KUNGFU.AI tailors strategies that propel businesses forward. 

The team, comprising experts in AI, machine learning, and software engineering, stays at the forefront of research to deliver cutting-edge solutions. By truly understanding client challenges and environments, KUNGFU.AI iteratively crafts solutions, optimizing AI operations at every stage. Their commitment extends beyond implementation to repeatability, reliability, and alignment with corporate objectives, empowering organizations to redirect resources toward innovation and sustainable advantage. 

Whether clients are initiating a data strategy or deploying production AI models, KUNGFU.AI ensures maximum return on investment in an AI-driven future.

Solution Architecture

Success Story

KUNGFU.AI designed and developed a computer vision detection and pricing system capable of detecting dents of various sizes and magnitudes to calculate price estimates for repairs.

The client, an automotive services company, sought innovative ways to automate service delivery and expand offerings through tech-enabled solutions. Our objective was to create a capability enabling customers to send images for automated damage inspection and cost estimates for common car repairs.

This capability provides the client with a new source of revenue while reducing delivery overhead. It has initiated channel distribution alliances with third-party auto-traders and attributed to valuation expansion for acquisition by Repairify.

How It Works

Over the course of 8-10 weeks, we will develop a governance framework, guide the design thinking process, and a roadmap for generative AI capabilities. Once viable use cases are identified, we will leverage the DataRobot AI platform to build, test, and fine-tune generative AI solutions tailored to each client, in a dedicated sandbox environment.

Key Deliverables

• Generative AI Strategy and Roadmap
• Governance Plan: Covers model management, ethics, regulatory risk evaluation, and test/evaluation criteria. 
• Custom Generative AI model

About the Partner

KUNGFU.AI is a management consulting and engineering firm focused exclusively on artificial intelligence. We empower CEOs and senior executives to leverage the full potential of AI so they remain competitive in a rapidly evolving world. 

Our expert team delivers AI strategy and bespoke production-grade solutions that allow clients to rapidly realize value. We stand apart because we implement our AI strategies into production quickly, safely, and responsibly.

Solution Architecture

Success Story

Challenge

Since November of 2022 with the unveiling of OpenAI’s ChatGPT, firms across the globe have begun racing to deploy such capabilities. Understanding fact vs science fiction in such an environment is very difficult. Iodine Software, seeing opportunities for LLM adoption across the organization, sought KUNGFU.AI’s help in evaluating a roadmap of use cases and guiding them with responsible implementation.

Solution

Iodine Software came to the table with several different use cases across the organization. KUNGFU.AI Strategy and Machine Learning personnel engaged to unpack each use case, understand key goals and risks, and evaluate up to 7 different LLMs and their efficacy with the given use case.

Outcome

Over the course of several weeks, KUNGFU.AI was able to identify and prioritize the use cases Iodine was considering, evaluating the technical requirements and ROI for each case. Using KUNGFU.AI’s LLM medical scorecard, the team was able to stack-rank available LLMs based on fine-tune-ability, use on HIPAA data, and more. Lastly, the KUNGFU team helped Iodine roadmap out the adoption of LLMs across the various use cases.

Challenges

Ongoing Economic Uncertainty: Navigating the ever-evolving regulatory landscape while maintaining model performance in changing market conditions is a constant challenge. The persistent economic uncertainty challenges lenders in accurately assessing the credit quality of borrowers and sectors vulnerable to economic changes.

Complexity and Interpretability: Our primary challenge is balancing the advanced algorithms of ML models with their interpretability. We need models that are not only powerful but also transparent and understandable, both for effective management and clear communication with stakeholders.

Fairness and Bias: Another critical issue is ensuring fairness and avoiding bias in ML models. These models can inadvertently reflect biases present in the data, potentially discriminating based on sensitive attributes. Our task involves continually scrutinising and adjusting models to prevent such biases.
Extensive Documentation: MRM teams are often required to develop comprehensive documentation on the development methodologies of credit risk models. These documents are lengthy and cover the full spectrum of experimentation and testing results. This is even more important for third-party model developers.

How it Works

Intelligent Quality Assessment:

Our cutting-edge credit scorecards goes beyond logistic regression models and uses algorithms which captures complex relationships in your data without compromising on the interpretability. Our solution uses sensitivity tests to assess risk, enabling you to set customised cutoff thresholds for auto approvals, reviews, and auto declines. We leverage the latest LLMs to perform contextual feature engineering to capture the underlying behavioural signals near real time.

Advanced Model Testing:

We provide detailed analyses of individual exposures and credit portfolios against potential future economic changes, including stressed conditions. Our services include regular, granular stress testing tailored to the nature of your credit portfolio, incorporating severe but plausible scenarios. We leverage latest mlops concepts like champion challenger to compare model versions to identify output trajectory and prediction drifts.

Seamless Integration:

We understand the importance of seamless integration within your existing origination systems. Our credit scorecard solution seamlessly integrates with common platforms such as Snowflake, Redshift, RDBMS, and third-party APIs like Equifax and Experian. We offer a wide range of deployment options from rule engines to real time APIs. These integrations ensure your customer facing teams have all the necessary information to make the decision. 

Continuous Monitoring:

Failure to identify and measure the ongoing deterioration in credit risk in real time may lead to higher future losses and capital reserve inadequacy. We constantly monitor the impact on borrowers from changing marketing conditions like rising interest rates, high inflation and market volatility. Constant monitoring and refinement ensure that our scoring systems remain accurate and aligned with your evolving risk appetite.

Customisation and Personalisation:

We recognise that each lending institution has its own unique requirements and risk tolerances. Scorecards can take several forms, from predicting Probability of Default to discrete numbers. We provide customisable scaling factors to convert the odds to logical score ranges for non-experts.Our AI-agent driven solution offers a competitive edge by providing declined loan applicants with clear explanations and tailored advice, encouraging future reapplications. These agents are fully configurable and ensure human oversight minimising any operation risk. 

Auto-compliance documentation:

Our solution leverages a combination of DataRobots Compliance Document and proprietary LLMs to develop documentation reports that provides the MRM team with appropriate transparency into the model development process related to design, theory , assumptions and logic.

Key Deliverables

• Quantifiable improvements in predictive accuracy and risk assessment.
• Customisable risk-scoring dashboards for continuous monitoring.
• API integration for real-time risk assessment in lending processes.
• Scalability and flexibility of the solution to accommodate various lending products and regulations.
• Automated Compliance and Governance documentation reports powered by Generative AI

About the Partner

Evolve AI is a machine learning research lab dedicated to addressing the challenges of resource concentration in the field. Inspired by collaborative scientific ventures, our mission is to provide technical services including advanced data analytics, bespoke machine learning solutions, and comprehensive AI infrastructure to organizations of various sizes. Our core focus is on solving real-world problems and delivering tangible value through machine learning, with a commitment to responsible and ethical development.

Solution Diagram

Credit decisioning scorecards by Evolve AI Labs, offer a state-of-the-art solution to address lending challenges. These scorecards utilise advanced machine learning technology for precise borrower credit assessments and minimal default risk.

Our commitment to regulatory compliance ensures high standards, with rigorous stability testing under diverse economic conditions. This comprehensive solution streamlines lending processes, enhances risk management, and adapts to evolving risk profiles within a robust governance framework.

Whether you choose implementation or consulting services, we provide comprehensive support. We ensure your AI/ML models meet compliance and performance standards, offering detailed reports and analyses to guide your team through testing phases. With Evolve AI Labs, you can trust in expert credit decisioning processes fit for the dynamic financial landscape.

Success Story

In the world of personal finance in Australia, a prominent lender faced challenges as the market evolved and borrower creditworthiness was impacted by COVID-19. They turned to Evolve AI Labs for assistance in revamping their credit scoring system to account for these changes and enhance their loan portfolio performance.

Evolve AI Labs, using DataRobot’s capabilities, developed advanced precision risk scorecard models tailored to various loan products where each model underwent rigorous testing and regulatory approval. Additionally, Evolve AI Labs were involved in developing the expected credit loss (ECL) framework to comply with accounting standards such as IFRS9 and AASB 9.  

Loan approval processes sped up thanks to more accurate risk assessments from the new internal scorecard. Borrowers benefited from personalized risk-based pricing, allowing for lower interest rates as their credit improved during the loan term. Additionally, Evolve AI Labs implemented a hybrid strategy that reduced bad rates for one segment of customers without compromising approval rates, while another segment maintained similar bad rates but with a 20% increase in approval rates.

The value delivered by Evolve AI Labs extended beyond utility, encompassing enhanced operational efficiency and the identification of previously overlooked market segments. This transformation led to a more robust and prosperous loan portfolio.

How it Works

The Process Mining Solution EVACO 4ward Mining predicts and visualises the next following and the final status of an operation in real time while the process is running. For this purpose, DataRobot ML extracts all information of logically linked process steps and the data traces of previous process runs.

All that is needed are 3 basic informations from upstream process steps: process ID, assigned
timestamp and assigned status information.

• Process-ID: unique identifier of each new instance (depending on the process, it can be, for example, an order#, ticket#, customer number#, serial number#, etc.).
• Status information: List of activities that a process ID can go through
• Time stamp (date/time)

Based on this knowledge EVACO 4ward Mining determines, with DataRobot, and visualises the next step and the final outcome of a process in EVACO’s own QLIK® extension.

AI-driven 4ward Mining is suited to help customers from all industries make data-based business decisions on-the-fly by initiating measures in the ongoing process. This helps improving all logically linked activities, both internal processes and the interaction with external partners (suppliers, customers, prospects, IoT machine interaction etc.).

Use cases are: Purchase-to-Pay/ Procure-to-Pay (P2P); Order-to-Cash (O2C); HR Application
processes, Lead-to-Cash; Churn & Retention Rate, IoT, etc. EVACO 4ward Mining enables adaptive business process management. Ex-post reacting is replaced by a process-immanent, proactive optimisation. Potential problems can be recognised at an early stage and countermeasures can be taken.

Key Deliverables

• ML Models in DataRobot
• All process-relevant data available in Qlik® Extension
• Visualisation in EVACO´s own Qlik® Extension* (* = optional)
• Consulting- and Implementation expertise for DataRobot and Qlik®

About the Partner

EVACO is a leading provider of user-driven business analytics solutions in D-A-CH. The Duisburg- based company has specialized in the implementation of modern analytical systems and in technologies for Intelligent Search and Artificial Intelligence. The company’s goal is to turn complex data into knowledge – knowledge that provides customers with the basis for making easier, faster and better decisions. Customers benefit from over 20 years of expertise in a wide range of industries and exclusive partnerships with selected international vendors.

Solution Diagram

Success Story

Mid-sized energy supplier ROTH ENERGIE uses EVACO Process Mining to gain insights from data and optimize processes. The central system provides a structured overview of key information from ten service areas. Customer IT project management emphasizes Qlik’s user-friendly interface and the EVACO expertise. Introducing process mining has helped ROTH ENERGIE transforming from re-active data analysis to pro-active business process management.The applications cover sales, maintenance contracts, service stations, and various business units, using 44 Qlik Sense applications. EVACO Process Mining extension helps to review and optimize business processes.

How it Works

By combining predictive and generative AI, Evolutio creates AI-powered applications with chat interfaces allowing both end users and IT support teams to streamline support ticket creation, categorization, prioritization, and response.

First, IT support tickets are clustered and labeled with categories—issues related to operating systems, VPN, network, hardware, etc. Additionally, machine learning models predict ticket priority, ensuring each new ticket is both clearly categorized and prioritized.

Next, historical IT support responses and documentations are used to create a vector database with a text embedding model. Relevant information from these proprietary documents can now be retrieved based on customer query and used as additional context for the LLM to generate responses both to end users and the IT support team.

Key Deliverables

• End-to-end generative + predictive AI solution
• Frontend applications to provide Generative AI response for both end users and IT support team, with user feedback captured
• Predictive AI model for IT ticket clustering, categorization, and prioritization
• Retrieval augmentation generative workflow for high quality response, with customer’s choice of embedding model, vector database, and LLM
• Documents data and tabular data preparation for both generative and predictive AI
• LLM observability in production with custom metrics to monitor LLM performance
• Implementation of LLM guardrails to block unwanted prompt and LLM response

About the Partner

Evolutio specializes in helping organizations address the operational challenges of building and scaling complex enterprise applications. The professional service team deploys and optimizes proven technologies to increase revenue, enhance brand loyalty, and provide a premium digital experience.

With extensive industry experience in custom development, data engineering, analytics, machine learning, and MLOps, Evolutio offers proven enterprise project management experience, as well as technical certifications with DataRobot, AWS, GCP, Azure, Snowflake, SAP, Cisco.

Evolutio Data Science services are tailored to accelerate clients’ AI maturity, from first machine learning use case to the implementation of an enterprise AI Center of Excellence. Their expert team supports the full AI lifecycle, including enterprise ML experimentation, deployment and custom application development. Client success stories include Propensity to Buy, Supply Chain Time Series Forecasting, Visual AI, GenAI Solutions, NLP, Employee and Client Retention, and Edge Compute Industrial ML Applications, etc.

Solution Diagram

Success Story

Evolutio has developed the support ticket response automation solution for an international manufacturer. The solution generates rapid and high quality automatic responses for their IT support team and the end customer user. With the solution, the customer is able to reduce labor and time to respond to repetitive IT support tickets. With saved time, the IT support team is able to focus their bandwidth on high priority and complex customer requests. The business result is improved customer approval rating for faster and higher quality IT support, with improved support team work satisfaction.

How it Works

Customer retention relates to a business’s strategies and activities to keep its current customers. Customer retention analytics can facilitate such activities by providing predictive metrics of customers who are expected to churn.

Data-driven customer retention strategies can be very rewarding, often driving profit. Research demonstrates that organisations that make extensive use of customer data analytics for business decision making can see profit improvements of over 100%, versus those that don’t (source: McKinsey & Company).

You can significantly reduce churn and improve customer retention through utilising your data in the following ways:

1. Formulate and implement a data strategy

To implement a successful data strategy, you should make use of the data you have, understand what the data is telling you and implement organisational change accordingly. Analytics should be viewed as a strategic driver of growth rather than using it in a silo.

Effective steps that you can take include:

• Ensuring that your organisational KPIs are automated, scalable and repeatable
• As a company, collectively define the priority problems that you strive to solve
• Group the problems by ‘data’ and ‘system’ issues, being mindful that issues don’t
  tend to be with data but with how people use or manage it
• Put tasks in priority order, at the same time as evaluating if your plan is technically
  feasible
• Review your progress on a three-monthly basis to keep on track
• Ensure behaviour change across your organisation
  

2. Identify and segment those who are less likely to churn

Customers who are similar to your target customers are less likely to churn. Use the data you have about the features and characteristics of your current customers and create a list of prospective customers, applying algorithms that compare the two. Similar characteristics include:

• Job title
• Industry
• Company size
• Annual spend

Segment your customers based on this information and this will enable you to identify the features of the prospective customers, or high-quality leads, that you should target.

3. Utilise machine learning methods to formulate predictive models

To create a robust customer retention strategy, predictive analytics is a valuable tool that you can use to make predictions about the future. Using historical data, relationships among varied metrics can be analysed to predict what customers like and dislike.

In terms of customer retention, machine learning will quickly and accurately expose the underlying reasons why customers are churning; predictive analytics for churn. It can also identify why customers are loyal to your brand. This is achieved through the analysis of data, statistics and probability to find correlations between variables to assist in optimising crucial outcomes such as customer retention. Such models are applied to new customer data to make informed predictions.

In comparison to human analysis, machine learning algorithms can deliver insights rapidly owing to their processing capabilities. Owing to their iterative nature, the more data they consume, the better they perform. The following examples demonstrate how predictive analytics can be a hugely valuable tool in the retail, financial services and manufacturing industries:

Retail: a retailer can use predictive analytics to identify which up-sell or cross-sell products will be the most relevant based on a customer’s past purchase or browsing history. They can utilise it to establish the optimal attainable prices for each customer and identify the right pricing to increase sales. With real-time machine learning, the impact of competitive pricing on sales, and the adequate frequency of price-based promotions can be found.

Financial services: an insurance company, for example, would benefit from using customer retention analytics to secure targeted insurance plans, speed up claims processing, and offer personalised customer experiences. This all creates a competitive advantage that will attract new customers and retain existing ones.

Manufacturing: those in the manufacturing industry would reap the benefits of predictive analytics in order to analyse the history of demands, providing valuable insight into consumer buying habits, the availability of raw materials, impacts of a trade war, supplier issues, shipping barriers, and other potential disruptions.

4. Utilise segmentation to increase customer retention

Use data analytics to segment your customers and prospects into different groups, identifying how each segment engages with your brand and your products and services. You will be able to draw insights from each subgroup so that you can tailor your communications and strategies accordingly in order to optimise retention of your most valued customers. Awareness of customer value will enable you to make important business decisions.

Customer data that is important to analyse includes:

• Demographics
• Lifestyle
• Products and services purchased by category and customer type
• Frequency of purchase
• Purchase value

This will enable you to establish what types of customers are generating the most revenue.

Effective customer segmentation will enable you to create highly targeted product and
service recommendation offers. You can segment by, for example:

• Historical value
• Lifetime value
• Value over the next 12 months
• Average customer value by segme

Utilising the correct segmentation means that you can create highly targeted product and service recommendations for your customers and prospects.

A retailer, for example could offer its customers discounts based on different channels, such as online, mobile app or in-store, with different customers receiving different offers based upon their purchase value. Monitoring the seasonality and time-sensitivity of their promotional codes will enable an understanding of how a demographic responds to their sales communications and take relevant actions accordingly to maximise potential revenue

Key Deliverables

Retaining your existing customers will enable you to see the following benefits:

1. Retaining your current customers costs less

Attracting a new customer can be five times more expensive than retaining an existing customer. Loyal customers therefore are valuable assets that you should seek to keep. Finding out what makes your current customers stay and why they keep buying from you means that you can take the right steps in keeping the right customers.

2. The opportunity to up or cross sell to current customers

Current customers are much easier to market and sell to. Selling a new product or service to
your existing customer base is far less costly than selling them to new customers; customer
acquisition activities can be costly.

3. Increasing customer retention rates can enable sustainable growth

Increasing customer retention rates by just 5% can increase profits from anywhere between
25 – 95% (source: Harvard Business Review). Sustainable business growth is much more
likely to thrive by retaining your existing customers.

About the Partner

Since 2014, Catalyst BI has been at the forefront of revolutionising the way businesses make decisions. As trusted business intelligence consultants, we help teams and organisations in both the public and private sectors throughout the UK harness the power of data to drive smart and strategic decisions. With a proven track record of success, we’ve helped over 450 satisfied customers achieve their business goals and accelerate their transformation objectives. Whether you’re looking to make sense of complex data or gain valuable insights into your operations, Catalyst BI has the expertise and solutions you need to take your business to new heights of success.

Solution Diagram

Success Story

Company

Global Pharmaceutical Company

Situation

The company experienced rapid growth but faced a setback with a double-digit loss in annual revenue due to customer churn. This churn involved replacing 3,500 customer-product combinations to meet sales targets. Recognising the potential of AI to predict and mitigate this churn, but with no internal skillset or tools, the company sought an external solution to empower its sales team.

Partner Selection

The company chose to collaborate with DataRobot due to their strong partnership with Catalyst and based on Gartner’s AutoML recommendations.

Requirements

1. Explainability: Clear understanding of the factors contributing to customer churn predictions.
2. User Interface: Integration with other relevant data, like sales history, so that sales reps could access comprehensive information in one place.
3. Feedback Loop: The system needed to allow sales reps to add notes, change statuses, and communicate strategies to management.

Results

• DataRobot serves as the predictive engine which integrates with user-friendly BI interfaces. The integration was seamless, allowing for daily data updates between the systems.
• A 20% churn reduction was observed in the trial area, equivalent to a £2.3 million revenue increase.
• Multiple use cases and high-value projects are now in progress.
• The company has dedicated resources for AI and data science, maintaining a consistent development approach but remaining flexible in deployment strategies.

Additional Outcomes

• A growing network of Data Science professionals for further development and collaboration.

How it Works

We provided access to the Ready Signal UI and API for integration with DataRobot to augment your time series projects. An example workflow is laid out in the AI Accelerator. Ready Signal’s AutoDiscovery feature automatically compares your target variable to the 500+ control features within Ready Signal. It identifies the features that are most useful for predicting the target variable. Machine learning model can be then developed and deployed with augmented dataset in DataRobot.

The workflow can be applied to any time series forecasting project. Data Science consulting and implementation services are available through Ready Signal. 

Key Deliverables

• All data sources available in Ready Signal
• Feature aggregation and signal discovery in Ready Signal
• Time Series models in DataRobot
• Expertise from both DataRobot and Ready Signal to provide consulting and implementation services

About the Partner

Ready Signal is an AI-powered data platform that provides access to over 500 normalized, aggregated, and automatically updated data sources for predictive modeling, experimentation, business intelligence, and other data enrichment needs. Ready Signal provides a comprehensive control data catalog with an AI-powered recommendation engine to help DataRobot users find highly relevant leading indicators from external sources and seamlessly integrate them into your ML Experimentation & Models. The data catalog includes micro/macro-economic indicators, labor statistics, demographics, weather, and more. Its AI recommendation engine and auto feature engineering capabilities make it easy to integrate with existing data pipelines and analytics tooling, accelerating and enhancing how relevant third-party data is leveraged. Ready Signal saves data science organizations time from having to manually procure, test, and engineer control data that has a material impact on model performance and predictions. 

Solution Diagram

Success Story

Outsell

At Outsell, we are focused on helping car dealerships drive sales and service revenue by improving customer relationships through our AI based customer lifecycle marketing platform. Ready Signal’s control data improves the accuracy of our models and I never have to worry about the data being up to date. Ready Signal keeps everything current for us. -Matt Kristo, Sr. Manager, Analytics Services

How it Works

Our Propensity to Buy solution leverages targeted marketing data from financial institutions, including credit unions and credit reporting agencies, to increase your marketing return on investment, by identifying consumers most likely to purchase financial products and services.

With this solution, institutions such as banks or credit unions can stack-rank their consumers with a propensity to buy score for each product in their portfolio based on their data profile and past behaviors. Machine learning models pick up patterns from the data and accurately predict future consumer behavior, superseding traditional analytics for population segmentation and filtering.

Evolutio’s data science team develops and deploys propensity to buy models for each product line identified. Leveraging Snowflake into the production workflow enables seamless scheduling and monitoring daily batch predictions through DataRobot, with the prediction results written into Snowflake automatically. For institutions that use other database technologies, Evolutio provides services for different database integrations into DataRobot, for more information see supported databases.

The predicted propensity to buy scores are also delivered to the financial institution’s customer relationship management (CRM) system, with prediction explanations that enable sales teams to prioritize offerings and deliver relevant suggestions to their customers.

Evolutio’s team has worked extensively to maximize value for financial institutions by building an industry suite of models including customer lifetime value, know-your-customer (KYC), and client retention models. These models enable banks, credit unions, and credit agencies to attract, engage, and retain their customers. Additionally, our deep experience partnering with credit agencies amplifies our ability to derive value from data assets.

Key Deliverables

• Complimentary AI Strategy Session to validate data availability and business value of the Use Case
• Expertise in data engineering to extract and combine data from diverse client data sources
• Seamless migration, data pipeline creation, and production deployment
• Optimization of database technology and DataRobot integration capabilities
• Clear analytics insights and actionable guidance on targeted marketing, client satisfaction, and retention
• Results delivered seamlessly into existing technology platforms

About the Partner

Evolutio specializes in helping organizations address the operational challenges of building and scaling complex enterprise applications. The professional service team deploys and optimizes proven technologies to increase revenue, enhance brand loyalty, and provide a premium digital experience.

With extensive industry experience in custom development, data engineering, analytics, machine learning, and MLOps, Evolutio offers proven enterprise project management experience, as well as technical certifications with DataRobot, AWS, GCP, Azure, Snowflake, SAP, Cisco.

Evolutio Data Science services are tailored to accelerate clients’ AI maturity, from first machine learning use case to the implementation of an enterprise AI Center of Excellence. Their expert team supports the full AI lifecycle, including enterprise ML experimentation, deployment and custom application development. Client success stories include Propensity to Buy, Supply Chain Time Series Forecasting, Visual AI, GenAI Solutions, NLP, Employee and Client Retention, and Edge Compute Industrial ML Applications, etc.

Solution Diagram

Success Story

How it Works

EPIC is the most common EHR (electronic health record) system used in the United States. Many hospital systems and providers use EPIC as their main data store and to host machine learning models as well. Models deployed in EPIC can be consumed by end-users, such as physicians and nurses. Model predictions can be used to create dashboards or written into patient records.

Leveraging your data from EPIC to develop customized machine learning model with DataRobot is crucial to generate predictions and deliver value to the business. DataRobot models are tailored to your actual data with insights and transparency, instead of an off-the-shelf black box model trained on generic data. DataRobot also enables you to run many different type of modeling approaches in one go, as well as host and monitor model in production.

AHEAD can help you to develop end-to-end machine learning solutions with data in EPIC Charity or Caboodle, together with other data sources using Fivetran, in either Cloud or on-prem setup. The data integration pipeline also can be established with EPIC FHIR API endpoint.

After model experimentation, the best model can be deployed into EPIC using Nebula and monitored with DataRobot monitoring agent. Alternatively, the models can be deployed in DataRobot directly, with the benefit of native model monitoring capability in production.

The solution is HIPAA compliant with security and governance.

Key Deliverables

• Data integration for EPIC Clarity and Caboodle with other data sources using Fivetran
• JDBC connector or other cloud data integration services
• Data ingestion into Snowflake or another database
• Data engineering and transformation pipeline developed with dbt Labs
• Develop machine learning workflow and models in DataRobot
• Model deployment and monitoring in EPIC or DataRobot
• Secure and HIPPA compliant end-to-end data pipeline and machine learning workflow

About the Partner

AHEAD excels at building digital platforms, modernizing infrastructure with cloud-native capabilities in software engineering, data engineering and data science. AHEAD helps customers to accelerate their digital transformation through digital strategy consulting and professional services using modern tech stack and platforms. 

AHEAD provides services around cloud transformation and migration, agile coaching and DevOps enablement along with software engineering, data engineering and data science with modern data and AI architecture. For the data and analytics services portfolio, AHEAD delivers data strategy roadmap aligned to business strategy, data migration from legacy database to cloud, data pipelines and actionable insights integrated with end-to-end machine learning solution using DataRobot.

Healthcare is one of the most focused vertical for AHEAD, with around 40% clients in the industry. AHEAD has worked with over 150+ Epic customers for their infrastructure deployment projects. Here are several AHEAD’s referenceable healthcare provider customers: Lurie Children’s Hospital, Northwestern Memorial Hospital, Mayo Clinic, Community Health Systems, Beth Israel Deaconess Medical Center, Beaumont Health and HCA Healthcare.

Solution Diagram

Are you ready to visualize the power of our solution? In the ‘Solution Diagram’ section, we present an intuitive visual representation of how our cutting-edge technology addresses your challenges. Our innovative approach, combined with your valuable content, will create a compelling and comprehensive solution diagram. Let’s collaborate to showcase the seamless integration of our platform and empower our users. Fill this section with your expertise and insights to bring our solution to life!

Success Story

AHEAD FHIR Health API experience

AHEAD helped the client to build a single customer access portal for all their patients to access. The client has acquired various healthcare companies over the last few years creating a disparate system that made it very inconvenient for patients to access their healthcare information in a central place. AHEAD built a gateway that allows a single point of entry for communications to the various EPIC and other health systems sources via FHIRs, thus allowing the patients a single point of access. The unified data in the backend can be then used for machine learning projects across all the systems.