Operational Mortgage Analytics

And how to use it to build better mortgage operations
Vova Pylypchatin
Founder and CTO @ MortgageFlow

Operational mortgage analytics enables mortgage companies to operate more efficiently, reduce costs, manage risks, and improve customer satisfaction.

However, the term operational analytics is relatively uncommon, and what it represents may not be clear.

So in this issue, I’ll share my analysis of:

  • What’s is operational mortgage analytics
  • How is it different from traditional mortgage analytics
  • How does operational mortgage analytics work
  • How it can be used in mortgage operations

I hope it will clarify what Operational mortgage analytics is and if it can benefit your mortgage operations.

What’s operational mortgage analytics

Operational mortgage analytics refers to applying data analytics specifically to the operational aspects of mortgage lending.

It focuses on the efficiency of the operational processes involved in mortgage lending. This includes loan processing, underwriting, servicing, and default management.

The primary goal is to optimize operational performance, reduce costs, and enhance the customer experience through data-driven insights.

It involves analyzing data related to workflow efficiency, process bottlenecks, and staff productivity.

How it relates to operational mortgage automation, apps and data

Operational Mortgage Analytics is deeply intertwined with Operational Mortgage Automation, Apps, and Data.

It relies on the shared foundation, Operational Mortgage Data consumed by all 3 Analytics, Automation, and Apps.

Unlike Automation and Apps, analytics uses data only to answer questions based on data. Automation is used to act on the data, and apps are used to create/read/update/delete data.

How operational mortgage analytics is different from traditional analytics

Traditional analytics have been adopted by the mortgage industry for a while. So, it naturally raises the question of how operational analytics is different. Below is my analysis of the key differences between operational and traditional mortgage analytics.

Focus and application

Operational Mortgage Analytics focuses explicitly on the operational aspects of mortgage lending. It deals with analyzing and improving processes like loan origination, underwriting, servicing, and default management.

Traditional Mortgage Analytics focused on broader aspects like market trends, borrower behavior, loan performance, risk assessment, and portfolio analysis. The emphasis is on understanding the market dynamics, assessing credit risk, and making strategic decisions based on long-term trends and borrower demographics.

Time horizon

Operational Mortgage Analytics relies on real-time or near-real-time data to assist in day-to-day decision-making and make immediate improvements in operations.

Traditional Mortgage Analytics typically uses historical data to identify trends and make predictions. The analysis might be more retrospective, focusing on long-term patterns and strategic insights rather than immediate operational improvements.

Objectives and outcomes

Operational Mortgage Analytics is primarily used to streamline operations, reduce costs, improve customer experience, and enhance compliance within daily operational processes. It's more about tactical improvements in how mortgage operations are conducted.

Traditional Mortgage Analytics focuses on strategic decision-making, risk management, and market positioning. It's more about guiding long-term business strategies and understanding broader market risks and opportunities.

Scope and level of detail

Operational Mortgage Analytics focuses on granular, process-specific data. This includes detailed information on loan processing times, underwriting decisions, application turnaround times, error rates in document processing, compliance checks, customer interaction logs, and staff productivity metrics.

Traditional Mortgage Analytics utilizes broader market and customer data. This includes borrower credit profiles, loan-to-value ratios, historical default rates, macroeconomic indicators, housing market trends, and borrower demographics. While it can also be detailed, it often looks at longer-term trends and patterns, requiring a more aggregated view of data over time.

How operational mortgage analytics works

Like traditional, operational analytics consists of 2 parts:

  • Data → Raw information
  • Views → Visual representation of the data

How each part works is what makes it different from traditional analytics.

Below is an overview of the key concepts of operational mortgage analytics.

(Data) Events

An event is defined as something that occurred at a particular point in time.

Usually, it represents an action performed on/by an object or person.

In operational mortgage analytics, an event is usually an action within a specific process.

Here are some examples of the events:

  • Application taken
  • Loan estimate sent
  • Loan taken into processing
  • Appraisal received
  • Rate locked
  • Rate lock expired
  • Financing contingency expired
  • Deal closed
  • etc.

Here’s an example of the event structure:

  • Id → what’s a unique identifier of the event
  • Type → what event happened
  • Date → when the event happened
  • Loan → what loan was subject to the event
  • Properties → contextual information about the event

Events are continuously streamed to the permanent storage to enable real-time reporting.

(Data) Entities

When an event occurs, it is generally related to a particular entity.

Entities represent a person or an object subject to the event.

Here are some examples of the entities in mortgage operations:

  • Loan
  • Loan Officer
  • Borrower

Entities store reach and contextual information that enables more granular data analysis.

(Views) Queries

Query refers to extracting the data (events and entities) from the database.

Queries pull specific data needed to answer questions.

In operational analytics, most of the answers come from queuing events. While entities add more context to the queried events.

You can apply different methods (example below) to narrow your query to get as granular as you need.

Filtering

You can filter events by:

  • Event type
  • Event date
  • Event properties
  • Entity properties

For example, you can filter events by event type “Taken into processing” to get a list of loans taken into processing.

Aggregation

You can aggregate events

  • Count of events
  • Time between events
  • Conversion rate between events
  • etc.

For example, aggregation total count of the “Loan taken into processing” will give you total number of the loans taken into processing by all loan processors.

Grouping

You can use grouping to see the aggregation for each property value.

For example, grouping the total count of “taken into processing” events by Loan Processor will give you the total count of loans taken into processing by each Loan processor.

Time series aggregation

You can aggregate data across different time periods. For example you can aggregate results of the query above by month to see how each Loan Processor taken loans into processing each month.

(Views) Visualisation

Visualisation refers to the visual representation of the data in a chart.

Queries define WHAT data to present. While Visualisations determine HOW to present data.

Queries make it easier to comprehend the results of the query.

For example, you can quickly identify outliers by visualizing the query results above as a line chart.

There are plenty of ways you can visualize the data; here are some common examples:

  • Line/bar charts
  • Timeline
  • Table
  • Funnel
  • Number

How to use operational analytics to improve mortgage operations

The primary function of operational mortgage analytics is answering questions based on real-time operational data.

You can achieve different goals or solve various problems depending on your questions.

Below are a few common use cases for operational analytics in mortgage operations.

Process optimization and automation

By examining operational data, you can pinpoint where inefficiencies or errors are occurring within teams.

For example, suppose you’re looking to improve loan processing turnaround time. In that case, analytics can help identify steps in the loan processing chain that consistently experience delays.

The insights gained from operational mortgage analytics are crucial for identifying mortgage operation processes ripe for automation.

After implementing change into the process, analytics can be used to assess the effectiveness of these changes. This includes measuring improvements in efficiency, accuracy, and processing times.

Team performance monitoring & accountability

You can use operational analytics to quantify team and individual performance through measurable metrics.

Managers can objectively assess performance by analyzing data like the number of loans processed, underwriting accuracy, or number of customer interactions.

Clear metrics and performance data make it easier to hold team members accountable for their work. When everyone understands the benchmarks and expectations, it fosters a sense of responsibility.

These insights can be used to implement targeted training programs, process improvements, or redistribute workloads to enhance team efficiency.

Due date tracking

You can use operational analytics to automate the due date tracking process. By using real-time loan events as triggers, you can automatically calculate and capture due dates as events in the future.

For example, you can automatically create a “Financing contingency expired” event based on the “Offer Accepted” event and financing contingency days agreed on.

You can use future events to create reports that give a quick overview of where each loan stands in the pipeline and if any action is required.

Here, you can find an in-depth article on how to track mortgage due dates using Operational Analytics and Automation.

Company performance monitoring & forecasting

Managers and executives can understand the overall company standing by using operational data to create reports on KPIs.

Thanks to the event-based nature of operational analytics, you can retrieve current performance and how it has changed over time.

Performance over time can give insights needed to identify trends and build forecasts.

How to implement operational mortgage analytics and what tools to use

Here’s a high-level process for implementing operational analytics:

  1. Understand what questions you want to answer
  2. Define what charts/visualisations will answer the questions
  3. Define what data you’ll need to create the charts/visualisations
  4. Collect the data you need to answer these questions
  5. Setup reports that will give you answers to the data

Here’s the key tech used in the operational analytics:

  • Database for the entities (e.g., MongoDB, PostgreSQL, etc.)
  • Database for the events (e.g., ClickHouse, Swnoflake, etc.)
  • Event streaming technology (e.g., Kafka)
  • Event processing technology (e.g., Decodable)
  • BI/Reporting software (e.g., Streamlit, Mode, etc.)

An in-depth article on building operational mortgage analytics is coming soon.

What’s next

I hope this post gave you insight into Operational Mortgage Analytics and how it can be used in mortgage operations.

If you’d like to stay on top of the latest mortgage tech and how it can be applied to mortgage operations, consider joining our mortgage technology newsletter.

MORTGAGE TECH NEWSLETTER

Discover how technology can assist your mortgage company in reaching its strategic goals

A weekly newsletter about leveraging data, custom software, and modern technology to drive efficiency in mortgage operations.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Written by
Vova Pylypchatin
Founder and CTO @ MortgageFlow

I’m a software consultant with background in software engineering. Currently, I run a mortgage software consulting and development company that builds custom tools and automation solutions for mortgage lenders.