This is Part 2 of our series revisiting an early Q&A about what we expected the PositivityTech® platform and suite of applications would do. If you missed Part 1, you can read it here.

“Even if we didn’t achieve this, I feel like being able to do that and grind and go on a journey to try to achieve it would have been worth it as well, but this is definitely the cherry on top.”

— Jalen Brunson, New York Knicks

The world just watched the New York Knicks fight their way to an NBA championship. What made their achievement extraordinary was their determination to keep pushing through setbacks, pressure, and an agonizing 53-year drought. 

The win was the cherry on top, but the grind was what made it meaningful. In many ways, that mindset resonates deeply with what we’ve been building at PositivityTech.

In Part 1, we revisited the original idea behind PositivityTech: that customer conversations contain valuable signals about risk, behavior, and future performance. Today, that idea is no longer a hypothesis. We proved it. But proving business value and driving market adoption are two very different things.

Throughout my career, I’ve been an application builder. Whether it was creating and leading Decision Management at Citi or building PositivityTech, the challenge has always been the same: How do you take an untapped approach, transform it into a repeatable process, and create measurable business value?

That remains the goal behind PositivityTech. 

Most companies still approach customer conversation data through an operational lens, focusing on reducing costs, managing complaints, or replacing manual processes. The larger opportunity — using customer narratives to drive forward-looking business strategy and financial value — remains largely untapped.

From the beginning, we knew customer conversations could become a strategic data asset that enables innovative techniques and solutions for fraud detection, risk management, growth, and customer strategy. 

Over the last several years, we’ve refined PositivityTech as we learned what it takes to move to implementation — embedding customer conversation data into the processes, decisions, and systems organizations use every day.

As Angela and I discuss in the video below, that journey has helped define what PositivityTech is today:

Since launching PositivityTech, what have you learned about execution and implementation?

When we launched PositivityTech, we focused on proving that customer conversations contain hidden value.

What we’ve learned since then is that organizations need a way to operationalize and act upon that value. They need tools that fit into existing processes, support decision-making, and drive action. That realization has shaped how we’ve evolved the platform.

PositivityTech today is the combination of three interconnected capabilities:

  • A unique customer narrative dataset built from millions of customer conversations
  • A SaaS platform that visualizes insights derived from conversations and makes them actionable
  • Proprietary algorithms and models that transform raw narratives into predictive intelligence

Together, these capabilities help organizations identify and act on signals that are difficult or impossible to detect through traditional reporting.

What are the most powerful applications for PositivityTech today?

Over time, three applications have emerged as particularly powerful:

1. Fraud Detection: Separating Real Voices from Engineered Narratives

One of the biggest changes we’ve seen over the last several years is the rise of sophisticated fraud. Credit washing, AI-generated complaints, duplicate narratives, and engineered disputes are creating unprecedented challenges for financial institutions.

  • Credit washing activity has increased nearly 700% in two years.
  • An estimated $10 billion in debt has been removed through fraudulent credit repair schemes.
  • Nearly one-third of credit unions report scam activity increasing between 50% and 100% from 2024 to 2025.

At the same time, AI has made it easier than ever to generate convincing but deceptive narratives. This creates a fundamental challenge: How do you distinguish authentic customer friction from engineered complaints?

Early recognition of fraudulent transactions is a critical element in reducing credit washing. PositivityTech’s Bad Actor Detection™ (BAD) engine flags potential credit washing transactions by interrogating complaint data sent to the credit bureaus. BAD uses a proprietary artificial intelligence data management system and large language models to separate potential fraudulent complaints from authentic ones and determine which are most likely to create credit washing transactions. 

Our work has shown that four characteristics matter most: text duplication, word choice, timing, and complaint intensity.

These patterns become especially powerful when analyzed at scale. For example, no individual customer is submitting the same complaint 17,000 times. Yet, institutions may still attempt to process each one independently without realizing they are part of a coordinated fraud cohort. That creates enormous operational cost and inefficiency while obscuring genuine customer issues that deserve attention.

PositivityTech’s BAD engine flags individual credit reports suspicious of credit washing, enabling credit bureaus and lenders to take action accordingly. 
 

2. Risk Indicators: Seeing Economic Change Before It Happens

Customers often tell us what’s happening in their lives before those realities appear in traditional economic reports.

To capture those signals, PositivityTech developed proprietary Risk Indicators that connect customer narratives directly to business performance. These indicators — covering credit risk, revenue risk, balance sheet risk, and fraud risk — were built specifically from customer narrative data and are continuously refined through learning and analysis.

For example, our Credit Risk Indicator and Revenue Risk Indicator have signaled future increases in unemployment months before official data reflected those changes.

These indicators help organizations move from reacting to anticipating, giving leaders earlier visibility into emerging risks and opportunities.
 

3. Marketing Intelligence: Understanding Customers Before Competitors Do

Customer narratives reveal far more than complaints. They uncover changing needs, reasons for attrition, competitive opportunities, and emerging trends.

By combining customer conversations with benchmarking, segmentation, and proprietary algorithms such as Severity Score, Sentiment Score, and Multi-Dimensional Categorization, PositivityTech helps organizations identify opportunities for retention, growth, and improved customer experience.

Today, our data spans more than 9,000 institutions, enabling organizations to compare themselves against peers and regions, identify shifts in customer behavior, and uncover opportunities before competitors do.

The result is acting upon what customers are saying and what they are likely to do next.

Looking Ahead — The Cherry on Top

One of the most rewarding aspects of this journey has been seeing ideas that once existed only as hypotheses become proven capabilities. 

We’ve learned that customer conversations, when combined with technology, analytics, and business expertise, can help organizations identify fraud, manage risk, improve marketing performance, and create measurable financial value.

In Part 3, we’ll look at some of the milestones and achievements that have validated this work — and what comes next.

If you’d like to learn more about how PositivityTech can help your organization transform customer conversations into profitable business outcomes, connect with me at marcia.tal@positivitytech.com. I look forward to helping you transform negatives to positives.