In this episode of Signal & Noise, hosts Brian Lamar and Andrew DeCilles speak with Ben Cortese, Vice President of Decision Sciences and Innovation at KS&R, to explore how custom AI chatbots are reshaping the way research insights are accessed, shared, and expanded beyond traditional deliverables.

Ben explains how his team is combining qualitative and quantitative research with secure AI systems to build proprietary chatbots that allow clients to interact naturally with their own data. These tools help stakeholders ask new questions, explore nuances, and identify early opportunities without re-fielding studies or relying solely on static reports.

The discussion covers how KS&R structures these models to remain accurate, safe, and aligned with proven research findings. Ben also walks through the technical evolution of the tool, including early failures, the role of system guardrails, the importance of loading structured data, and why transparency and validation matter more than speed or novelty.

Key Takeaways:

• What it means to extend insights beyond traditional deliverables

• How custom chatbots are powered by proprietary qualitative content

• Why guardrails are essential to prevent drift, hallucination, and off-topic responses

• The process of grounding AI output to match known research narratives

• How AI can support ideation, early concept testing, and faster client exploration

• Where AI can accelerate the workflow versus where human judgment remains irreplaceable

• How researchers may evolve into curators, validators, and strategic translators

• The future role of AI in tracking, longitudinal analysis, and blended data environments

If you loved the episode, have comments, or want to appear on the show, connect with us down below!

Connect with us:

• LinkedIn

• YouTube

• ROI Rocket

Connect with Ben Cortese, PhD:

• LinkedIn
  

In this episode of Signal & Noise, hosts Brian Lamar and Andrew DeCilles welcome the team from Verisoul, a company redefining fraud detection in market research. The guests include Henry LeGard, Founder and CEO, Joey Maddox, Chief Strategy Officer, and Erinn Taylor, who joins the conversation to explore how Verisoul uses banking-grade technology to detect and eliminate fraud while keeping real respondents in the data set.

Henry and Joey explain how Verisoul’s background in cybersecurity and fintech enabled them to develop invisible fraud-prevention tools that identify bots, VPN spoofers, and fraud rings in real time. The team discusses how their system detects human behavior through device patterns, network signals, and even physics-based methods like latency and impossible travel analysis, all without disrupting survey participants.

Brian and Andrew dive into how legacy fraud detection tools often over-block legitimate respondents, creating bias and frustration. The Verisoul team outlines how their approach balances protection with precision, reducing false positives and improving data integrity across the research industry. They also share insights from Verisoul’s large-scale fraud study that analyzed more than 50 million sessions, revealing global fraud trends and the role of device types in fraud rates.

Key Takeaways:

• How Verisoul applies banking-level fraud detection to market research

• The global scale of organized fraud farms and location spoofing

• The impact of false positives on respondent experience and data quality

• Privacy compliance across global markets

• Insights from Verisoul’s fraud research and data quality report

• The concept of real-time fraud and value scoring

• The growing importance of authenticated AI agents in fraud prevention

If you loved the episode, have comments, or want to appear on the show, connect with us down below!

Connect with us:

• LinkedIn

• YouTube

• ROI Rocket

Connect with Joey:

• LinkedIn

Connect with Henry:

• LinkedIn

Connect with Erinn:

• LinkedIn

Learn more about Verisoul:

• Website

In this no-filter episode of Signal & Noise, hosts Brian and Andrew dive into one of the most pressing issues in market research: the misuse of fraud detection platforms. They unpack how digital fingerprinting tools, while valuable, are often implemented in ways that damage data quality and block legitimate respondents.

Brian and Andrew explain how overly-aggressive fraud settings can create non-response bias, limit sample diversity, and frustrate real participants. They reveal why using multiple fraud detection tools or relying solely on automation can lead to distorted data and wasted resources.

They share lessons from industry research, stories of clients who turned off their fraud filters and saw better outcomes, and practical insights into balancing data protection with respondent experience. The conversation builds toward Brian’s “Eight Steps to Winning,” a clear framework to improve quality, restore trust, and move the industry forward.

Key Takeaways:

• Why misusing fraud detection tools can do more harm than good

• How false positives block real people and skew your data

• The hidden consequences of using multiple fraud platforms

• The Eight Steps to Winning framework for better data quality

If you loved the episode, have comments, or want to appear on the show, connect with us down below!

Connect with us:

• LinkedIn

• YouTube

• ROI Rocket