Revolutionizing Market Research with GASP: How NewtonX is Transforming Participant Targeting and Fraud Prevention

By Kavita Anand, Vice President of Product & Design

When I first joined NewtonX, I was immediately struck by the team’s dedication to maintaining a high bar for data quality for our research clients. As I quickly learned, the market research industry is fraught with bad quality data, especially for B2B surveys. NewtonX had built its reputation on providing high-quality survey data, borne from its precise approach to recruiting research participants. But, as clients increasingly demand more specialized expertise, recruiting participants is getting more time-consuming and resource-intensive, straining project margins.

Clients increasingly want more narrowly curated groups of participants with specific combinations of professional experiences, tool usage, and functional expertise for their research projects. This trend toward hyper-specialization presents a significant challenge: identifying and engaging these niche professionals while maintaining quality at scale. Companies need insights from CTOs who’ve implemented specific security solutions or marketing leaders who’ve used particular ad platforms in specific stages of growth, and they need large enough samples of these specialists for statistical analysis.

Doing such specialized research at scale encounters three key challenges:

1. The targeting challenge. Traditional approaches cast wide nets, inviting participants who are not well-matched to the research criteria. This leads to a higher rate of participants screening out, creating poor engagement, wasted resources, and ultimately, lower-quality insights.
2. The fraud and misrepresentation challenge. When participants aren’t well-matched to research opportunities, some feel incentivized to misrepresent their backgrounds or expertise to qualify for studies, compromising data integrity at its source.
3. The attribute visibility challenge. Businesses often need to segment research participants by specialized attributes that aren’t readily available in standard databases. Want to talk exclusively with IT Directors who recently migrated from AWS to Google Cloud? Good luck finding that information on LinkedIn profiles.

These challenges directly impact the reliability of insights that drive multi-million-dollar business decisions. As a product leader, I recognized that solving these three problems required more than incremental improvements to existing approaches.

Enter GASP: Granular Attribute Search and Prediction

This realization was the genesis of our Granular Attribute Search and Prediction (GASP) initiative. The concept emerged from a simple yet powerful insight: years of large-scale research operations at NewtonX had created a strategic moat of proprietary data that no competitor could replicate. This was uniquely valuable data about research participants, their attributes, expertise, behaviors, and engagement patterns.

The lightbulb moment came when we connected this unique data asset to the industry’s core challenges. What if we could leverage machine learning to transform this raw data into predictive intelligence about participant attributes and behaviors? What if we could use these predictions to dramatically improve targeting precision, reduce fraud risk, and increase visibility into otherwise unavailable attributes?

Bringing GASP to Life

Developing GASP required deep collaboration across product, business, and engineering teams, each adding a critical ingredient for success. 

Once we set the product vision, our Client Delivery (CD) team provided the domain expertise required to ensure our solution addressed real-world needs. The CD team is responsible for recruiting project participants and possesses a nuanced understanding of recruitment criteria by research type and frequently requested attributes for each segment. This helped my team prioritize attributes for GASP based on client and business needs.

Meanwhile, our product and engineering teams tackled laying the right ETL (extract, transform, load) pipeline: the data processing framework that pulls information from various sources, standardizes it, and makes it usable for our models. It was paramount to focus on a scalable architecture that could support not just a handful of attribute models, but potentially dozens as we expanded the initiative. 

We took a two-part approach to GASP, requiring designing complementary systems – one to improve targeting precision with proprietary data and another to predictively infer hidden attributes that clients increasingly demand.

GASP Part 1: Improving Recruiting Precision for a Better Participant Experience

GASP delivers tangible impact across multiple business dimensions by more precisely matching research participants to relevant opportunities. This significantly improves engagement rates by sending participants more relevant opportunities and more chances to contribute in their genuine areas of expertise—creating a better experience for everyone involved.

By proactively filtering out participants who would be poor fits for specific research criteria, GASP reduces the incidences of fraud and misrepresentation. This preventative approach to survey fraud mitigation complements the industry’s more common reactive fraud detection methods that occur during or after survey responses are collected.

GASP Part 2: Inferring Hidden Attributes for Deeply Curated Research Participant Pools

GASP’s inference capabilities enable us to predict participant attributes that are otherwise unavailable via traditional recruiting methods. When a client needs to conduct research with users of specific enterprise software or professionals with niche expertise, GASP’s models predict likely matches in our database—dramatically expanding the range of specialized research we can support.

The result is a virtuous cycle: better participant targeting enhances participant engagement, which leads to more accurate data being fed into our models that further improves their targeting precision. Each successful project strengthens our data advantage and widens the gap between NewtonX and traditional market research approaches.

Beyond Targeting: The Ripple Effects of GASP

While GASP began as a solution to targeting challenges, its impact will extend far beyond that initial focus. By improving the fundamental quality of participant matching, we enhance the reliability of research insights across all our offerings. Clients receive not just data, but data they can trust to drive critical business decisions.

This quality improvement has positioned NewtonX as the go-to partner for specialized research that would be difficult or impossible to execute through traditional methods. When companies need to understand the perspectives of specific decision-maker segments or users of particular technologies, they increasingly turn to NewtonX because of capabilities that simply don’t exist elsewhere in the market.

GASP represents something equally valuable for our team members: proof that NewtonX is a place where innovative thinking can thrive and transform an industry. Building and scaling GASP has required us to blend data science with deep domain expertise, bringing multiple teams together to solve this problem for our clients.

Building for the Future: What’s Next for GASP and NewtonX

As proud as I am of what we’ve accomplished with GASP, I’m even more excited about what lies ahead. We’re continuously expanding the range of attributes our models can predict, opening new possibilities for specialized research. We’re enhancing our accuracy through model improvements and the growing dataset from each successful project.

Leading the development of GASP has reinforced a core belief that has guided my product career: the most valuable innovations often come not from creating entirely new technologies, but from applying existing technologies in novel ways to solve persistent industry problems.

At NewtonX, we’re committed to continuing this journey of data-driven innovation. Whether you’re a potential client seeking research capabilities that go beyond what traditional methods can offer or a market research professional looking to understand the industry’s direction, please don’t hesitate to contact us.