With a heavy focus on AI and advanced technology, the critical contributions that human intervention makes in ensuring the accuracy, consistency, and reliability of data can be overlooked.  In marketing research, in particular, it is crucial that humans remain at the center of the process to interpret the data, give it context, and draw and communicate actionable and meaningful insights. Human-centered design provides the tools to ensure that human expertise and experience are an integral part of improving and maintaining data quality.

Key takeaways:

1. Discover the benefits of human-centered design and examine the critical role of humans in improving and maintaining data quality.
2. Learn how to facilitate at least 3 different human-centered design methodologies so that you can apply them right away.
3. Hear stories and best practices about how human-centered design has improved data quality for peer organizations.

- Presented by Rebecca Farabaugh

Deep dive into the evolving landscape of data quality and trust with Full Circle Research, A+E Global Media, and Procter & Gamble as groundbreaking insights from a recent data quality test are shared. The test compares survey results from panel versus non-panel sources and evaluates the impact of security checks on data integrity.

Procter & Gamble will also discuss their bold move to mandate ISO 20252 certification for all online quantitative suppliers - a decision that’s setting a new standard for data quality and transparency in the industry. This session will provide actionable takeaways on how to secure data integrity, improve respondent quality, and build trust at scale.

Alexandrine de Montera, Full Circle Research, Vanilynne Gulla, A+E Global Media, Megan Budzynski, Procter & Gamble

AI has the potential to enhance data quality, but only when applied correctly.

When AI is thoughtfully integrated with human oversight, it not only safeguards data quality but also enhances the accuracy of insights while minimizing inefficiencies. 

Join this session to learn how: 

• By streamlining validation and analysis, AI can save time and resources, while misuse can have the opposite effect.
• Over-filtering can shrink respondent pools, while weak tools may allow poor-quality data to slip through.
• AI-powered tools for open-ended response validation—such as text clustering, biometrics, and automated analysis—help improve the integrity of qualitative data.
• When researchers actively flag issues and suppliers refine their respondent sources, AI detection capabilities improve.
• Relying on outdated fraud detection methods puts data quality at risk.
• A structured feedback loop further enhances detection, reducing the manual workload for researchers and ensuring greater efficiency.

Presented by: Mike Herrel, Director of Data Solutions at The Directions Group & Mayank Agrawal, Co-founder and CEO of Roundtable

While researchers prioritize fraud prevention, they often overlook respondent disengagement, which can undermine data quality and business insights. A poor survey experience can negatively impact not only a single study but the broader research ecosystem, potentially leading to respondent loss. Dynata has developed a measurement system with eight key variables (e.g., interview length, abandon rate) to assess a study’s impact on respondent engagement. This system helps clients optimize their surveys for better data quality and retention. In this session, we’ll explore these measurement components, share real respondent feedback, and discuss actionable ways to enhance the survey experience.

- Presented by Tammy Rosner, Ph.D., Principal, Research, Dynata

Data quality control questions (also known as “red herrings”, attention checks, etc.) have been used in online surveys for decades now.  They can improve the data quality of the survey; however, if too convoluted, there is a risk of “over-cleaning” the data, alienating, and losing valuable responses. aytm conducted research to assess the utility of a new generation of data quality control questions, Qualchas, that aim to combat not only inattentive survey respondents, but AI-generated survey responses and survey farms, as well while making survey taking experience more engaging for thoughtful survey takers.  

Goals of this research include: 

• Helping other researchers design effective data quality control questions. This includes measuring attention and helping identify potential survey farms/professional respondents as well as LLM-enabled bots.
• Providing insights into cultural differences in respondents’ perceptions of data quality control questions.
• Teaching insights professionals how to apply an empathetic approach to quality control design (We should never forget the real respondents who are actually paying attention to the survey).

- Presented by Rossi Dobrikova, aytm