Brookmark is a full-service, custom market research firm. In 2020, Brookmark significantly invested in managing data quality. Programmatic buying and selling of data is a huge boost for information availability, but has brought challenges with regards to quality. As more and more fraudulent respondents enter the ecosystem via programmatic delivery, a lack of true answers, and a lack of engagement threatens the core product of our industry, which is actionable insights.
Brookmark is partnering with SampleChain, a market research data quality platform to ensure the integrity of the data and the data sources of Brookmark’s research. Brookmark will invest in the following technologies to ensure that data quality continues to be of a high and acceptable standard.
Digital Fingerprinting
Digital Fingerprinting is a standard form of machine identification. Different identifiers from a respondent’s machine are tracked to ensure that duplicative respondents do not participate in research . These details include items like browser information, machine information, cookie or cookie-less based tracking. Other appropriate data points that help identify click farm usage, bot traffic, participants who have been a part of e-commerce or financial fraud, etc. will also be tracked, identified, and prevented in real-time.
Any form of digital fingerprinting will adhere to the CCPA (California Consumer Privacy Act) and General Data Protection Regulation (GDPR) standards.
Real time Text Analytics
Brookmark will also leverage custom built text analytics to ensure that answers that are original and real will be accepted. These checks include identifying plagiarism from across the internet, gibberish, contextual answers, and copy/pasted responses. Foul language, and other forms of unacceptable answers will be removed as well.
Text analytics is currently the best indicator of future research validity, and correlates with a high degree of precision with scrubbed out respondents.
Machine Learning
All of Brookmark’s traffic will be subject to rigorous machine learning principles to ensure that every single respondent participating in Brookmark’s surveys is of high quality. The machine learning principles include regression techniques to help analyze past respondent behavior, feature engineering to ensure that only engaged, non-professional survey takers are participating in research studies. Furthermore, anyone seen participating in multiple, simultaneous surveys will be eliminated from the data sets as well.
Engagement Tracking
Last but not least, Brookmark will also ensure that professional survey takers and over extended respondents are removed from our datasets. Research has consistently shown that professional survey takers have an outsized, adverse effect on research results. RoR has consistently shown these respondents tend to over state usage of products, and provide highly positive sentiment on product usage.
Brookmark and SampleChain will continue to invest in these and other forms of research technology to guarantee data quality. As more and more supply overlaps and is programmatically delivered via networks, the industry needs to bring robust fraud detection techniques to market.
About Brookmark Research:
Brookmark is a research operations company focused on managing the critical components of the research process including project management, survey design, programming, hosting, sample design & management, data management and reporting solutions. Brookmark provides a full outsource solution for all your research needs.