Quality should be the North Star of the consumer data industry, but bad incentives and lack of education has delayed the industry’s appetite to fix it. Buyers pushed for lower prices and faster turnaround times as key factors determining with whom they work, so suppliers prioritized speed and cost over quality. When you consider that $1 trillion of business and advertising decisions depend on the accuracy of market research, this is unacceptable.
But that is changing. We can no longer ignore the glaring survey quality issues. Realeyes is proud to be part of a recent Quality Pledge, which currently has 31 and counting signatories. Both buyers and sellers are acknowledging that we have to do something about data quality today.
For the consumer data industry to grow and for buyers to trust the data they’re getting, the industry now recognizes the need for a new industry standard.
Addressing Fraud Head On
The types of false data and fraud have proliferated in recent years. From bot farms to intentional and unintentional duplicates, there are strong incentives for bad actors to commit fraud against data providers. It’s a lucrative field, and the major culprits will continue to explore ways to stay ahead of quality checks.
The next big arms race will revolve around AI; both for industry organizations to improve how they block fraud and for bad actors to try to thwart those checks.
|
“When there’s money to be made, there’s going to be fraud, and when there’s big money to be made there’s going to be big fraud” Tia Maurer, Group Scientist at Procter & Gamble |
Respondent UX
Of course, the quest to protect surveys against fraud has either involved very little verification or onerous checks like having a hold an ID up to a web camera. Any successful anti-fraud technology will need to be intelligent enough to avoid letting through fraud while also avoiding false positives and lightweight enough that it doesn’t deter real panelists from participating. There are several promising applications using lightweight AI technology powering facial verification that could solve this issue.
Discussing the Quality Pledge
We believe the entire industry needs to come together to advocate for market research untainted by fake engagement. That’s why we joined a quality pledge and recently hosted a webinar with other industry leaders like Procter & Gamble, Kantar Profiles, ESOMAR, and Innovate MR.
In a study of supposed quality individuals supplied by some major panel providers, Realeyes found 40% were of dubious or poor quality.
An impetus behind the pledge is the importance of sharing indicators of quality. Some of the quality pledgers have begun sharing anonymous data. Analyzing over 300,000 respondents, we’ve found only 62.72% are currently deemed good quality.
|
“I’m a big fan of the pledge because you’re not going to get good, quality answers from good, quality people if it continues the way it’s going” Caroline Frankum, CEO at Kantar Profiles |
Pledgers are also committing to ensuring all staff and suppliers are trained in best practices on quality, including the principles of this Pledge; and sharing examples of bad practices.
“I’m a big fan of the pledge idea because you’re not going to get good, quality answers from good, quality people if it continues the way it’s going,” said Caroline Frankum, CEO at Kantar Profiles.