How A Better User Experience = Better Data Quality

By James Wilson

Veriglif is a blockchain and smart contract powered personal data marketplace that validates and links individual records across multiple sources. With permission and transparency, these records are able to be used in real-time for a variety of use cases such as advertising targeting, survey sample and data enrichment.

During the early stages of the COVID-19 pandemic, while most companies appeared internally focused on cost cutting and preservation of existing revenue, Veriglif set about helping a partner to grow revenue. This partner interacts with approximately 6 million individuals each year and was looking to leverage this with other income opportunities. As a soft launch, a portion of these individuals were targeted and invited to join the Veriglif network, and to our pleasant surprise over 60% opted-in to view, control and profit from their personal data.

With our growing network of members, we also set out to improve the experience of taking surveys in the existing sample marketplaces. We found that because the survey rejection rate was so high (even with programmatic profile matching) and because there were so many middleware routers asking the same pre-screening questions that after only two or three survey attempts our members were simply giving up and not coming back — the experience was not what they expected, and we were embarrassed.

In order to accurately match members with the best opportunities, a complete profile is required. However, asking profile questions during signup, especially during the survey pre-screening process, caused our members to drop off and lose interest exponentially. We needed to passively and transparently profile our members as they use the network.

When a panel sends one of its members into a survey it’s largely a black box, and what comes back is a status of whether the individual completed the survey or not and occasionally some pre-screening questions; if they were asked by the marketplace itself, not the additional middleware routers. Because there is so much unknown, questions arise such as:

• Was the survey mobile friendly, difficult to complete, or simply broken?
• Were there multiple middleware routers in-between the marketplace and the survey, each asking more pre-screening questions and taking their slice of the CPI?
• Were there additional qualifications in the survey not listed upfront?
• Did my member answer differently in this survey than one taken previously to get past qualifications?
• Did the survey attempt to recruit our members into another panel?

To address these problems of a poor experience, difficulty in obtaining a complete profile and transparency, Veriglif built a browser extension and a soon-to-be released mobile application. We tested it both internally with our members and externally with a white labeled version to our partners’ members. In our trial during January 2021 we captured 4,817 surveys and 8,579 questions — our results showed many opportunities for improvement.

This solution detects when the member is taking a survey from one of our 100+ supported platforms and helps in a number of ways:

• Uses advanced AI to categorize and auto-complete the pre-screening and survey profile questions from the members profile saving them time. In our trials each member spends 36% of their time completing profile questions, imagine how happy they are to not answer these repetitive questions again and again.
• Captures and auto-updates the members profile in real-time. This means we can onboard our members with a simple 3 question profile and auto-update it as they complete surveys. This saves everyone time and money, the panel no-longer has to incentivize their members periodically to update their profile and the member only needs to answer the same question once because it’s remembered.
• Provides full transparency and provenance to the member and panel about what exact data was provided to each site. This is especially important if they have given a few pieces of PII to a number of middleware routers before actually starting the survey — imagine the challenge of requesting all that information be removed, where would you start? This transparency also allows the panel to see exactly where a potential recurring issue is and if required remove that survey and prevent other members from having the same issue.
• Protects the survey from data quality issues by flagging those that have different data from one survey to another. For example, if a member changes their employment industry for each survey to attempt to qualify our system will flag it automatically and in real-time. In our trial, 34% of profile data questions had inconsistencies — not always due to fraud, but other issues such as laziness or simply the question options did not match — but regardless of the cause we can flag that survey as having data quality issues.

We believe this has many more applications than those listed above and our models show at least a 3X ROI. It’s time to innovate not just the way personal data is managed and used, but the actual underlying processes that create time inefficiencies, user churn, throughput breakage, economic waste and fraud. We continue to reverse engineer and deconstruct the existing system to find ways to address these challenges and bring them into production. The more we do that, the more we see that the Veriglif model continues to be the right path to transforming the entire market, and that is very exciting indeed!

If you are interested in exploring this solution as a white labeled option for your members contact us at james.wilson@veriglif.com or call +1 888 864 0640.