Cyber security continues to be one of the biggest threats impacting modern society, as most of our day to day experiences are played out virtually. We are hyperconnected through smart devices that have reshaped culture and transformed how we live and work. From online wallets to social media, smart speakers to drones, as consumers, we have served up a virtual smorgasbord of data points for hackers to prey on. Companies amassing these data points are especially at risk as they are the keepers of purchase behavior, transaction history, financial information, and more.
Which brings me to the online sample industry, which is 100% online. Given the consumer data we collect, i.e., demographics, email addresses, etc., and our methodology for sourcing respondents, we, too, are exposed to cyber-attacks that compromise the quality of our data. To mitigate the risk of delivering bad sample to your clients, consider these best practices:
The integrity of your sample is your competitive advantage. So, all sample companies brag about their quality control practices. In terms of how that’s done, we see the usual suspects: proxy survey detection, digital fingerprinting, IP filtering, speed checks, and trap questions. These tools are the industry standard and are by far the most effective methods to prevent or catch fraudulent respondents. However, some fraudulent respondents or what I call “sample hackers” still get through by deploying ingenious tactics like using multiple devices with a combination of bots and URL manipulation to complete surveys without actually taking the whole survey just to get the incentive. As you can see, more is needed to prevent fraud than just software. Let’s add two more layers to the security process.
The recruitment process sample companies use to recruit respondents matters. At ThinkNow, our hands-on approach to recruiting, including validations by phone and double opt-ins, enables us to build online panels that give us control over verifying panelists. Programmatic Sample companies, on the other hand, don’t have that luxury as there is no human intervention. They must thoroughly vet the sample companies they partner with to ensure those data quality checks are met. Our call center representatives process approximately 2,000 calls per month. Since the bulk of sample hackers come from the internet, validating potential panelists by phone enables us to eliminate a larger percentage of fraudulent respondents, which leads me to our last layer – your team.
The final layer of data security happens on the ground. Create a team responsible for analyzing the data and weeding out what your software or validation methods failed to detect. Be on the lookout for inconsistent responses, quick response times, response times at odd hours, or suspicious e-mail addresses, then take the proper steps to remove that data.
Ultimately, all three layers are needed to gather the best data quality and fight off sample hackers. Data quality controls are very effective, but they don’t eliminate all fraud. This three-pronged approach is your best bet to ensuring online sample data integrity.