How can marketers better leverage the scads of data at their disposal? Hundreds, if not thousands, of articles and books have been written on the topic. While most marketers globally admit the value of tapping into big data, many are challenged with converting the possibilities into reality.
A critical mistake companies often make is to overlook an important type of data that can provide valuable insight: first-party data, according to Jay Marwaha, president and CEO of Syntasa, a behavioral analytics company. But in the quest to convert big data into “big insight,” a company’s own customer behavior data is the most valuable. Granted, converting customer data into insight is a complex process, but the competitive advantage gained by doing so is powerful.
I recently spoke with Marwaha about the value of first-party data. Our conversation included the differences among the types of data marketers can access, how first-party data can be leveraged to make better decisions, and steps marketers can take to identify and tap into internally generated data.
Whitler: What is first-party data, and how does if differ from second- or third-party data?
Marwaha: In order to understand first-, second-, or third-party data, let’s talk about how companies collect data. First, companies collect known and unknown visitor data through online and offline interactions with their brands. Every visit to a company’s Web site from different channels is considered an event. Some of the information related to these events are recorded in cookies. Companies place these cookies on a consumer’s device so that they can provide the consumer with an improved experience—for example, topics or products that they have demonstrated a past interest in.
Another way companies collect consumer interaction data is by placing page tags on their sites. Page tags are a small piece of software code that records your every interaction with the site. These are one form of first-party data, with others being customer records, purchase records, and loyalty program data. Think of first-party data as data collected around your direct interaction with customers. Second-party data can be defined as shared first-party data. For example, an airline may share its first-party data with a rental car company. Third-party data is simply external data. Examples of this are purchased data from a vendor, census data, weather, or geolocation information.
Whitler: Why are companies overlooking first-party data?
Marwaha: I think there are several challenges working with first-party data:
• Sheer volume of data: Companies are collecting a massive amount of data from every visitor to their sites and apps. Trying to gain valuable insight from this huge and continuous flow of data is quite the formidable task. It’s like drinking from a fire hose.
• Separation of marketers from typical IT and enterprise data: Over time, the marketing organization has grown into a technology silo unto itself. While general infrastructure support still exists within the IT function, the job of understanding how technology can be used to drive e-commerce and marketing analytics from first-party data has moved to the marketing organization. So marketers have been struggling to understand the technology available as the amount of data and competitive pressures grow. Just ask any marketer if they understand how to utilize something like Hadoop. I doubt you will get a clean and concise answer.
• The complexity of the data itself: The vast and continuous stream of first-party data coming from a site alone can be daunting to try and consume. Furthermore, there is a significant challenge associated with trying to integrate different data sets. Just organizing and integrating the data is a significant issue, which is why marketers turn to third-party data providers; the data is already organized, integrated, and cleaned. The marketers are not only overwhelmed due to the amount of the data, but the complexity of the data is equally beyond their realm of understanding or ability.
• Whitler: Do you have an example of a company that is effectively leveraging first-party data?
Marwaha: Many retailers and e-commerce companies are still struggling to understand the visitors to their Web propertyies. The majority of the opportunity is with those visitors who are not known to the marketer, or anonymous visitors. The most effective way to address anonymous visitors is based on their actual behavior. By understanding historical behavior from first-party data and developing analytical models that can learn and evolve over time, we can now determine what behavior is most likely to conclude in a successful conversion. Delivering content and offers specific to the behavior of the visitor is much closer to actually personalizing an unknown visitor’s experience and increasing the likelihood they will make a purchase. Leveraging the first-party data to understand visitor intentions on that visit is, by far, more effective than simply analyzing previous purchases.
As an example, we worked with a technology firm to help predict which customers were at risk of leaving based on their Web behavior activities. Based on this insight, the company was able to retain 80% of the customers at risk by developing and deploying an outbound calling program. During these discussions, we found out that most of these customers were experiencing technical or product issues and needed help. The different behavioral patterns online helped us identify there was an issue.
Whitler: What advice would you give CMOs to begin identifying and leveraging first-party data?
Marwaha: Consider first-party data as the closest version of the truth related to what is happening on your Web site. Because it is. While insight regarding third-party data and analytics is important, the foundation of your consumer understanding has to come from your consumers and what they are intending to do while on your site. You spend a lot to get visitors to your site. Capitalize on their visit by understanding their behavior and intended behavior. And then try to influence that behavior to your benefit.