In business, as in life, the "five P's" continue to hold true: Proper Planning Prevents Poor Performance. But as we're now in the midst of the digital revolution, we would probably do well to modernize and improve that venerable bromide by adding the "five V's": volume, velocity, veracity, variety and value.
Let me break that down: the five Vs are critical criteria in assessing when and how to capitalize on the benefits of a digital revolution that has snuck up on -- and is constantly being fueled by -- all of us in every aspect of our professional and personal lives, according to a compelling new report from the University of Pennsylvania’s Wharton School of Business.
According to the study, smart data revolves around the five aspects of volume, velocity, veracity, variety and value.
Citing an example from the healthcare world, the report shows why those five V's are essential in creating an effective digital strategy in today's always-on world:
At a hospital, volume is important in the case of patient-level data on medical history, physical history and genetic family traits. Similarly, clinical data related to laboratory tests and physician’s visits or administrative data on payments are increasingly becoming voluminous.
Velocity refers to the need to digest and conduct real-time analysis of the data that is being generated at a rapidly increasing pace owing to technological advancements. For example, in healthcare, velocity would be important if patient data relating to MRI or CT scans, or weight changes, are captured and transmitted in real time over cloud platforms to others involved in a treatment regimen such as a medical device manufacturer, a pharmaceutical company or a diagnostic center.
Veracity is critical for such treatment-related data, especially if it is sourced from systems at several organizations in the healthcare ecosystem.
Pharmaceutical companies, among others, will no doubt find themselves grappling with a variety of data feeds, such as from their marketing channels, social networks, surveys, physician networks and medical or scientific journals. Here, standardization of information would play a critical role.
Value is a critical aspect as it’s important to ensure that the right insights are generated from the data and they lead to measurable improvements in patient outcomes.
Identifying the right metrics is also important in teasing outsmart data, the report says. For example, tracking the time consumers spend on a website may be a “false metric” because that website might be poorly designed, making navigation difficult and time-consuming. Also, the study notes that the revenue streams at most businesses are irregular, not subscription-oriented and have non-contractual settings.
Any organization embarking on its smart-data journey must progress through three levels of maturity, the report finds. These include:
• Basic data analytics: Here, organizations could get a reasonably good first glimpse with fairly simple processes. For instance, previously a consumer packaged goods company would typically track customer usage patterns and attitudes, point-of-sale data, household survey data and shipment data. Now, they could harmonize all that with data from social networks, mobile tracking systems and so forth. “This is basic stuff, not sophisticated analytics, and it could generate up to 200 metrics,” the report finds. “But with these alone, they have gotten so much smarter about their data to get a 360-degree view of a brand or a category or a customer. More and more companies are realizing the power of doing this at the first stage.”
• Advanced analytics: Here, sophisticated analytics models are used, going beyond the conventional regression methods that are often not very predictable. Often, regression methods show correlation but not causality. Companies also invest in applications to automate basic analytics models to reduce processing times and improve accuracy.
• Cognitive or intelligent analytics: Insights gained from advanced analytics lead to new-age tools such as artificial intelligence, machine learning and smart data discovery. Artificial intelligence/machine learning tools are the foundational elements of a cognitive organization, and they enable an organization-wide approach to analytics. The algorithms that such tools use iteratively learn from data and find hidden insights without being explicitly instructed where to look. For example, machine learning could help a retailer run a personalized campaign with offerings tailored for specific individuals, and on a mass scale. In such exercises, a retailer would do rapid experimentation by testing and retesting a campaign offering, feed the results in real time to its personalization model, thereby constantly refining it for each individual customer. “Ideally, every person is a market segment,” the report says.
The process of extracting smart data begins with identifying the end-issues that need to be addressed. These could be outcomes like cost savings, efficient decision making, the ability to serve customers better or improve customer engagement.
Asking the right questions is critical. Give us a call so we can walk through your questions together.