“Big” and “fast” are two words that aren’t normally associated with one another: Things that are big (oil tankers, continental plates, etc.) tend to move slowly. But what about with big data? In life sciences and health care many argue that a slower and more calculated approach to big data will provide higher quality and more valuable insights over the long run. But on the flip side, in an industry where decisions mean life or death for a patient, the data often needs to be fast. Moreover, insights are perishable – and especially in this industry, their value simply may decrease over time.
Is there such a thing as “fast data” in a big data world? Or “fast data” in life sciences and health care? Absolutely. The concept of fast data is not entirely new. Many other industries have been actively pursuing this for at least a couple of years. And now, propelled forward by several major industry developments in analytics, it’s likely on the verge of becoming an everyday reality among life sciences and health care organizations.
Here’s what many innovative companies are starting to do:
• Giving data its own platform. Many large organizations have their own dedicated web server, email server, app server, you name it. But a dedicated data server? For some reason data has often been handled differently over the years – but that’s changing. Users are engaging with data at an API level, and deriving insights. And as these insights are derived, scaling this knowledge and making it repeatable is becoming a challenge. As it changes, organizations are beginning to implement platforms with dedicated data servers. Essentially, this can allow a wider range of people across the organization to draw from the data, extracting only what they need, when they need it, in highly visual formats that make it easier to digest. As insights become additively built on top of shared knowledge, the speed to market is considerably increasing.
• Commoditize analytics. It’s a well-established fact that the analytics world is becoming more commoditized by the day – and in many important ways that’s a good thing. Commoditization can make analytics accessible. It’s no longer just the elite data programmers who can interpret what data is telling us, this move allows almost anyone to become citizen data scientists.
• Move from big data to fast data. Big data isn’t going away – but instead of focusing on amassing and managing huge volumes of data, many are seeking to work with data instantly, in order to generate insights instantly. This is a significant shift. Take a slice, examine it, compare it to similarly constrained slices of data from elsewhere (either inside or outside the organization, or both), shake some insights loose, rinse, repeat.
What will these shifts allow? Consider how you plan a trip. Today, maybe you visit several websites as part of your planning process – one for flights and hotels, another for reviews, another to scope out the attractions in your destination, another for weather, and so on. That’s roughly how analytics may currently work in your organization, with users going to different sources to cobble together what they need.
But now and in the future, with developments like those above (not least the trend towards single data-dedicated platforms), it’s possible for individual users to have their own personalized “sites” where they can explore data for their own use. This site could draw from all the other relevant sites and data sources (and, let’s be honest, manual spreadsheets), both internal and external, allowing users to accomplish in a matter of minutes what used to take days of searching and lots of analytics muscle. What’s more, this approach could allow for the research to be replicated and shared, giving the user a solid starting place instead of reinventing the wheel every time. Fast data, faster insights.
Is that what analytics looks like in your organization today? If not, you likely should be asking more questions about what’s possible. Here at HIMSS, I’ve fielded lots of questions about these very issues, from leaders who can’t afford the status quo. Can you?