cc: Life Science
cc: Life Science Podcast
What is a Data Strategy and Why You Need One
0:00
Current time: 0:00 / Total time: -26:22
-26:22

I met Charles Baldridge, CEO of Sol Analytica on a Zoom call. In an eclectic group of filmmakers, artists, music publishers (and data scientists), Charles mentioned that his company helps others with their data strategy. That set off my curiosity and I invited him to this podcast.

I hadn’t really thought about data strategy and didn’t know what it was, but it certainly sounded like something many life science businesses should have.

It will be no surprise to anyone that we are awash in data. And we leave a trail of it everywhere we go. (Charles calls it digital exhaust.) The question is how can we take advantage of it in our business?

Data Strategy- What is it?

…having a data strategy as a company means that kind of from the top down from CEO, all the way down through your kind of frontline workers, when somebody has a question or somebody has a, a problem that's in front of them, if they have a strong data strategy, they're thinking for.

What does the history look like? What do I have available to me to show what that is? And that's data, that's the information that is available to you. And so I think of a data strategy from that perspective, as it's the first thing that people go to. When they're thinking about making a decision as what does the data say, what information do we have available to us so that we can do a good clean job of making that decision and defending it?

It used to be expensive but it’s not any more. CEOs know in the back of their heads that they should do something. Of course, the most important thing is knowing what would you like to improve? What problem would you like to solve? Do you have data that can help? The process is iterative. With each answer you might have another question that makes you dig deeper. And of course if you can’t answer that, you may have to figure out how to get the data you need. Which brings us to the idea of having a data culture.

I’m always curious about company culture so the idea of a data culture captured my attention.

Data Culture

Building a data culture is about collecting and storing data “in a way that gets everybody speaking the same language within your cohort or your people that you're talking with.” It means actively thinking about what data you may want in the future. What should you capture? How should it be stored so you can use it when you need it?

I can’t help but picture a garage full of stuff that has been stored without a lot of organization. It can be overwhelming. Data is like that. Too much of it can keep you from doing anything because you don’t know where to start. I may take a shot at the garage this weekend. Pizza and beer if you show up.

Your data has value to others

Having a data culture is also recognizing that your data may have value outside of your own organization. Charles gave an example of medical billing. A company with a large data set could see which insurance companies paid the fastest and the highest percentage of what was billed. That data may would have value to others to forecast their own accounts receivable, for example, and even to prioritize their own billing to optimize cash flow.

Distilling data into wisdom

There are many levels to the value in the data. There is data itself which can be useful for someone else to answer questions that may be different from what you are interested in. Then there are analytics that provide insight that may be valuable to others. At some point, you might be able to create a data visualization that has value because of patterns it reveals that provides insight to anther group. Companies pay for access to that kind of information.

Using (and supplementing) your data for AI

Eventually we got around to artificial intelligence. Just as for analytics, you need to know what you are looking for. What is the problem you are trying to solve? The next step is data wrangling - apt metaphor for getting your data together, making sure you know what’s there and what’s missing. More on that in a moment.

Finally Charles suggests care in selecting the people who will help you with the right algorithms for your project.

A lot of this ties together. In some cases you might not have all the information you need. There is a good chance that someone else might have an overlapping data set that can help. You can get these through a data exchange.

…the exchange is the place where you go out and either provide that data or purchase that data.

And so in the context of AI, we mentioned, you want to know what your questions are. Well, it's iterative. You're not going to know every question. And in fact, if you're doing science appropriately, when you get the answers to questions, that actually makes more questions. And when you do those more questions, you may realize that there's additional data that you need to answer those questions and you don't have it.

So you might go to a data exchange to collect and find that information and pull it in bring it into your data infrastructure into your warehouse or your data lake. And then expose it again to your analytics tools and to your artificial intelligence tools. And then, you might be able to go and answer that question and it makes some more questions.

Think of it as a way to add columns to your spreadsheet. For example, you have a name and an address. Someone else might have name and place of birth. Now you have an extra column on your table for understanding how far people migrate over the course of their lives. Once you understand that, it may bring up other questions. How does that relate to income, for example?

I could use some data

I’d be interested to hear in the comments if data strategy is obvious to everyone or if this is something new to think about. I’m still learning. More importantly, I ‘m interested in what you would like to learn from this podcast in the future.


Contact Charles about data strategy.

Chat with Chris Conner about content for demand generation.

cc: Life Science
cc: Life Science Podcast
How will AI, blockchain and other new technologies impact life science?