Allstate Identifies Out-of-Pattern Trends

Marta Magnuszewska, Special Investigation Unit Analyst


Allstate is the largest publicly-held personal lines insurer in the U.S.

Before it was always numbers in rows and columns. Now I’m eager to see when I do a new analysis, because it's so much easier to actually see the trends, the story, and the power the data brings.

The Allstate Corporation, America’s largest publicly held personal lines insurer, is known for its "You're In Good Hands With Allstate®" slogan. Now celebrating its 80th anniversary as an insurer, Allstate is reinventing protection and retirement to help nearly 16 million households insure what they have today and better prepare for tomorrow through insurance products for auto, home, life and retirement. At the 2011 U.S. Tableau Customer Conference, special investigation unit analyst Marta Magnuszewska shared how Tableau empowers her team spot out-of-pattern trends so field workers can investigate them to discover the root cause.

THIS IS NOT AN ENDORSMENT. Neither Allstate nor its subsidiaries endorse or recommend the use of Tableau Software products. While Allstate utilizes Tableau, you should review the product in detail to make sure it meets your needs prior to purchasing it. Further, Allstate is not providing any advice on the use or application of the product in conducting the demonstration. Data used to create this demonstration has been adjusted and does not include any proprietary Allstate information.

Tableau: How are you using Tableau at Allstate?
Marta: I work at the special investigation unit for claims department and for the data analytics team. We take data and we look at the trends of the data, and try to see if there are any out-of-pattern trends. We use Tableau to do a lot of that analysis, and once we see those out-of-pattern trends, with Tableau capabilities we are able to drill into the data that's behind those out-of-pattern trends.
Tableau: What kind of insight does Tableau help you discover?
Marta: We do a lot of geographical analysis. So, we compare state to state or a state with its own territories, and see which geography is driving those analyses, and then we try to see from the claims perspective, who is driving those claims. And our unit does a lot with the medical claims data, so then we are able to get a little bit further and see a little bit more information from the medical perspective of the treatments and details. We share our work with the adjustment and the investigators in the field. Then they can take further steps upon what the data is showing us.
Tableau: How has Tableau changed the way employees in the field work?
Marta: It's easier for our investigators and adjusters to work with that data, because now it's more in one place versus many different data files. And that's how it helped them tremendously on saving time.
Tableau: Has Tableau helped communication with partners outside of the company?
Marta: Well, visual examples are great for us to explain the story, and not only to our investigators and adjusters, but then we are able to take the data elements that we have and all other work and just share this with our attorneys and the law enforcement to take action.
Tableau: What’s the overall impact on your organization?
Marta: It actually helps us to see everything from a big picture. I always give an example of peeling down the onion, because once we see the trends, once we see the overall picture, then we can peel it down and see who is driving those trends, where are we going with hose trends, how can we can take actions to maybe stop those trends.
Tableau: How has Tableau impacted you personally?
Marta: For me personally I think Tableau make me more creative of a person and helped me to appreciate data. Before it was always numbers in rows and columns, where now I am actually eager to see when I do a new analysis what my trends are, because it's so much more easier to actually see the trends, the story, and the power the data brings.
Tableau: What do you mean by the “story”?
Marta: When I create my work in Tableau, I always think about the story. There has to be a beginning, it has to start somewhere. It has to evolve to more information, and then it has to end with the capability of drilling behind selected elements of the data.
Tableau: How do you use Tableau to present data as a story?
Marta: I use dashboards, because it gives me the ability to combine my work. My data is not at one place all the time. Sometimes I use different datasets. So, I'm able to use dashboards to show two different data elements at the same time, and it looks nicer to the end users and to my audience.
Tableau: Next to dashboards, what other Tableau features do you find helpful?
Marta: We use maps all the time, and I have been experimenting with those maps a lot through polygons and many different geographical areas. I really like mapping because it's not only showing us where it is geographically located, but we can drill into the street level. We can calculate distances. That's actually one of the biggest wins that we really like, because we can see the distances, and mapping and polygons. All of that is just bringing that much more information and visualization for us to work with.
Tableau: Can you describe the volume of data you work with?
Marta: When we started using Tableau, I was only connecting to text files in Excel. But as I moved forward with the implementation of Tableau and getting to know Tableau better, and the capabilities of Tableau, we also outgrew that, and those were not big enough files for us to store the data. So, right now we are connecting to Oracle and SQL Servers, and our data sizes run from few thousand records to as much as 20 million records. And that's just great because we have everything in one place, we don't have to restrain this to an area or a geographical area or just the particular analysis; we can look at the big picture.
Tableau: What about the speed to insight possible with Tableau vs. Excel?
Marta: One major thing is Excel will not be able to hold 20 million records. The second thing is the capabilities of different graphs. So, although we do have graphs and functions within Excel, it's much more of a manual process. In Tableau, graphs get prettier. And definitely it's just quicker to drag and drop versus making sure you select everything in Excel. So I look at Excel as more of a manual process, where in Tableau it's more automatic, faster, and efficient.
Tableau: Any final thoughts on using Tableau?
Marta: Data brings all the facts, and the facts within the data. And when you don't visualize them, when you don't show them in a correct way, you might be missing something that might play a big part in your organization. The power of Tableau visualization, the capability of Tableau visualization, and the power of data, if you combine that together, it can take any organization to the next level.

You might also be interested in...