Big Data Improves The Customer Experience Through Silo Integration

Buffer

I had the privilege of delivering a talk on the topic of Big Data and Customer Experience as part of the Customer Experience Professional Association (CXPA) Bay Area Local Networking Event in San Mateo, CA on February 26, 2014. I would like to thank the CXPA for hosting and organizing this event. Below are the slides to my talk, followed by some information about the slides.

The goal of my talk was to provide an overview of the field of Big Data and to show how the integration of different data sources can give you deeper insights about your customers than any single data source by itself. By connecting different data sources to customer feedback and analyzing these data in a “customer-centric” way, you can identify the causes and consequences of satisfaction with the customer experience.

Don’t You Ever Talk about Big Data – Slide 2

Because I was presenting in the Bay Area and I’m from Seattle, I thought I would shock the audience into attention by showing the Seahawks logo. In the spirit of Richard Sherman and the Legion of Boom, “Don’t you ever talk about Big Data!”

Big Data Overview – Slides 3 – 9

While the three Vs provide a good description of the type of data we are now managing, I think Big Data is less about the data and more about the value we get out of it. We have always had Big Data; my graduate school professor told me that, for his dissertation, he did a factor analysis (with rotation, mind you) on a calculator (computers weren’t invented yet). He had a Big Data problem over 50 years ago. To me:

Big Data is an amalgamation of different areas* that help us get a handle on, insight from and use out of data.

* includes technology (data capture, storage and management, BI reporting) and analytics. I might now include “organizational development.”

The Field of Big Data is growing. Wikibon estimates that the Big Data industry will grow to become a $50B market by 2017 (it was $19B in 2013). Even though businesses continue to adopt Big Data solutions to primarily analyze internal data, they say that they are not seeing improvements in decision-making. Big Data vendors are trying to solve this problem by focusing on analytics to deliver solutions that improve marketing, sales and service processes.

Analytics of Integrated Data – Slides 10 – 11

Research has shown that companies who use analytics outperform those who do not (see here, here and here). Additionally, companies who integrate their customer feedback data with other business data (e.g., operational, financial, constituency) outperform their counterparts who keep their customer feedback data isolated. You can learn a lot more about your customers by looking at all your business data together.

Customer-Centric Approach to Analyzing Data – Slides 12 – 25

Getting value from your integrated data requires that your analysis be focused on the right data sets with corresponding metrics. Depending on the question you are trying to answer, you will need different metrics from different data silos, organized in the right way. I offer a few general customer-centric data models that help you answer your business questions, whether they address strategic issues (i.e., What is the ROI of our customer experience program?) or tactical issues (i.e., Which operational metrics have the biggest impact on customer satisfaction?).

The Value of Integrating Data Silos: An Example – Slides 26 – 34

Selecting your first Big Data project requires that you identify all your data. Define your problem statement and establish a compelling use case to help you measure ROI. I offer an example of a Big Data project by using real customer experience data from US hospital (Thanks, Medicare) and other US hospital data sets that housed health outcome metrics, process metrics and financial metrics. While each of these four data sets alone provides much information to help healthcare consumers (e.g., map of US hospitals using patient experience ratings, map of US hospitals using health outcome metrics), integrating them allows us to understand the interplay of these different metrics. We can test new hypotheses with this broader data set: Is patient experience related to health outcome metrics (yes, they are related; hospitals who provide better patient experience also have higher survival rates)?; Is healthcare spending related to the customer experience (no relationship found between patient experience and Medicare spend)?

Data Science and Data Veracity- Slides 35 – 40

Business leaders are relying on data scientists to improve how they acquire data, determine its value, analyze it and build algorithms for the ultimate purpose of improving how they do business. Data science skills fall into five skills/abilities/talents. These are: 1) quantitative methods, 2) computer engineering, 3) business acumen, 4) communication and 5) the scientific method. The scientific method reflects how to approach a business problem with the use of hypothesis testing. Hypothesis testing helps us understand how the world really works. This should be our ultimate goal as data scientists.

Data veracity is about the accuracy and truthfulness of your data and the analytic outcomes of those data. While other Vs have dominated the discussion in this data-abundant world, I think that Veracity is the most important V of data, big or small. The scientific method accompanied by solid critical thinking skills, however, can help you verify the veracity of your data as well as your claims. To improve the data veracity of your Big Data projects, you need to: 1) Have a hypothesis: With huge data sets and powerful analytics, we are more likely to find a significant yet spurious correlations; 2) Be aware of your biases: so, don’t cherry-pick those spurious correlations to support your current beliefs; 3) Know your data source: Poor decisions will be made when poor data are used, so know what your data mean; and 4) Know your customer metrics: Before using customer metrics, be clear on what the metric is measuring, how the metric is calculated and ensure it has good measurement properties and is useful for the problem at hand.

Implications and Online Resources – Slides 41 – 43

There are three broad implications of Big Data to the field of customer experience management: 1) You can now ask/answer bigger questions; 2) You can build your company around the customer through data integration and customer-centric analytics and 3) You can use real loyalty metrics rather than relying solely on self-reported loyalty metrics.

I’ve been following the Big Data space for over two years now, reading and writing on the topic. I provide some online resources that I use on a regular basis.

Additional Slides – 44 – 51

I provided some additional slides I didn’t discuss in my talk, including Wikibon’s estimate of Big Data Market, Gartner’s Hype Cycle for Big Data, and a Customer Loyalty Measurement Framework.