Big Data

Using Big Data Analytics To Improve the User Experience

Bill Schmarzo By Bill Schmarzo August 29, 2012

For those of you who have followed my blogs, you know that one of my favorite subjects to rail against is the “unintelligent” user experience.  This is a problem caused by, in my humble opinion, the lack of effort by IT staffs to fully understand and deliver business intelligence (BI), data warehousing, and advanced analytic solutions that truly support an organization’s key business initiatives.  And this user experience problem is only exacerbated by Big Data.

The Un-intelligent User Experience

Here is a real-world example of how NOT to leverage data and analytics to communicate with your customers.  My daughter recently got the email below from our cell phone provider warning her that she was about to exceed her monthly data usage limit of 2GB.  She was very worried that she was about to go over her limit and it would start costing her (actually, me) an additional $10.00 per GB over the limit.  (Note:  the red circled “Monday, August 13, 2012” date will play an important role in this example later in the blog).

Figure 1:  Cellular Provider Email

I asked my daughter what information she thought she needed in order to make a decision about altering her Facebook and Instagram usage (since those are the two data hog culprits in her case) so that she would not exceed her data plan limits.  She came up with the following questions:

  1. How much of my data plan do I have left?
  2. When does my new month start?
  3. At my current rate, when will I run over for this period?

Understand The Decision To Identify The Relevant Questions

This use case provides a good example of the process that one can employ in order to identify the most relevant questions that need to be answered in order to support an organization’s key business decisions.  And it all starts by understanding your organization’s key business initiatives.

  • Step 1:  Understand your organization’s key business initiatives (in this case case, don’t over run your monthly data usage plan).
  • Step 2: Capture the decisions that an organization needs to make in order to support the organization’s key business initiatives (e.g., alter Facebook and Instagram usage).
  • Step 3: Identify those questions that need to be answered in order to facilitate making the decisions (e.g., How much of my data plan do I have left?  When does my new month start?  When will I run over for my current period?)

Understand The Questions To Provide A Relevant User Experience

Understanding the relationship between your customer’s objectives, decisions, and questions that need to be answered is key to creating a user experience that provides the right information (and recommendations) to the right customer to make the right decisions at the right time.

So to continue the cellular provider story, I went online to research my daughter’s key questions:

So given the results of my analysis, my daughter has nothing to worry about as she would have to consume nearly as much bandwidth in her final 24 hours (assuming that she does not sleep) as she’s done the previous 30 days.  The probability of that happening:  near zero (or the same probability of me beating Usain Bolt in the 100 meter dash).   The bottom-line is that the email should have never sent, as there was really nothing to be worried about.

Using Big Data Analytics To Improve Your Customer Engagements

Let’s say the situation was different and there was some probability greater than zero that my daughter was going to go over her data plan.  Then our cellular company should have provided a user experience that highlighted the key data and insights necessary to help make a decision about data usage.  The user experience could have looked something like the below email message (see Figure 2).

Figure 2:  Enhanced Subscriber Email

This sample email has all the information that is needed is in one spot, including:

  • Actual usage to date (65%)
  • A forecast of usage by the end of the period (67%)
  • The date when the data plan will reset (in 1 day on August 14)

With this information, my daughter is now in a position to make the “right” decision.


Our cellular company should implement an analytic process that calculates daily the likelihood that each subscriber’s data usage will end up in the red zone in Figure 2 (the 90%+ of data usage range) before the end of the usage period. The email should not only alert the subscriber of the potential problem, but should recommend some actions, such as:

  • “We notice that you use Facebook and Instagram frequently, try to use a WIFI connection option more often to reduce your network data usage,” or
  • “We notice that you approached the data usage limit each of the past 3 months.  We would like to offer you an upgrade to the next level of data plan at a 75% discount.”

And my provider is not alone in missing opportunities to leverage the data that they know about their customers in order to provide a more relevant, more meaningful customer experience.  Organizations know a lot about their customers and their buying and usage patterns, but little of that information is packaged in a manner that improves the user experience.  Big Data will only exacerbate this problem, and organizations will either learn to jump on Big Data as an opportunity to improve their user experience, or they will get buried by the data and continue to provide irrelevant customer engagements.

And if you haven’t already registered, don’t miss the opportunity to attend my upcoming webcast “Analyze This! Best Practices For Big And Fast Data” where we will discuss one of the most important trends in data management – the emergence of big and fast data!

Bill Schmarzo

About Bill Schmarzo

Read More

Share this Story
Join the Conversation

Our Team becomes stronger with every person who adds to the conversation. So please join the conversation. Comment on our posts and share!

Leave a Reply

Your email address will not be published. Required fields are marked *

0 thoughts on “Using Big Data Analytics To Improve the User Experience

  1. Great framework for looking at how to convey key data to customers. Seems like the most important step is to go beyond the first try at the report, to looking at what customers will see in a number of different scenarios should you send that report to them. Thanks!

  2. Thanks Lee for the feedback. I think you got it spot on – which is to take the time to understand what the users are looking for when they view that report (i.e., how do they ascertain what’s good and what’s a concern) and what sorts of actions will they take based upon the data in the report.