Page 2 of 2

Three Ways to Gain Value from Analytics

Do analytics provide value? Absolutely! In my 2014 Best Practices Report on predictive analytics, fully 45 percent of the respondents had measured either top-line revenue impact or cost savings or both from their predictive analytics efforts. I am always interested in how organizations gainvalue from their analytics endeavors. I've even started tweeting daily about best practices in analytics (@fhalper). Here are three ways to gain value from analytics.

1. Tie analytics to a metric

A metric is a quantifiable measurement that relates to a business activity. When properly defined, metrics should relate to each other and provide a framework for analysis. Typically, organizations can track metrics, which is a good thing because metrics can also help drive transparency. At a recent TDWI Executive Summit in Boston, many of the case study and expert speakers highlighted the value of metrics when it comes to analytics. It was also highlighted in our conference workshop with Nauman Shiekh. In my predictive analytics study, many of the organizations that measured value in predictive analytics made a point of including measurements into their analytics process.

When starting out with analytics, it can make sense to begin by thinking about a metric that the business already measures -- such as sales volume, percent outages, or percent churn -- and then see if you can predict it. For example, if you typically measure percent downtime, take that metric and see if you can predict it. If you can, then clearly that is a value add because your organization can take action on it, whereas previously your analytics was simply reporting a value.

2. Operationalize your analytics

Operationalize means to make analytics part of a business process. When analytics are part of a process, it can help to drive action -- either manually or automatically.

There are a number of ways that analytics can be operationalized. You can score models and drive an action depending on the outcome. That action might be to route a potentially fraudulent call to a person at a call center for action or direct a customer at risk of churn to a call center agent. You can embed analytics into systems operations to make a recommendation to a website visitor about a product to buy. While running analytics against very fast data in motion is still in its infancy, it's another way to operationalize analytics.

Operationalization can happen at various time intervals, from daily to less than every second. You can gain a tremendous amount of value from operationalizing analytics in cost efficiencies, productivity gains, and in new revenue opportunities. TDWI is seeing an increase in operational analytics in our surveys and we expect the trend to continue.

3. Add in new kinds of data

Most users still employ structured data with some demographic data thrown in. TDWI research suggests that as users become more mature in analytics, they start to add disparate data to the mix. However, vendors are making it easier to perform analysis with different kinds of data. For instance, many analytics vendors are now providing the ability to add geolocation data (such as maps) to your analysis. In effect, you can layer various kinds of location data to your analysis. This often makes it easier to see patterns in the data. Other vendors are providing parsing technology or text analytics as part of their analytics package to extract useful pieces of information from text data that can be added to the analysis.

Different kinds of data can often provide new insights to analysis. In modeling, it can provide "lift" to the analysis. All of this can help to drive value, especially if action is taken on the results.

Learn more about the potential of BI, analytics and big data at the TDWI Executive Forum in San Diego, California on September 22-23, 2014.

About the Author

Fern Halper, Ph.D., is vice president and senior director of TDWI Research for advanced analytics. She is well known in the analytics community, having been published hundreds of times on data mining and information technology over the past 20 years. Halper is also co-author of several Dummies books on cloud computing and big data. She focuses on advanced analytics, including predictive analytics, text and social media analysis, machine-learning, AI, cognitive computing and big data analytics approaches. She has been a partner at industry analyst firm Hurwitz & Associates and a lead data analyst for Bell Labs. Her Ph.D. is from Texas A&M University. You can reach her by email (, on Twitter (, and on LinkedIn (

TDWI Membership

TDWI Members Get FREE anytime access to all reports and publications

Individual, Student, & Team memberships available.