4 Data Analytics Success Stories

by Drew Robb
4 Data Analytics Success Stories

Yes, there is lots of hype about data analytics. But there are also some compelling success stories, involving areas like predictive analytics.

There is a lot of hype out there about Big Data's ability to transform your business and how the latest analytics can revolutionize every aspect of human existence. But does the hype match up to reality? Enterprise Apps Today looked around for some successful analytics use cases and found four interesting ones.

Analytics? It's Electric

Who would have thought that theft of electricity would be a big issue? Yet Elizabeth Fletcher, deputy director of the Smart Metering and Infrastructure Program at Canadian utility BC Hydro, tells of losses of $100 million in one year due to pilfering, much of it from marijuana growers who need constant light for their crops. On a small scale, they steal it by bypassing the meter at a household level. The more enterprising do it at the transformer, substation or even directly from high voltage lines.

BC Hydro already employed field investigators to find wrongdoers, but it realized that feet on the ground weren’t enough. They needed to more efficiently zero in on areas of potential theft.

To combat the issue, the company rolled out 1.9 million smart meters at houses and businesses. The meter data is fed into a Pivotal Greenplum database and an EMC data lake where it is analyzed. The company ties this data into SAP systems and its own proprietary systems which have geospatial features to detect discrepancies in voltage patterns.

"These discrepancies are sometimes a data issue, but often help our investigators identify theft," said Fletcher. "Analytics is helping us achieve our goal of a 75 percent reduction in electrical theft."

The company is now adding in SAS for predictive analytics in order to conduct what is known as energy balancing. This entails being able to measure exactly the amount of energy coming into the grid and reconcile it against the amount consumed. The company has elaborate methods of estimating this metric, but it is now assembling an exact way to calculate it and then analyze it. SAS will help the company view this information across the entire power grid, then drill down into regions and districts too.

Analytics and Wearable Tech

Geneia is a health care technology and consulting company working with clinical information and medical records to help health care organizations deliver better care at a lower cost. It wants to improve decision making, both by health care providers and patients, so that the right actions are taken at the right time and unnecessary actions are eliminated.

The company is building a platform known as Theon to make this possible. While a huge amount of medical information exists in hospital systems, it is spread across disparate databases and repositories. While consolidating that information can help, Geneia wants to go beyond simple data consolidation. So it is incorporating wearable devices to do such things as take vitals and record EKG readings.

Having a flood of valuable data increases the likelihood of correct decision making. The challenge now is to narrow it all down in enough time to make a difference. Geneia is adding SAS analytics to analyze medical information in real time and predict when intervention is necessary.

For example, health care providers want to know how likely a patient is to follow treatment or develop complications. The Geneia Theon platform sits on top of a mountain of electronic medical record files. It is able to sift through all that data, estimate likely outcomes and suggest the best courses of action.

One health care provider utilizing Theon used this approach to develop a list of recently discharged people who could benefit from more thorough home monitoring to avoid emergency room visits or readmission. Analytics also spotted high risk areas, such as people who had run up large medical bills yet had not been seen by a physician in over a year.

Wearable devices add to the scope of Big Data analytics. With patients wearing them, data flows to Theon in constant streams. SAS has to analyze this, perform risk analytics and add value by focusing attention on where it needs to be.

"Wearable technology means a lot of data comes in, much of which is noise as well as a lot of false positives," said Avishek Mukherjee, CTO of Geneia. "Adding predictive real-time analytics helps us to improve the effectiveness of patient monitoring."

Analytics and the Supply Chain

The supply chain is an obvious place to harness the power of analytics. Chainalytics, a supply chain consulting and analytics firm, provides transportation market research to more than 140 organizations worldwide. It maintains data on more than $25 billion worth of transportation costs, as well as millions of transactions.

It uses Tableau Online to share information faster and provide self-service access to cloud-based visualizations used in decision making. This makes a big difference to buyers of transportation services, who spend anywhere from $5 million to over $1 billion per year for the intelligence. By communicating complex market pricing and research that explains where members stand against their peers, how transportation rates are trending and how their partners are supporting their cost and service goals through policy decisions, Chainalytics aims to help its clients cut transportation costs.

Matt Harding, vice president of Freight Market Intelligence Consortiums at Chainalytics, said spreadsheets and databases could not adequately support custom insights. Such reports only answered common questions and didn’t go deep enough With Tableau Online, the company added a standardized front-end on customer specific data that allows rapid and free-flowing navigation while providing a visualization experience not available from spreadsheets.

"Our customers can assess millions of shipment records within minutes or hours, compared to weeks before, and without any IT support or training," said Harding. "They can quickly determine evolving business strategies with their partners, communicate to the C-level and change the way people look at the business of transportation and the supply chain."

Competitive Intelligence Is No Game

Analytics is all too often given a rather vague problem of searching through a vast trove of data to find the elusive needle in the haystack. Quite often, there isn't even much idea what that needle will be or what value the insight will yield. But the area of competitive intelligence is quite different. Both the goal and the benefits are obvious: What are your competitors doing, how do you compare to them, and what should you be doing about it in order to gain market share?

That is what a tool named Periscope by a company named wefi aims to do. This expansion of its existing mobile data analytics platform provides insights such as how much time users of game apps spend playing a particular game per day. The game provider can also detect patterns in usage trends – times of day, periods of the week or seasonally important factors. This analytics service can also rank them against various benchmarks to see how they are performing.

Armed with this data, game application developers can make better strategic decisions toward capturing greater market share. It could also be used by travel app developers, for example, to divine insight on their sector. As the service is still in beta, no users were able to comment.

But the point is to take millions or even billons of data points each day, combine that with the combined repository of application usage from millions of devices, add in geo-context and then sift it all down to what a company needs to stay ahead in frenetic world of application development. It can fathom the degree of user engagement and overall reach, trend that data over time, and allow marketing to devise location-based campaigns to target areas where the company is strong and wants consolidate its position or weak and wishes to grow market share.

"Our data collection engine runs on mobile devices and collects anonymous information on user experience and behavior," said Alexander Zaidelson, wefi's vice president of Product Management. "The engine collects over 3,000 data points per user per day on all applications running on a device, allowing various kinds of analysis, including competitive analysis."

Drew Robb is a freelance writer specializing in technology and engineering. Currently living in Florida, he is originally from Scotland, where he received a degree in geology and geography from the University of Strathclyde. He is the author of Server Disk Management in a Windows Environment (CRC Press).

  This article was originally published on Thursday May 7th 2015
Mobile Site | Full Site