by Jeffrey Walker
| Feb 20, 2015
Data, data, and more data. It’s all we hear about anymore. Everywhere you turn today businesses are talking about their challenges with how to collect, prepare, analyze, and visualize various types of complex data. By the way you hear them talk, you’d think that data is gold. Well, actually, data is a major asset, and these streams of information are easily equated with big ROI opportunities if they can just lead to more customers.
Most companies focus on the “front-end” data, or the kind of data that will influence customer actions and behaviors in relation to their brands, products, and services. But there’s another side to the data game that you really need to pay attention to if you want to maximize your business efficiency and revenue, and it’s not front-facing at all. This concerns the huge amount of information that gets generated behind the scenes, especially in the form of message queues, sensor data, GPS data, and IT log data (application logs, point of sale logs, server logs, virtual machine logs, web proxy logs, ect.).
This vast storehouse of information is referred to as “machine data” and it’s increasingly becoming valued as a source of key business insights. The rapid expansion of Internet of Things and M2M (machine to machine) communications are all fueled by this kind of data as well, and businesses are hard-pressed today to find ways to capture and utilize this treasure trove of hidden insights. In fact, IDC forecasts that machine-generated data will increase to 42% of all data by 2020, up from 11% in 2005.
How well is your organization leveraging IT/machine-generated data? Let’s explore a few examples of how you can turn your hidden data into valuable insights that can generate new forms of ROI for your organization.
1. Clickstream Data
A clickstream data is the record of the screens that a visitor to your website clicks on while visiting the site. As the user clicks anywhere in the webpage or application the data gets stored. Clickstream data can be found within your web server, routers, proxy servers, and ad servers. If captured and used, this kind of data can provide valuable insights about customer behaviors and customer experience issues, can help drive real-time ad placement, and can also be used to detect shopping cart abandonments.
2. Sensor Data
The drop in cost of sensors, RFID chips, and the ubiquity of 3G and 4G wireless technology has led to a dramatic increase in the volume of sensor data, and made it much more efficient for businesses to start experimenting with the kind of operational data that these systems can produce. Examples include fleet management, product monitoring, smart meters and buildings, car informatics for usage-based insurance, and the list goes on.
3. Call Records
Call Detail Records (CDRs) is data that is specific to a particular telecommunications transaction – but not the content of the transaction. It can provide useful insights into the volume and nature of calls and other meta-data such as the phone numbers of both the calling and receiving parties, the start time, and duration of that call. This information can be captured and used by organizations to determine service quality and customer experience issues, and is particularly helpful when joined in real-time with GPS and location data.
4. IP Router Syslog
Your IP network equipment uses syslog formation to capture connection status, capacity information, routing information, failure alerts, security alerts and performance data from your infrastructure and website. When this information is joined and queried and processed in real-time it can provide unique insights about your network, and also lead to predictive insights that will help forecaste uptime, downtime, and improve overall performance.
5. Application Logs
Many application servers generate log files in such areas as local files, log4j, log4net, Weblogic, WebSphere, JBoss, .NET, and PHP. All of this IT data provides critical insights into application and application server operation and performance, but also can lead to information on user activity and provide fraud detection.
The main takeaway here is this: with the massive amounts of machine-generated or IT data available today, businesses have almost a limitless number of opportunities to improve web and infrastructure operations and customer experience. The growing complexity of physical, virtual, and cloud environments today, along with the increasing amount of data coming online with Internet of Things and M2M communications, will mean enormous ROI for those businesses that can leverage the data efficiently. One place you might want to start is with Splunk. In recent years this company has been redefining the whole value proposition of machine data by showing how it can lead to increased performance and revenue using what it calls operational intelligence. If you haven’t done so, now is the perfect time to expand your Big Data 2015 strategy to include IT/machine-generated data.
Post Tagged with
