Technology company turns big data into insight Statistica customers using SupportAssist benefit from a world-class analytics platform that consolidates, manipulates and analyzes streams of data to fix systems before they fail. Case Study Company profile Company Dell Inc. Industry IT Country United States Employees 111,000 Business need To deliver a superior customer experience, Statistica s support team needed a way to proactively predict and address issues such as impending hardware failures before they negatively impact customers. Solution Statistica deployed the advanced analytics platform, Statistica, along with SupportAssist, to create, test and deploy effective predictive analytic models in order to predict and solve customers issues before they can impact the business. Benefits Delivers predictive analytics that enable issues to be addressed before they cause problems for customers Allows customers to spend up to 84 percent less time on the phone with tech support* Automates support case creation and issue resolution, resulting in up to 58 percent fewer steps in the support process* Enables the integration and governance of open source R into production systems *Based on Nov. 2014 Principled Technologies Test Report commissioned by Dell. Actual results will vary. Full report can be found at http://facts. pt/14dh1gl. SupportAssist not available on Venue 7 and 8 tablets.
With IT systems at the core of our lives and businesses today, enterprises and individuals alike cannot afford to be surprised by hardware failures. That s why Dell developed SupportAssist, a remote monitoring tool that provides automated issue detection, support case creation and dispatch of in-warranty parts. Whether you need automated monitoring on your home PC or need to monitor an entire server farm, SupportAssist reduces troubleshooting effort, accelerates resolution and ensures maximum uptime. Behind the scenes, SupportAssist has leveraged Statistica s advanced analytics platform, Statistica, to model predictive rules. Aggregating data from multiple sources enables better support For many technology vendors, customer support is limited to reaction responding to calls or emails reporting a problem that is already causing issues for the customer. For Dell, that is not good enough. That s why Dell created SupportAssist, a tool that delivers predictive issue prevention. Statistica enables us to create models from billions of data points, discover key relationships and extract predictive insights. - Michael Shepherd, Senior Strategist, Dell Inc. 2
Dell s global support and deployment services performance analytics group was charged with creating and maintaining Support- Assist. They knew that the challenge was not collecting data. For customers who opt in, the solution collects nonsensitive machine state data such as temperature, configuration data, version information, driver inventories and much more pouring over 100 million messages per week into the team s hands. Rather, the challenge was making that data useful to Dell customers. After all, many organizations today, especially larger ones, already have tools monitoring their data centers. To add value, the Dell support team needed to blend and contextualize data from multiple sources to discover emerging issues that might affect many customers. A customer may have several hundred servers and a tool that monitors them and issues alerts, explains Michael Shepherd, senior strategist at Dell. However, with SupportAssist, Dell not only sees those alerts but weighs the data against thousands of similar systems and supply chain test data collected from manufacturers, repair centers and so on. The single lens used by the SupportAssist Intelligence Engine provides us with a broad set of information that we can use to set alerts and proactively address issues before they impact our customers. Statistica automates the process from end to end, from data consolidation to data manipulation to modeling, so we can work more efficiently. - Michael Shepherd, Senior Strategist, Dell Inc. 3
Big data requires a powerful, scalable analytics platform The team needed an analytics solution that could help them aggregate and analyze that data to improve customers lives. The team began working with their legacy analytics platform, but quickly discovered that these solutions were not as scalable, required complex coding, and were not as open, flexible and user friendly compared to other platforms on the market. This limited the number of individuals who could create advanced analytic workflows within the organization and increased the complexity of automating critical tasks. Then, Dell s selection of Statistica brought the world-class advanced analytics platform to the team s doorstep. With Apache Kafka, Apache Solr, open source R and Statistica, the team is now able to create, deploy, test and rework models using the massive amounts of data streaming in from SupportAssist customers in order to predict and prevent issues. For example, using Statistica, the team was able to automate the comparison process between fault models for hard drives and, thus, generate rules to predict impending drive failures. As a result, Dell customers are able to replace failing drives before there is a loss of data and downtime. The research team used Statistica to automate the entire process of consolidating historical RAID controller logs, sampling, cleaning, modeling and, ultimately, reporting of results that could accurately predict when a hard drive is about to fail, says Shepherd. The models were validated by comparing the results of each prediction against the drives that actually required replacement. In other words, Statistica enabled the team to take a labor intensive research process and automate the discovery of key relationships and extract predictive insights. 4
Solving the problem, whatever its root cause Of course, the SupportAssist Intelligence Engine is used for more than predicting hardware failures. With this big data solution, the team is able to proactively identify anomalies, drill into details and pinpoint the root cause. Whether it is a hardware issue, a software issue, a configuration problem or a driver issue, the solution significantly reduces phone support by more than 80 percent while cutting the number of steps in the support process in half. In cases where the issue is with a Dell product, the issue can be quickly remediated through full automation and proactive contact with the customer. If this issue is with a component provided by a Dell supplier, the team is able to quickly share data and findings with the supplier to promptly resolve the issue and reduce the impact on the rest of the customer base. Predicting and preventing downtime and systems failures Helping customers predict and solve issues before they have the potential to harm their business is the primary goal of SupportAssist s predictive analytics, but the support team has also experienced other benefits, including enhanced internal productivity by as much as 10 times. Before we had Statistica and our other big data applications in place, support was a time-intensive, manual effort. A customer would report a problem; we would open a support case and then begin digging through mountains of cryptic log files, notes Shepherd. Now, the system and analytic tools automatically run on the rules we developed, enabling us to locate potential issues with minimal human effort. Within a few minutes, we have end-to-end case creation, customer contact and dispatch completed. 5
ReduCIng time spent on support calls by up to 84 percent Statistica also makes model development and deployment easy. Using its simple graphical interface, the team can create efficient, reusable modules and build workspaces with custom analyses. Moreover, Statistica eliminates the need for multiple tools, which reduces both costs and complexity. One of the biggest problems in analytics is having to use multiple tools to work with the data: to consolidate it, manipulate it, cleanse it, sample it, run it through models and so on, says Shepherd. Statistica automates the process from end to end, from data consolidation to data manipulation to modeling, so we can work more efficiently. We can build and run more models to deliver better results for our customers. As a result, customers spend 84 percent less time on the phone for technical support. About StatistICA Statistica s advanced analytics, big data and IoT offerings provide you endless possibilities to innovate your enterprise. Whether it s uncovering the genetic basis of a disease, reducing hospital readmissions, mitigating financial risk, or ensuring procedural validation, Statistica enables organizations to transform in new and exciting ways. By embedding analytics everywhere and empowering a wider community of citizen data scientists, you ll accelerate innovation, improve customer experiences, and streamline your enterprise for the future. http://statistica.io View more case studies at statistica.io/resources Statistica and the Quest Software logos and products as identified in this document are registered trademarks of Quest Software, Inc. in the U.S.A. and/or other countries. Other trademarks and trade names may be used in this document to refer to either the entities claiming the marks and names or their products. Quest Software disclaims any proprietary interest in the marks and names of others. Availability and terms of Quest Software, Solutions and Services vary by region. This case study is for informational purposes only. Quest Software makes no warranties express or implied - in this case study. March 2017, Quest Software Inc. All Rights Reserved. CASESTUDy-STATISTICADellSUPPORTASSIST-US-AA-00000