Application Modeling by Gsk Solutions Inc using Splunk Infrastructure
The retail industry is undergoing a profound transformation. For retailers, information technology has become a key source of competitive advantage to succeed in today’s markets. As part of this change, significant new developments in information technology have come about. These developments are driven by an increased focus on enabling mobile channels, embracing new payment processing technologies and using growing volumes of data to deliver a better customer experience.
The payment devices, mobile point of sale (POS) systems, mobile apps, online stores, applications, networks and databases that comprise a typical retailer’s online and store IT infrastructure generate a tremendous volume of data in the form of POS logs, payment systems and mobile device logs, application logs, server logs, syslog, message queues and clickstream data. Such data, also known as machine-generated data, is a critical source of value to retailers in providing new insights to IT, operations and the business. With increasing reliance on real-time visibility across a multitude of cross platform solutions, today’s retail technology teams face a formidable challenge in managing the big data generated by disparate systems. Unfortunately, the presence of obsolete technology infrastructures often prevents retailers from taking full competitive advantage of the wealth of information present in machine data. Oftentimes, the data is simply discarded to minimize “noise.”
Insight into machine data from payment systems can address the following needs:
Real-time analytics for loss prevention
Real-time analytics for loss prevention Studies have shown that decade over decade, retailers lose about 1.7% of sales to inventory shrinkage. For years, point of sale companies and vendors in the retail technology integration business have advocated the need for expensive, purpose-built loss prevention systems. These stand-alone loss prevention (LP) systems have significant drawbacks. Firstly, in almost all cases, these systems run in batch mode and are not available to leverage analytics in real time. Most systems are fed from post audited sales data and rely on another business group to close out the proceeding period to gain LP insight. Secondly, retailers are often bound to proprietary third party data parsing configurations. Introduction of new data elements, systems or applications require extensive re-work to ensure raw data elements are parsed correctly. This adds risk, can be disruptive to the loss prevention and operations team, and isn’t immediately available to the business. Machine data insights are well suited to enhance an existing LP solution for complete end-to-end LP functionality. With machine data, retailers can simultaneously accept real-time unedited or parsed sales data from multiple applications, eCommerce, and mobile devices to gain insight into meaningful metrics such as:
The Retailer Machine Data Opportunity
The Retailer Machine Data Opportunity As retailers consider the impact of big data on their organization, it is useful to explore opportunities in four key areas:
Payment processing availability
Payment device performance and availability is critical for any brick and mortar business. With the upcoming requirements for Europay, MasterCard and Visa (EMV) and other emerging payment technologies, the traditional support paradigm where IT has little to no visibility for all payment endpoints, payment transaction patterns and general processing effectiveness will no longer be acceptable. Downtime on these payment devices translates into loss of ability to upsell during the payment process and inability to capture payment at a specific lane—thereby decreasing total productivity significantly. With these trends, payment device uptime and availability will be even more critical moving forward. Organizations need a scalable solution that provides real-time insights into machine data generated by the payment-processing infrastructure.
Online and mobile store performance
For any retailer, the cost of online or mobile storefront downtime is huge and will only grow over time. Not only does it have a direct impact on revenues and profits, it can adversely affect the overall customer experience, leading to decreased loyalty.
While an online storefront may be very easy for a customer to navigate, the underlying applications, servers and networks are extremely complex. The IT infrastructure is often massive, distributed, virtualized and in the cloud. When a site is not performing well or is down, it is very difficult for IT and support personnel to understand the root cause of the problem because they do not have a consolidated view of machine data generated by all the discrete elements of their IT infrastructure. When machine data such as application logs, network logs, device logs and server logs can be viewed through a single pane of glass, retail IT can break down the departmental silos and correlate across data sources. By doing so, retailers can accelerate time to resolve issues, improve site uptime and deliver a superior customer experience. Examples of insights that machine data can provide include:
Efficient order management Ordering
Capabilities such as Amazon.com’s 1-Click have made order management extremely simple for end customers. However, the underlying order management process is very complex with many steps. A retailer’s IT infrastructure needs to support orders across multiple channels, devices and operating systems, all of which add multi-layered complexity to the ordering process. With such a sophisticated IT environment as the backdrop, it is often very difficult for retailers to accurately pinpoint when, where and why a customer order was not processed in a timely manner or why the order was lost. Here are some of the benefits that a retailer can realize from deploying a robust data management approach to analyzing machine-generated data across the end-to-end order management process:
Retailers are sitting on exponentially growing volumes of machine data—data that exists in the enterprise, can deliver tremendous value and yet is significantly underutilized. In this whitepaper, we have just scratched the surface with a small sampling of use cases where machine data insights help retailers. Other use cases include PCI Compliance, Security, Customer Intelligence, Retail Operational Sales Data and Virtual Infrastructure Management.