The age of digital disruption has arguably impacted retailers more than any other industry, with well-known brands going out of business or closing stores on a regular basis and consumers increasingly choosing online shopping over brick-and-mortar because of convenience. With pressure to stay competitive, when an outage of any kind (point-of-sale system, inventory management, etc.) happens at a retail store, the negative experience is sure to go viral. For instance, the recent outage experienced at Target certainly made things difficult and caused major headaches for customers and IT staff, not to mention the company’s reputation. E-commerce is just as challenging, if not more so, in terms of keeping customers happy, as demonstrated last year during the holidays when several major retailers faced technical difficulties.
To remain competitive, retailers must continually make enhancements to ensure the customer experience remains flawless and meets growing expectations. With digital innovation and the addition of new retail features, technology is more important than ever, but this comes with major challenges. Innovation creates a more complex technology architecture that doesn’t easily scale and cannot be fixed by simply throwing more manpower at the problem.
Traditionally, IT has worked in silos, which is problematic. Each component of the technology stack is split among teams (developers, network, security, and infrastructure), and business leaders are completely out of the loop and unaware of what’s happening holistically. With the complexity of adding new applications and capabilities, you can quickly see where processes can break down and failures are inevitable.
1. Automate problem resolution.
AIOps, or artificial intelligence for operations, is a new method retailers need to inject into their operations strategies. By definition from Gartner, “AIOps is the application of machine learning (ML) and data science to IT operations problems.” The goal is to knock down silos and build an environment capable of providing holistic monitoring across all aspects of the application flow, leveraging domain knowledge across the IT department. Additionally, enabling automation is critical to identifying and resolving issues in real time, as soon as they crop up and before the customer is impacted. The real benefit of AIOps and automation is the troubleshooting time it saves IT teams so they can stay focused on innovation.
2. Quickly identify the root cause of issues.
When you break down performance issues, the first hurdle is quickly determining the root cause analysis (RCA) to be able to resolve the issue at hand. A recent study of 6,000 IT leaders indicated that this is still greater than seven hours, costing enterprises approximately $400,000 in IT spend. This doesn’t include the potentially massive loss of revenue and damage to the brand. To drastically reduce this time, companies need to think about how they can consolidate their cross-environment monitoring events, logs, errors and anomalies into a centralized AI engine that can quickly correlate all those metrics, pinpoint issues quickly, and automate remediation. Once an organization is able to quickly determine RCA, it can automate remediation, reducing potential outages to mere minutes as opposed to hours, ultimately resolving problems before any customer (and the brand’s reputation) is impacted.
3. Free up time for innovation.
Post-mortem meetings are useful after any business performance problem, but they can consume many hours or even days for several key IT staff members as they dig in to identify what triggered the issue. This goal is often elusive due to the complexity of the environment. With an AIOps strategy and framework, these tasks are eliminated, with automated emails explaining exactly what happened and when, what was done to fix an issue, and how much business was impacted. This frees up even more time for innovation, reduces IT costs, and eliminates the risk of losing customers.
The bottom line is retailers without an AIOps strategy will have a tough uphill climb to stay ahead of their competition. They should understand the concept of AIOps, and from there determine what can be automated, focusing on the most mundane to the most mission critical. AIOps can be implemented and enhanced over time, but the time to start is now. Otherwise, retailers will be left behind with a platform ripe for outages, incurring damage to the bottom line and brand reputation.
Gregg Ostrowski is regional chief technology officer at AppDynamics, an application performance monitoring and management platform.
Related story: Retailers Believe in the Power of Artificial Intelligence
Gregg Ostrowski is regional chief technology officer at AppDynamics, an application performance monitoring and management platform.
Gregg is a senior executive and thought leader with over 20 years’ experience in leadership positions for companies like Research in Motion and Samsung. He has worked with many F1000 customers, Government Agencies and partners on digital transformation, mobility application deployments, DevOps strategies, Analytics and high ROI business solutions.