Cloud computing promises to enhance business flexibility, efficiency, scalability, and reliability. According to a recent survey from O’Reilly, cloud adoption is steadily rising across industries, with 90% of organizations using cloud computing. Not only is cloud adoption growing, but enterprises are approaching cloud migration aggressively. Often, it is incorrectly assumed that services rendered by cloud either as platform or infrastructure will itself deliver the business outcomes of improved availability, business performance, security, and efficiency provided the application delivers the functionality. This is far from true. Cloud applications must be optimally architected, tested and operated for realizing the intended business outcomes. The considerations, choices and decisions taken in each of these lifecycle phases impact technical performance and as a result business outcome. Architecture considerations include decisions on choice of cloud services to be employed, design principles to apply, configuration parameters to tune, etc. Testing considerations include simulation of load and fault conditions that closely mirror runtime conditions and mining behavior for anomalies and faults. Operations considerations include monitoring critical parameters, identifying imminent failures (a.k.a incidents), taking preventive decisions, detecting failures, and taking recovery decisions etc. The challenge for businesses is that the cloud application runtime behavior consequences of these decisions are not easily envisaged by architects, developers, testers, and operators leading to sub-optimal business outcomes, reliability, and security issues. In addition, the decisions taken by architects, testers and operators are often inconsistent and incompatible with each other further aggravating the problem. In this Industry Track Paper (single pager) we intend to present these challenges and call for industry-academic collaboration to explore mining and modeling approaches to address the challenges