Understanding user experience is becoming critically important to the success of all companies. I\u2019ve interviewed dozens of business leaders on their digital transformation initiatives, and I can sort them into two larger buckets: increasing workforce productivity and improving customer experience.\u00a0\nThose may seem somewhat unrelated, other than they used digital technologies, but there is another point of commonality and it\u2019s that applications play a key role.\n\nBy 2020, customer experience will be the #1 brand differentiator, topping price, product, or any other metric you can think of. While a web or mobile app experience isn\u2019t the only thing that creates a good or bad experience, it\u2019s often the first touch point for customers \u2014 and a bad one could drive them away.\nAlso, businesses are becoming increasingly distributed, which is why there has been such a strong adoption of conferencing applications and video. These let people in different geographic locations collaborate as if they were in the same room.\nWe are clearly in the app economy now, and poor app performance means lost revenue. If you\u2019re looking for that to be quantified, I recently calculated the cost of downtime across a number of different verticals and found the average to be about $28,000 per minute. The lesson here? Downtime and poor app performance costs companies real money.\nDetermining the cause of system downtime, poor application performance\nA decade or so ago, the cause of downtime was pretty easy to find. Every company had its visual dashboard, and when things went from green to red, that indicated an outage. Today, the concept of an outage is much more nebulous.\nFirst, we built our infrastructure to be so redundant that someone could flick off a router and things would keep working. The bigger problem is when everything is green and stuff isn\u2019t working. Adding to the challenge is that apps have moved to the cloud, so the source of an outage can be anywhere between the cloud service and the user. It\u2019s enough to give even the most experienced network professional serious angst.\nWe are clearly in the app economy now, and poor app performance means lost revenue.\nThis week, ThousandEyes added a new capability to bring visibility to the end-to-end application path. The company is best known for its ability to look at performance across all internet paths and has now added a feature called Device Layer that provides the health status of network devices. This data provides network endpoint context to the ThousandEyes Path Visualization, enabling network operations teams to deliver better application and service performance.\nDespite the uninventive name, Device Layer is easy to deploy. It automatically discovers the network devices, such as routers, switches, firewalls and application delivery controllers, that are in the path of the business applications and services. The added data provides engineers with visibility into how each network endpoint impacts the delivery of services through an integrated view of application performance, network and routing behavior, as well as device health.\u00a0\nOnce the tool discovers the devices, it dynamically maps and visualizes the network topology of the enterprise WAN, including the internet, which has been a \u201cblack box\u201d to most network and application performance tools and continually updates as the environment changes.\u00a0\n ThousandEyes \nThousandEyes Device Layer dashboard displays network device metrics, such as\u00a0throughput, errors and discards.\n\nDevice Layer tracks metrics such as throughput, errors and discards to infer health. For example, if a WAN interface on a router started reporting a massive amount of errors, that wouldn\u2019t necessarily trigger an \u201coutage\u201d but it would degrade the performance of an application. The ThousandEyes dashboard would see this and indicate a connectivity problem, and the remediation process could begin.\nSolving application problems can be one of the most difficult things for network engineers. I\u2019ll correct that: Identifying the source of a problem is the most difficult, as my research has found that 90 percent of the time taken to solve problems is in the identification phase. By combining network device context into its visualization tool, ThousandEyes obviates the need to switch between consoles and manually correlate information, which can significantly reduce the time to resolve a problem.