How do you fully quantify every aspect of a unique viewer’s real-world viewing experience? While we have a multitude of quality of experience (QoE) metrics to quantify various aspects of experience, it is still a challenge to uncover niche quality issues or errors as they occur for even the smallest set of viewers.
Often, crucial quality issues go undetected because they nominally impact multiple KPIs. How do you know what devices are experiencing poor average bitrate, if you easily can’t identify which devices had plays less than 20 seconds, a high bitrate, and high rebuffering? The ability to drill down and filter across multiple QoE metrics allows for richer analysis and lets operations and engineering teams uncover granular quality issues they might have missed.
Coming Soon: Metric Filtering
Flexibility in filtering is crucial to effective QoE management, and that’s why we are soon releasing enhanced Metric Filtering capabilities to allow users to filter data down by any metric of interest. Users can drill into specific subsets of viewers and devices that match a set of parameters to spot quality issues that would otherwise go undetected if viewed in the context of an entire data set.
This feature update allows for more sophisticated analysis, ensuring your team can monitor challenges unique to diverse sets of viewers. While Conviva already offers multi-dimensional drill down analysis across limitless dimensions, Metric Filtering allows you to zoom into a specific subset of your viewers that are experiencing a change with a specific performance metric and identify how that impacted metric might cause additional deviations in quality.
It can also help to measure the magnitude of any given issue. If a specific device type exhibits more video start failures, your team’s response may be different if the issue impacted 1,000 viewers than if it impacted 1,000,000 viewers.
With millions of unique viewer sessions at any given time, executing this sort of granular analysis at scale would require massive time and resources. Plus, you would need to perform those cumbersome analyses again any time you want to compare new time frames. The in-depth, cross-metric analysis that can be achieved in a few steps unlocks an entirely new level of responsiveness, agility, and monitoring that was not previously possible.
Now, your team has the tools to act quickly and reduce the risk of lost viewership and revenue.
Uncover Quality Outliers In a Few Clicks
So how will metric filtering help your team uncover complex quality issues? It’s simple.
Now that your team can filter by multiple performance metrics, you can find trends, patterns, and anomalies more quickly and easily. Some quality errors might not impact a massive subset of users, and thus might go undetected if looking at your entire viewer base. With metric filtering you can uncover outliers who might be experiencing multiple issues, for example high rebuffering and low bitrate, by drilling into the bitrate of those viewers and then further filter into those with high rebuffering.
With metric filtering, your team has the power to make sense of increasingly complex sequences of events across limitless dimensions. You can quickly uncover even the tiniest subset of users experiencing issues to ensure that you are continuously improving performance for all users. These types of analyses can be time– consuming to execute manually, and thus having the ability to uncover these issues in just a few clicks ensures that your team can prioritize continuous optimization and deliver a quality viewer experience to your entire viewer base.