Video Quality Report

Overview

The Video Quality Report shows HD/SD video quality metrics for each of your products.

  • HD Markets:  The number of markets that are rated HD for the selected product.

  • HD Users: The percentage of users that are rated HD for the selected product. A user is HD if they are in an HD market.

  • Also displayed are values of each metric 30 days ago. No Change means that the change is less than 0.5%

Product Overview Section 

The RATED HD/SD/LD (weighted) columns represent how a product would be rated, if we weight locations by the number of streams coming from that location.

The RATED HD/SD/LD columns represent a simple fraction of how many locations have that particular quality rating.

You can switch between viewing cities, regions, or countries using the location selector above.

Previous Hop Overview Section 

This section shows playbacks and quality rating percentages broken down by the previous hop ASN in BGP before it reached your network. For example, a playback with an AS-path of 15169-X-Y-Z-<your- network> will have a previous hop of Z.

Unknown playbacks are playbacks where Google is unable to compute the previous hop ASN but have data about the playback.

Other playbacks are playbacks where Google doesn't have enough information to give useful information.

Global Quality Section 

This section shows a list of the locations that you’re serving and their respective quality rating.

You can switch between viewing cities or regions using the location selector to the left.

Hourly Breakdown Graph 

The hourly rating view is intended to show time of day trends for user capability (the highest bitrate users can stream without interruption). The ratings view shows observed throughput for the best 90% of streams. The view also shows layback percentage.

For each hour, we consider all playbacks that occurred during that hour of the day for the past 7 days. Playback percentage is computed as the percentage of views in the past 7 days that occurred in that hour.

A bar-chart shows the breakdown of each hour's playbacks by user capability level.

You can select a city, region, country, or ASN below to view the hourly trend for that location.

Prefix Breakdown Table 

The Prefix Breakdown table is similar to the Per-Prefix Report. 95th-percentile egress and 95th-percentile retransmits are calculated by grouping the last 7 days into 15 minute intervals and computing at the 95th-percentile group. Currently they include all traffic to the prefix, not just traffic geolocated to the selected location.

The HD playbacks column shows the percentage of HD playbacks going to this prefix within the selected location.

The playbacks column shows the percentage of playbacks going to this prefix within the selected location.

The HD playbacks / playbacks column is a simple fraction of the two columns. Values > 1 mean that a prefix is serving more HD prefixes relative to other prefixes in the selected location. Values < 1 mean that a prefix is serving fewer HD prefixes relative to other prefixes in the selected location. Values > 2 are highlighted in green. Values < 0.5 are highlighted in red. This allows you to quickly see areas that are having the greatest positive or negative impact on playback quality.

Only prefixes with a 95th% egress of >10Mb and an impact of >1% are shown.

You can click any of the table headings to sort the table.

You can select a city, region, or country to view the prefix breakdown for that location.

Location Selector

Click on any location row to update all visualizations to contain only data for users in the selected location.

With the location dropdown, you can view data at the city, region, or country level.

The Quality column shows the current rating of each location.

The Playbacks column shows the percentage of all traffic going to this location.

You can click the table headings to sort the table.

Was this helpful?

How can we improve it?
Search
Clear search
Close search
Google apps
Main menu
2134145379693866507
true
Search Help Center
true
true
true
false
false
false
false