Guide to Analytics

ACTIVITY DASHBOARD #

FILTERS #

Starting at the top of the Activity page is a line of filters allowing you to only see the data that is relevant for your query. Want to know about your activity in the last quarter? Click on Start Time to and change it to 1 Quarter. Curious about your Video on Demand events? Change Type to VoD.

Once you have made your selections, click the circular arrow in the upper right. This will update the dashboard to only show you the data that you want to see.

TILES #

The first stripe of tiles will show you the average Quality of Experience (QoE), Total Views, Average Peering Efficiency, and Bandwidth Savings. This is a great way to get a snapshot of how your organization’s events have been going.

• Avg QoE Score – On average, how was the QoE for your viewer? This metric ranges from Bad to Excellent and is calculated using buffering time and time to the first frame. The less time your viewer spends waiting for the content, the better the experience!
• Total Views – How many total viewers did you have for these events? This is a sum of views over all the events in the time frame specified by the filter settings.
• Avg Peering Efficiency – How did the ECDN perform? This is the grade that we like to give ourselves. Peering efficiency is the percentage of bytes delivered through peering when the size of the peering cluster is taken into account. Basically, in an ideal peering group, one viewer will pull content from the origin and everyone else in the group can peer off of that content. Peering efficiency is how well the ECDN performed when compared to the ideal peering group. A high percentage is a good grade, a low percentage means we should look at your network configuration so you can realize the highest possible bandwidth savings.
• Bandwidth Savings – This metric is the percentage of bytes you were able to save using peering during your events. If this is a low number, that likely means that many of your viewers were remote or peering disabled and therefore not able to peer.

TABLES #

There are two tables on this page as well: Live Right Now and Recent Activity.

• Live Right Now – If your company has an event that is happening at this moment, it will show up here. If you click on the title of the event this will allow you to drilldown into the Quick Stats dashboard or the Producer Dashboard, so you are able to see your event data in near real time.
• Recent Activity – Looking for an event that already happened? Find it in this table and click on the title to go to the Event Analytics and Network Analytics dashboards.

DOWNLOADS #

All the data from your events is yours to download. If you want to download the information on a tile, hover the curser over it and three dots will appear in the upper right corner. If you click on these dots, it will give you the option to download the data. Alternatively, you can click on the three dots in the upper right of this page and download all the data from the Activity dashboard.

Now that we understand the Activity dashboard, let’s go over all the details to be found in our Viewership Analytics and Network Analytics dashboards.

VIEWERSHIP ANALYTICS #

VIEWS & EXPERIENCE – HOW WAS YOUR EVENT CONSUMED? #

To start off, our analytics will give you a high-level view of your event with important metrics and key performance indicators. This is where to find the answers to questions such as:

1. How many people consumed your content?
2. How long did people watch?
3. How was the viewer’s experience?

Here is how we break down this information:

• QoE Score – The ECDN’s analytics tells you not only your average QoE score but also breaks down the percentage of viewers who experienced each type of QoE from Bad to Excellent in the donut visualization.
• Total Reach – Want to know how many people from your organization consumed the content? We report the Total Reach which is the total number of users who viewed the content.
• Avg View Duration – This metric reports how long the average users spent viewing your content. Ideally you want this number to be as close to the duration of the event as possible.
• Event Reach – This handy visualization allows you to easily see the maximum reach over time, so you know how many viewers you had for each minute of the event. On this chart, we also point out the Max Reach which is the maximum number of viewers you had at one time. (It’s important to note that this visualization is primarily for live events. Data shown on this chart encompasses up to 48 hours from the start of an event. This visualization is available in the dashboard for 3 months.)

DELIVERIES & CONSUMPTION – WHERE AND HOW WAS YOUR EVENT CONSUMED? #

Now that you have a high-level overview of how your event went, the analytics dashboards provide more information on where viewers were located as well as OS and browser distinctions.

• Delivery Map – This will show you where your viewers were located geographically. It’s the easiest way to get a snapshot of where the content was seen.
• View Tiles – The four tiles to the right of the map will further break down your viewership stats. In these tiles you can see Views by External IP, Views by Locality, Views by City, and Views by Country. The two tiles below the map break down the Views by OS and the Views by Browser.

USER PERFORMANCE #

Here you will find important metrics of how the event went, broken down by the individual user. You can find IP addresses, platform information and viewer experience stats on the micro level for each one of your users in a concise table.

NETWORK ANALYTICS #

Some of the information in the Network Analytics dashboard overlaps with the data from Viewership Analytics. However, in this dashboard we breakdown more of the network and peering data. If you want to get to the Network Analytics dashboard from Viewership Analytics, just click on the event title and you can toggle back and forth.

NETWORK PERFORMANCE – HOW MUCH BANDWIDTH WAS SAVED? #

At this point, the ECDN’s analytics will help you get into the nitty gritty details of how your network performed during the event. In this section, you can find out how much bandwidth you were able to save by using peering technology, as well as areas of improvement to capture even more savings!

• Bandwidth Savings – This metric is the percentage of bytes you were able to save using peering during the event. First, we remove the peering disabled users from the savings calculation and only include viewers who were capable of peering. Next, we take the bytes saved and divide it by the total number of bytes for these viewers, which is the percentage of bandwidth savings.
• Peering Efficiency – Peering efficiency is the percentage of bytes delivered through peering when the size of the peering cluster is taken into account. Basically, in an ideal peering group, one viewer will pull content from the origin and everyone else in the group can peer off of that content. Peering efficiency is how well the ECDN did when compared to the ideal peering group. A high percentage is a good grade, a low percentage means we should look at your network configuration so you can realize the highest possible bandwidth savings.
• Non-Peering Users – Is there room for improvement? In this metric, we show you what percentage of your users who were capable of peering were alone in their cluster. You want this number to be as close to zero as possible. If this is a large percentage, the likely culprit is that there are many VPN/remote users that have not been identified as such. Moving these users to the peering disabled category should drive this percentage down to zero.
• Bandwidth Over Time – This visualization shows you the Megabytes per second over the course of the event. The rate is further broken down between bytes that were delivered to peering disabled users, bytes pulled from the origin (not including peering disabled), bytes pulled from EdgeCache and bytes peered. (It’s important to note that this visualization is primarily for live events. Data shown on this chart encompasses up to 48 hours from the start of an event. This visualization is available in the dashboard for 3 months.)
• GB Delivered – The bar chart is a handy way to see what number of gigabytes were delivered from the origin (not including peering disabled), delivered from EdgeCache, peered, and those that were specifically delivered to peering disabled users.

DELIVERY PERFORMANCE #

While this stripe is very similar to the one in Viewership Analytics, here we color the location bars by the amount of content consumed from the origin and the amount that was peered. The dots on these tiles represent peering efficiency. This is a great way to understand your content delivery around the globe.

CLUSTER PERFORMANCE #

If you are curious about the peering groups in your event, here we breakdown the size of the clusters, the savings and peering efficiency for each cluster, and much more in the Cluster Details table!

One of our exciting features that you can find from the Cluster Details table is the Cluster Network. If you click on a specific Cluster ID in the table, you have the option to view this graph where you can see the peering in that cluster by node. Each node is colored by Quality of Experience and if you click on the node, you can see details for that user such as IP address and bytes. Click and drag a node to redistribute as you see fit!

QUICK STATS #

Quick Stats is an excellent way to see how your live event is going in real-time. This is a way for you to get a snapshot of important metrics while the event is in progress. This dashboard refreshes every two minutes to automatically keep you up-to-date!

The high-level statistics are shown at the top in Views & Experience. As you scroll down, you’ll find more details on viewership and network performance.

VIEWS & EXPERIENCE – HOW WAS YOUR EVENT CONSUMED? #

Here is how we give an overview of the viewership data:

• QoE Score – Kollective Analytics tells you not only your average QoE score but also breaks down the percentage of viewers who experienced each type of QoE from Bad to Excellent in the donut visualization.
• Total Reach – Want to know how many people from your organization consumed the content? We report the Total Reach which is the total number of users who viewed the content.
• Avg View Duration – This metric reports how long the average users spent viewing your content. Ideally you want this number to be as close to the duration of the event as possible.
• Event Reach – This handy visualization allows you to easily see the maximum reach over time, so you know how many viewers you had for each minute of the event. On this chart, we also point out the Max Reach which is the maximum number of viewers you had at one time. (It’s important to note that this visualization is primarily for live events. Data shown on this chart encompasses up to 48 hours from the start of an event. This visualization is available in the dashboard for 3 months.)

From that high-level viewership information, we then break down the location information and how the video is being viewed in Delivery & Consumption.

DELIVERIES & CONSUMPTION – WHERE AND HOW WAS YOUR EVENT CONSUMED? #

All of this viewership information should give you a great idea how and where your content was consumed.

• Delivery Map – This will show you where your viewers were located geographically. It’s the easiest way to get a snapshot of where the content was seen.
• View Tiles – The four tiles to the right of the map will further break down your viewership stats. In these tiles you can see Views by External IP, Views by Locality, Views by City, and Views by Country. The two tiles below the map break down the Views by OS and the Views by Browser.

You probably also want to know how the network is performing and what kind of savings Kollective is providing. For that we have the Network Performance stripe so you can get great information about the peering clusters and bandwidth savings.

NETWORK PERFORMANCE – HOW MUCH BANDWIDTH WAS SAVED? #

This section is where you can find out how much bandwidth you were able to save by using Kollective’s peering technology, as well as areas of improvement to capture even more savings!

• Bandwidth Savings – This metric is the percentage of bytes you were able to save using peering during the event. First, we remove the peering disabled users from the savings calculation and only include viewers who were capable of peering. Next, we take the bytes saved and divide it by the total number of bytes for these viewers, which is the percentage of bandwidth savings.
• Peering Efficiency – Peering efficiency is the percentage of bytes delivered through peering when the size of the peering cluster is taken into account. Basically, in an ideal peering group, one viewer will pull content from the origin and everyone else in the group can peer off of that content. Peering efficiency is how well Kollective did when compared to the ideal peering group. A high percentage is a good grade, a low percentage means we should look at your network configuration so you can realize the highest possible bandwidth savings.
• Non-Peering Users – Is there room for improvement? In this metric, we show you what percentage of your users who were capable of peering were alone in their cluster. You want this number to be as close to zero as possible. If this is a large percentage, the likely culprit is that there are many VPN/remote users that have not been identified as such. Moving these users to the peering disabled category should drive this percentage down to zero.
• Bandwidth Over Time – This visualization shows you the Megabytes per second over the course of the event. The rate is further broken down between bytes that were delivered to peering disabled users, bytes pulled from the origin (not including peering disabled), bytes pulled from EdgeCache and bytes peered. (It’s important to note that this visualization is primarily for live events. Data shown on this chart encompasses up to 48 hours from the start of an event. This visualization is available in the dashboard for 3 months.)
• GB Delivered – The bar chart is a handy way to see what number of gigabytes were delivered from the origin (not including peering disabled), delivered from EdgeCache, peered, and those that were specifically delivered to peering disabled users.

USER PERFORMANCE #

Here you will find important metrics of how the event is going, broken down by the individual user. You can find IP addresses, platform information and viewer experience stats on the micro level for each one of your users in a concise table.

PRODUCER DASHBOARD #

In addition to Quick Stats, we also have the Producer Dashboard for live events. This is a bare-bones dashboard that is designed to answer three questions for you:

1. Is my event running?
2. Is there peering happening in my event?
3. What’s the current and total reach?

The answers to these crucial questions can be read at a glance from this quickly loading dashboard.

QUESTIONS? #

If you have any remaining questions, please reach out to your Customer Success Manager. We’re happy to help!