Streaming Analytics Demo: MuxData
Learn more about streaming analytics at Streaming Media's next event.
Watch the complete presentation from Streaming Media East, VES203. Accessing Real-Time Metrics for Live Streams at Scale, in the Streaming Media Conference Video Portal.
Read the complete transcript of this clip:
Robert Reinhardt: One of the reasons I like Mux is because if you want quick-and-dirty analytics where you don't have to spend a whole lot of time designing your whole experience around getting this data out, you can use something like Mux. So, Mux has--I'm already logged in--Mux gives you a lot of information up front. More importantly--and I'm not gonna--this isn't a whole tour of Mux, but you can define workspaces essentially.
Right now, I have a development workspace that I called Streaming Media for this demo. It does show me two concurrent users right now, too, but it gives me a lot more video-centric information than Google's gonna tell me unless I start formatting custom reports and know how to do that in Google, right? So, it's gonna give me, right now I don't have enough users to give me an accurate rebuffering percentage, but like if I'm even with just a few hundred people, I can find out really quickly just by coming into this control panel 10 minutes into a broadcast and see what this rebuffering percentage is. And if it's high, that means that a lot of people are having problems watching that stream potentially if there's a rebuffering state happening. And rebuffering, of course, means you've buffered but then you've gone back to having to rebuffer it again. And it's showing me that, you know, I gave it a label.
And I'm gonna just show you really quickly how easy it is to--again, most of these services wanna make it easy for you to put their code into a player. They've got several integration libraries. And, one of them is a JW Player integration. So, again, I have a JW Player library. This could be video.js, it could be any number, remember they invented video.js encoder guy, so of course they're gonna have a video.js tie-in to this. They have a JW Player Mux library that I have to make sure is available on the page with the player.
And more importantly, after I initiate JW Player, and again, this is all code that they give you samples of, they want to establish the page-load start time. So there's a player_init_time whenever you initialize the player, since I initialize it as soon as the page loads, I'm just using the current date, timestamp. And these are all the properties that I can specify and pass into the events that Mux is capturing. So, I can go through, if users were logging in, I could pass it a user ID.
And again, this could all, in a production scenario, I wouldn't expose all this information like on the main page in a view source like this. It would be hidden in its own call, and we could pass all these attributes dynamically from the database into the page, so that all this looks like just a bunch of variables and not hard-coded values like they are now.
But again, for most of the events that I'm doing, they're one-off events, they don't run in a series or anything like that, where there's even a centralized management system for the events themselves. But you could easily put a lot of dynamic information into this. The biggest important thing with most of these vendors is having a key that identifies you and your account. Just like you have a Google Analytics ID. I skipped that, but I presume most of you are familiar with Google Analytics. I'm gonna go back there real quick. That's the only thing I had to change in this script, is my ID for that site. So for my VideoRx Google Analytics sub account, that's the tracking ID for it, so that's the only value I needed to know. Same thing over here with Mux. I need some kind of ID that recognizes this so I know which area to go into my Mux account.
And again, it's sort of an all-in-one solution. I haven't really played around with exporting data from Mux. I'm gonna just show you quickly that this is just the real-time dashboard. I think that costs a little more to get real-time. But there's other viewer metrics that you can get access to. They have this Overall Viewer Experience score, a 100 being great. So, so far it looks pretty good at 89%, Playback Success is at 97, Startup Time, these are all based on zero to 100. The closer we are to a 100, the better.
So they try to make visualizing things easy for, again, stakeholders. Right, I can get total video views. They have definitions of everything that's presented on this page. So if you brand new you can go in and just click these categories--see, oh look at that. We've got--you can see these experience scores. Like we were talking about quality of experience is broken down by browser. So if I started to see--oh it looks like Chrome had the worst experience, but everything else was pretty good.
And you can sort of go back and say, "Well, do we need to fix something for Chrome? Like, what's throwing an error there?" Maybe it's JW Player, who knows? But you can go down, and, again, I don't wanna spend too much time drilling down, but, a lot of information. I should also mention that all these vendors want your business so it's super easy, usually, to get trial accounts at least for two weeks. And, if you're serious about it, they'll extend it for probably whatever time period you need for your stakeholders to make a decision about it.
Bitmovin's Steve Geiger offers essential tips on which streaming/encoding analytics to track, how to read heatmaps, and how to use analytics to deliver better video to your audience in this clip from his presentation at Streaming Media West 2019.
VideoRx CTO Robert Reinhardt demos Datazoom analytics in his Video Engineering Summit presentation at Streaming Media East 2019.
VideoRx CTO Robert Reinhardt demos SnowPlow Analytics in his Video Engineering Summit presentation at Streaming Media East 2019.
VideoRx CTO Robert Reinhardt identifying the key streaming data points stakeholders care about in this clip from his Video Engineering Summit presentation at Streaming Media East 2019.
GigCasters' Casey Charvet and Cheesehead TV's Aaron Nagler discuss how to contextualize, interpret, and leverage social streaming metrics in this clip from their Live Streaming Summit panel at Streaming Media East 2019.
SSIMWAVE Chief Science Officer Zhou Wang discusses the shortcomings of traditional QoE metrics in this clip from his Video Engineering Summit presentation at Streaming Media West 2019.
Companies and Suppliers Mentioned