Understanding Cloud Media Workloads: Content Creation, Media Supply Chain & Content Distribution

The HPA Tech Retreat is the pre-eminent gathering of industry-leading thinkers and innovation-focused companies engaged in the creation, management, and dissemination of content. At the HPA Tech Retreat, the most compelling topics facing the media and entertainment landscape today are presented, explored and debated.

The impact of the Cloud on Media & Entertainment is profound. Gain a holistic executive viewpoint of key insights into the M&E industry, end-to-end critical considerations and risks across the three primary cloud media workloads (Content Creation, Media Supply Chain, and Content Distribution), and how the rapid pace of change and innovations in technology are impacting and evolving solutions within this ecosystem.

Featuring:

• Gary Ballabio, Head of Technology Partnerships at Cloudinary
• Philippe Brodeur, CEO & Founder at Overcast HQ
• Teddy Zeskind, Partner Manager at Bitmovin
• Joaquin Lippincott, CEO at Metal Toad (moderator)

Joaquin Lippincott 

Welcome, everybody. We are here today for the HPA Tech Retreat. We're doing a panel called "Understanding Cloud Media Workloads." Really the genesis of this is we're talking about how much the media and entertainment industry -- and the recording industry, streaming industry -- have been impacted by technology, and specifically, cloud technology. The HPA Tech Retreat, for some context, gathers speakers and attendees from some of the brightest minds in the industry. The goal here is to have a discussion between experts that is candid, unbiased, unvarnished, and authentic, in a way that is unique to this forum. So with that in mind, I want to introduce our esteemed panelists. I'm going to start with Philippe Brodeur of Overcast. Welcome, Philippe.

Philippe Brodeur 

Thanks very much, Joaquin. It's really nice to be here.

Joaquin Lippincott  

And we also are joined by Gary Ballabio of Cloudinary.

Gary Ballabio  

Hi, everybody. Thanks for having me here. 

Joaquin Lippincott  

And then last, but certainly not least, we have Teddy Zeskind of Bitmovin. 

Teddy Zeskind  

Hello, happy to be here. 

Joaquin Lippincott   

Philippe, you and I talked a little bit about live events. What is a live event? What's the nature of a live event when it's a challenge to bring people into the same room?

Philippe Brodeur  

Live events — and we're having a live event now, but it's a pre-recorded live. So now, we've talked about live events as well — which are live events. It's like live television. What's live television? Is it scheduled television, or is it live television? Terms are important. They're there. They're being redefined everywhere we look at the moment. So, you know, you look at something like cloud and MAM, DAM, PAM — they don't exist in cloud, right? It's much more composable. But when it comes to live events, you can run a live event and then have live events running off that live event. It can be multiple live events running all at once. So you've got your live sport event that might be going on, — you might have live, Twitter feeds live, lots of live events, that are related to that. You could even have live events happening within the stadium, which are different feeds going to different people and different interactivity happening.There are no rules here. There are no rules here. What the business leaders need to understand is that we're going through this transformation. In which case, we have to try different things and understand what the technology is capable of, in order to be able to have the business solutions and make money at the end of the day. Because not making money is a challenge in this world when revenues are certainly under pressure.

Joaquin Lippincott

Yeah, the business model has to adapt. It's uncharted territory, both in terms of the, "What do you charge?" but also the cost. There was a really interesting article that I just read. When we look at Peacock and HBO Max — some of these streaming providers — they're spending literally billions of dollars, and no matter what they're going to make on the front end of this, in terms of what the consumer is going to pay — the cost, the cost benefits for the cloud are incredible. But the potential to spend infinite amounts of money, or for all practical purposes, infinite amount of money is really intense as well. I think that also, private businesses are sort of trying to figure out what the cloud actually means. How much are they going to leverage the public cloud? How much are they going to do themselves? The risks are really significant. Let's switch gears and talk about the media supply chain. So when we talk about the complexity of the media supply chain today, in the conductivity, the interactivity; it's never been more complex, or robust. So Teddy, I'd like to ask you, how do you debug issues? We were just talking about this in terms of, if I'm watching Ancient Aliens and the application crashes, what was it? Was it the app? How do you do that? How do you even know what's going on? 

Teddy Zeskind   

Yeah, "What was it?" is the main question there. At Bitmovin, when we speak to our customers, one of the most important themes we're hearing around debugging is sort of reducing that cost of errors. So, I think it was Vimeo, who just released information that 6% of subscribers will churn your service due to technical reasons. So, as an operator, not only do you need to have apps on all the different platforms, but they need to work. Roku needs to work, Samsung Tiizen needs to work, LG needs to work. And you need fast startup times. You need limited buffering. You need no rebuffering on these low power TVs that maybe were built in 2016 or 2017. In a meaningful way that our customers combat the cost of errors on those sort of lower-powered TVs, is to ensure that playback is smooth with a battle tested video player — and also robust analytics to then troubleshoot those issues if they do pop up.

Joaquin Lippincott  

The challenge of the hardware is significant, although it's gotten a lot better. We had a project at Metal Toad where we had to deploy something to the first generation Smart TV coming from a large manufacturer, I won't specify who. But when we looked at animation, and the customer said, "Hey, I want to animate something," we looked at how many frames per second we were getting. Back then it was one frame per second, so that was not going to work. Gary, how about you? We're talking about interconnectivity. Talk about Cloudinary and what you guys are seeing over there. 

Gary Ballabio   

I think, just to reiterate a little bit what Teddy said — I think the telemetry is so critical. And making sure that you're collecting data. You have to really pinpoint the spots along the path — where you need to be collecting data from — so you can really pinpoint, "Is it an end user connectivity problem?" Where along the chain is it? Is it on the CDN layer? For live streaming it can be even more complicated, right, in terms of ingestion and the transcoding that's happening. So the telemetry is really key. I would say, it's one of those that you try to set it up, and if you have it set up right, then you'll be able to resolve those tickets very quickly. But then you learn as you go along, and it's so critical to talk to other people and understand where they've seen problems in the past. Is it at the employer level? Is it at the end? How do you debug end user issues? Having the right tools, getting the right telemetry, is just so important for us. Luckily, at Cloudinary, we're dealing with VOD right now. There's still complications in dealing with VOD, especially if we're talking about ad insertion in different systems that are coming together. But really, it's the right time of telemetry. Where are those windows, at the points where the systems are coming together and making sure that you're collecting the right data to pinpoint which system is it that's having the problem? 

Joaquin Lippincott  

Philippe, what are your thoughts on that?

Philippe Brodeur  

I was just going to pick up on that because the permutations of what can go wrong are enormous. We look at it in a very linear way. So formats, hundreds; workflows, literally thousands; tech tools, thousands; publishing platforms, dozens; devices, tens. You multiply all those together, you're up to about 100 million different permutations. Things can go wrong anywhere along that chain of that technology. And that is a massive challenge.

Joaquin Lippincott 

Yeah, and I think beyond that, there's also the diminishing returns, right? That's, I think, why it's so critical for these large studios, or even small studios. Because one of the things that we've seen is really the democratization because these tools are on demand. People who want to focus on creating content, it's critical to have the right partners in the room to really think about, "How do we do the content creation, the media supply chain distribution?" Because that's not core to the business. The core to the business, in many cases, is the content, the audience — all of that — and that's a full time job. More and more, we are able to move to sort of an a la carte system. How is that impacting your guys' business models? When customers are saying, "Hey, I want this feature, but I don't want that." It's very different from where we've been in the past. Gary, how is that impacting you? 

Gary Ballabio  

From a Cloudinary standpoint, we're certainly right in the middle of that, where our customers have the choice to really leverage the best of breed systems that they want to leverage. That's great. They can take the systems to be able to connect them. Certainly, there's an upside to that because now you're taking the one system that has the exact features that you're looking for, and it's a best fit for your use case. You're chaining them all together. That's great. Of course, there is the downside to that in terms of, you have different contracts with all these different companies. It's not just buying. It's not a one-stop shop anymore. So now you have different contracts to negotiate and manage and everything, and also from a support standpoint — you know, going back to that whole thing of debugging and having the right telemetry in place. So now you may need to be contacting different support teams to kind of help you with issues, especially if they're production related issues. So, we're definitely right in the middle of that — offering the right APIs to help with the right telemetry offering and make it easier to connect different systems and to be extensible, essentially. 

Joaquin Lippincott   

And Teddy, are you seeing the same thing on your side? 

Teddy Zeskind 

Yeah, you know, we've always taken that component approach here a Bitmovin. Stefan and Chris, when they developed our encoder player in analytics, I think they had this vision to essentially slot them into any existing video workflow. We're hearing more and more from our customers that they're preferring to piece together, like Gary said, best in breed solutions and build that strategy themselves, or go to someone like Metal Toad, to build it for them. I would say one of the bigger lessons we learned this year is it also pays to have  that adaptability and that flexibility. The ability to swap out areas that might be cost centers in a year or two or and replace it with the latest and greatest is pretty invaluable. 

Joaquin Lippincott 

Yeah, that the whole amortization question within the cloud is really interesting because it's no longer, "We bought this equipment. We're going to amortize it over the next five years," or whatever the case may be. Philippe, I'd like to shift the question a little bit and talk about, what's the strategy on your end? Or what do you see the right strategy for technology partners to be in this kind of interactive, cross functional, a la carte card world? Do you think specialization is the answer? What does a win look like, when a customer might be dealing with different vendors? What do you think?

Philippe Brodeur 

So I think, picking up what the other guys said, yes, composability is absolutely central to what everyone has to build at the moment, right? Be that microservices, or be that individuals solutions, and then also having an API that can bring in best of breed. So there's two things, right? It's affected us significantly recently. And so we broke down one of our products into nine different products. And becomes configurable, as opposed to customizable.And the very big difference in that is that one doesn't need a developer in order to implement it. You want configuration, you want less customization, and the software solutions of old are all around customization. I think the main challenge at the moment for the CIO, so the technology side equation, is the pace of change, right? And with the pace of change as it is at the moment, that can produce two things. One is paralysis. So don't do anything because I don't know where to move. And the second one is the uncertainty of backing the wrong horse. Nobody wants either of those. So I think to start with, you really need to start with the idea that you're going after a composable solution. To be fair, all three of us here are in that space. And It's actually exactly what you do at Metal Toad, as you bring partners together. And that is absolutely core to what is what's happening going forward, and to try and win the trust. Now, what is a win? I think the first win is actually just winning the trust of both the CEO and the CIO, at the same time.

Joaquin Lippincott  

So, I want to ask sort of like the big picture prompt. You know, if you had the ideal customer in the room with you — I want to ask all three of you, but I'd like to start with you Philippe and just ask you — what would you say? Like, what do they really need to know about what's happening in the industry today? The risks? You know, how would you frame that?

Philippe Brodeur 

Right, that is a very good question, and it's a big challenge. I think the first thing you really need to understand is where are they on the digital and business transformation roadmap, okay? So, where are you, and where do you want to get to? So, that's the first thing. The second thing is let's talk about streamlined business processes first. Okay, I know this is a technology discussion. But there's no point in bringing in new technologies if you're not going to have a business outcome and a business model. So let's talk about those business outcomes that we would really like to see, like what is possible with AI? What is possible with machine learning? How do we — if you can automate a whole bunch of mundane tasks by using third party services — like you said, Lambda completely reducing a cost to almost nil -- then what are the opportunities out of that and how much time are you saving and what can you put that time into doing? And these are tough questions to answer, but they're very tough questions to ask as well. Because you need to get into the head of the organization — the heads of the — the combine the collaborative heads of the organization and say, "Look, the pace of change is fast. We all don't know where we are. But we do know one thing. If we start from the bottom up, with a composable solution, then we can try different things. We can bring them in and swap them out." And by doing that, you will future-proof your solutions into the next decade. So, that's really where I think you need to start with a company, and then take it from there because, you know, we all know that there are a lot of cowboys out there, right?. We see it in their marketing. They say that they can do this, that and the other, and they're selling off paper. Now, we've got to not sell off paper. We've got to sell what we can configure, as opposed to customize, and then take it from there.

Joaquin Lippincott  

Interesting. So Teddy, same prompt to you. Same question. If you had your ideal customer — CIO, CEO, CTO, whoever that might be in the room — what do you tell them? What do they need to know about Bitmovin? About why they should be using your service? 

Teddy Zeskind 

Yeah, so I guess the another way to answer the question is what's important to our customers when we talk to them. And there's really three things that our customers consistently are coming to us about. Number one is providing a really great experience. So video content -- it's delivered. It's consumed on a multitude of devices and platforms. You know, when you're spending large sums of money to create this content, it's also equally important to invest in delivering the content in the highest possible quality that you can. The second point is, you know, reducing time to market. How do you beat the competition? How are you a first mover? Are your suite of solutions enabling your business to free up internal resources so that you can bring the focus back to making that great content and staying ahead of the curve? And then the last point is just reducing costs, right? So what's that operational optimization look like? Are you able to control costs? Are you able to mitigate risk? Are you able to manage this complex workflow that's vital to the success of your business? 

Joaquin Lippincott  

And Gary, same prompt. Ideal customer — what do you tell them? 

Gary Ballabio 

Starting with their goals, understanding the end user experience that they're trying to move towards, aspirationally, where do they want to go? Are you trying to be more like a Disney+ type of experience? A Netflix type of experience like what is it that you're trying to accomplish? At Cloudinary, we're very much an enablement platform. So there's a lot you can do with our APIs to help deliver that type of experience. But it does start with the goals and the metrics that you're seeking to, to measure basically. So that's the first thing and then from there, just understanding the composability. I do think that there's still a ramp that's occurring, in terms of understanding. How you can take, first of all, what are the boxes in the architecture and then how you can fill those boxes with different technologies to achieve the end result that you're looking for? So helping, you know, just helping them understand that and helping them embrace that is really ideal for us. 

Joaquin Lippincott 

Thanks to all of you. It's been an absolute pleasure and really interesting. I think we're just at a really interesting inflection point in our industry.