Integration Therapy’s Rebecca Avery Talks Metadata Transformation and AI's Impact on Streaming
Integration Therapy Owner and Principal Rebecca Avery joins Future Frames Podcast’s Doug Daulton to discuss the challenges of technology integration and metadata transformation and risk mitigation in this candid and wide-ranging interview from Streaming Media 2025.
Streaming Network Transformation and Metadata Translation
Avery says that her consultancy, Integration Therapy, which aims to help “streaming media operations that leak revenue,” has that name for two reasons: “‘Integration’ because streaming networks are complex things, streaming networks, and ‘Therapy’ because I know how it feels to try to scale a company when things aren't working.”
She goes on to say that what she and her company are mainly focused on now is working with “streaming networks who are either getting ready for a major transformation—which a lot of broadcast networks are streaming arms are—or they're experiencing hypergrowth going out into new markets or expanding their content libraries. And most of the work that I have been called in to do is about changing from many to one, streamlining what I call the 'content superhighway' system.”
Asked by Daulton to expand a bit on what that means, Avery says, “There are actually several different things that you need to look at when you are changing out what feels like something that's just tech ops or tech, the first being strategic alignment. Industry benchmarks actually put lack of strategic alignment at a 30% revenue leak for many companies," she continues. So Integration Therapy's aim is to help these companies "really understand what they're trying to do and why they're trying to do it. What's efficient. Metadata translation. So, my sweet spot is metadata. I am the chair of the Metadata Working Group for the SVTA. It's my first love. And then also there's team development then technology integration as part of that. When you change one of those five things, it affects the other four. So that's part of what I'm there to oversee."
She goes on to explain that "where most streaming companies have been in the past is one of two paths. One is either 'We're just going to change our metadata 'cause we just need to do a heat-seeking missile targeted fix,' or 'We're going to call in McKinsey and we're going to do a gigantic transformation that's very expensive and very disruptive to the company.' I'm here to provide that middle path and to start as a heat-seeking missile and then pull back for a second and say, 'If you want to move fast, start slow and, when we make these changes, here are the things that's going to affect and here's how we can mitigate that risk and not lose the pace that your company needs to function.'"
Daulton concurs on the "start slow to move fast" approach. "If you're doing that consistently, you're going to be getting better results quicker rather than just trying to throw everything at the wall and then see what sticks."
“I think that's more important now as technology lifecycles are getting shorter and as there are more and more levers for streaming companies to pull to go for monetization because they want to pivot so much,” Avery says. “If you're not stable at your core, you're just going to start falling apart.”
Reflecting on her experience at the conference so far, Avery highlights the AI discussions on various panels. “This whole community has been making AI such a part of the conversation and I've really enjoyed the talks, the panels that I've been in and the panels that I've gotten to watch about how that conversation is starting to shift and say, 'okay, we understand that we are in the age of AI. Given that context, where are we with everything else? What do we want to become?'"
“Talk to me a little bit about your perspective on AI in the space,” Daulton says.
"I think it is a huge game-changer," Avery affirms. "Like I said, my first love is metadata, and that is one of the first places that AI has really gone in. I recently published an article that says we talk too much about AI because we've been kind of obsessed with, 'Well, what is AI? How does it work? What are its limitations?' And I think as a result, a lot of the companies that are developing AI for the media space are lacking a little bit of clarity on what they need."
Avery goes on to say that one area where AI has made an immediate, demonstrable impact is localization (aka localisation). “That is one of the few places that media immediately went, 'Oh, we need to localize, not just translate, but we need to be able to push out in Europe.' AI has done a really amazing job stepping up to that challenge. And now I am lucky enough to be part of the intelligentsia who's defining what all of the next things are for discoverability rights management and all the places AI can go."
"I think it's not so much that we're talking about AI too much, it's that maybe we're talking about the wrong applications: the focus is on gen AI and the creative piece,” Daulton contends. Really, the immediate bang for the buck is on these things like metadata acquisition, cataloging, and localization, getting these things that both make the cost of creation lower and also create and expand markets that. Things that are not as sexy, but they're integral to success. Would you agree with that?”
"Definitely," she says. "I remember the days when metadata wasn't sexy and it's finally getting its time in the sun. It just won an Emmy, so that's great. I would also argue that we've been looking kind of at the wrong lens. So I have been working with giant enterprise companies as well as smaller startups and something that really happened probably starting about three years ago when ChatGPT was about to release and people were kind of on the edge of their seats. The conversation was, how can AI help solve my problems? To me that sounds like how can that monkey wrench help fix my Jeep? And I really think that we can flip the conversation and say, 'Here's what I want. Here's how I want to optimize my operation.'"
One of the challenges of predicting or assessing AI's likely impact in the streaming space or trying to explore long-range possibilities and cultivate future development is that in this industry, "tech lifecycles are fairly short." On the one hand, Avery says, "we have these green pastures of saying, 'How do we want to monetize? What do we want the experience to be? How are we going to reshape our UI in a way that doesn't just look like a recycled version of somebody else's and make it interactive and do something special and really challenge technology?'" But too often, the conversation is limited to, "What's been developed today?" And in a sense, she concludes, "It doesn't matter because by the time you learn it, it's going to be in a new generation anyway."
“I think that's an interesting perspective," Daulton replies. "So, as you're trying to figure out what the problem is in the first place or what the solutions might be, you may define something that you can't solve right now, but you put it on the board and say, 'Somewhere down the line, we're going to get some kind of solution.' So you're not trying to solve a problem that exists, but there's not a monkey wrench to apply to it yet.”
"I think to solve the problems that exist, to understand how to prioritize those, you should be looking at your longterm," Avery advises. "What do you want to do in three years or five years beyond five years is very tricky right now, but you should be looking at, 'Where do we want to be?' And I think that there is a huge horse race right now of all these AI companies that are trying to win. I think media has the potential to be a very influential customer of AI, and I think that if there is enough clarity of vision in this industry, we can actually help impact how AI gets developed and for whom."
Ethical Applications for and Implications of AI
"One of the interesting things about the work you're doing," Daulton says, shifting the conversation a bit, is that "you're not coming at it from the technology side or [viewing AI through a] technology lens to begin with. You're coming at it from a human lens. 'How is this going to benefit the person in the pipeline that's doing the job? How is it going to affect the business bottom line? How's it going to affect the audience and the creation of more and better content or greater access?' And one of the things that near and dear to my heart is the idea of ethical applications. Could you speak a little bit to that?"
"There are a lot of different ways that companies can win in more than one way," Avery says. "There are ethical applications from the point of view of, 'How are you treating your internal team when you are putting in AI?' And I think that there is actually a very neutral way to do it. One I always say when I'm talking about accessibility panels is, 'If you're not a good person, don't worry. I probably have a fiscal reason for you to do the right thing anyway.'"
"I love that so much because doing the right thing is a benefit in and of itself, and particularly when you think about all this generative AI stuff," Daulton chimes in exuberantly. "If you're doing the right thing and you're bringing things on-prem and you're doing things where you're building an IP chain that may cost you some money, it might be doing the right thing, but it also protects you in the long run from legal challenges and other things."
"It does,” Avery says, “and it protects you from over-laying off and then realizing your AI can't do what you thought it could do and accumulating a huge amount of data debt and then having to rehire back people, which honestly is a thing that's kind of happening in some ways and it's just creating more confusion. There's huge cost overhead to having that kind of employee churn."
Going further into the consequences of excessive layoffs (AI-driven or otherwise), Avery goes on to say, "There is a term called 'organizational trauma' that happens whenever there's layoffs. There's been this new thing of 'I've been affected by the layoff' and my first thought is always, 'No, man, you got laid off and the people left behind were affected by the layoff.' This hurts everybody. It affects the trust of the teams. The quality of your product invariably sinks. It affects the trust of your audience, it impacts the economy. The ripple effects are almost non-stop."
Join us December 9-11 and tune in for more great conversations at Streaming Media Connect! Registration is free and open now!
Related Articles
Streaming Video Technology Association (SVTA) Live Operations Expert Bhavesh Upadhyaya argues that it's time to move on from vague discussions of AI as streaming's "next big thing" to emphasizing practical applications, addressing real-world streaming workflow and delivery problems AI can solve, as well as preparing for the inevitable ways AI will impact the streaming business in this interview with Future Frames' Doug Daulton on the show floor at Streaming Media 2025
14 Nov 2025
Bruno Giner, VP Sales, Global, at Nice People At Work (NPAW), discusses the importance of insights in streaming optimization and how NPAW works with its collaborators and partners to cultivate a data-driven and data-sharing culture across the streaming industry in this interview with Streaming Media contributing editor Timothy Fore-Siglin from Streaming Media 2025.
07 Nov 2025
In a candid interview at Streaming Media 2025, EZDRM COO and co-founder Olga Kornienko and Future Frames producer Doug Daulton trade helpful analogies in discussing the growing importance of content provenance in the age of generative AI and deepfakes, as well as the role that the C2PA standard and DRM providers like EZDRM can play in protecting content authenticity and content creators' rights in a changing media landscape.
24 Oct 2025
Almost live from Streaming Media 2025, Vevo Head of Data, Research & Measurement Natasha Potashnik discusses the many uses of streaming metadata with Streaming Media Contributing Editor Timothy Fore-Siglin.
10 Oct 2025