About Us | Login | Follow CITO Research:

Profile in Nerdage: Noah Broadwater, CTO, Sesame Workshop Part I: Asset Management

Nerdage Brain

Noah Broadwater is the CTO of Sesame Workshop, the nonprofit behind the Sesame Street children’s media juggernaut. Dan Woods, CTO and Editor of CITO Research, sat down with Broadwater to discuss asset and portfolio management from the perspective of the CTO, the effect and dictates of consumerization on the modern CTO, and his thoughts on demystifying the cloud.

In this segment, Broadwater discusses asset management.

Dan Woods: How do you define asset management, and why is it so hard?

Noah Broadway

Noah Broadwater: Asset management is quite simply the ability to actually understand, catalog, store, and reuse your data in many ways, the components of which run your business. For a media company, assets include images, photographs, music, videos, scripts for the videos, and sound effects. An asset can be as simple as a sound: “doink.” It can as complex as an entire show.

So the word “assets” encompasses a lot, and I think in most companies—especially media companies—when people talk about asset management, they think about cataloging absolutely everything so that they have their entire history fully listed, catalogued, and stored. As a company that’s been around for 43 years, to catalog everything, our data center would need to be the size of our building.

Woods: So what’s the right approach?

Broadwater: It’s not a technical issue. It’s culturally and politically impossible. If you try to make everyone happy, nobody will be happy. Trying to do everything just doesn’t work. Pick the low-hanging fruit.

Woods: What tends to be the low-hanging fruit in asset management?

Broadwater: A lot of companies will say it’s image assets. Others will say it’s video. UPS doesn’t have a lot of video, so video is easy for them. We’re Sesame Street. Forty-three years of Sesame Street is a ton of video. It’ll take us a long time just to process it. And politically, that’s also harder, because everyone has a viewpoint.

Right now, we’re focusing on audio. If you have a small group, you focus on that group. You’ve got a team of five people and they are the only people who deal with audio. They can focus on it and put in the time and really attack it. It’s not 100 people and 100 viewpoints. It’s five people, and they get it done.

The same thing happened in publishing. Publishing had six people. We got them into a room, heard their business case, worked together to find a solution, and today we have a great publishing workflow. It’s about understanding how to achieve small wins. And once you have a small win, everyone else wants to know how to do it too.”

Woods: This sounds like an object-oriented analysis, where you try to isolate the little cluster of information that is tightly related, to which a given solution might lend itself, either through a vendor solution or in some other way. When the information is in a tight cluster, the political, transformational, and workflow issues are manageable, allowing them to be solved in a smaller arena.

It sounds like what you’re saying is that you create a culture of asset management, where you show that you can be successful on a modest scale. You then gradually create islands of success, until eventually, more and more of the problem is solved. Do you ever have to get to that vision of everything categorized?

Broadwater: Maybe we don’t want to get there. Because, if you start out with this idea of, “We want to catalog everything,” that’s nice from a library-services perspective. But step back and think about it from a business perspective. Do you need to? We probably have over a million photos. Roughly a few hundred thousand are duplicates, or just slightly different. I took a picture of Elmo from this angle, and then I took another one just slightly different. Do you need both? Not really.

So you don’t want to digitize everything. You don’t want to put everything into asset management because you don’t need to. You have to think about what is going to generate return. And it’s not always just financial return.

For example, we do our retrospective every ten years. We have a 40-year anniversary book. We’ll have a 50-year anniversary book. We know we want pictures of Jim Henson, so those have to be cataloged. Do we want pictures of Jim Henson’s foot? Not really.

Do you want every single aspect of Elmo? Well, you want one series of that, but you don’t need 20. From a cost standpoint, it just doesn’t make sense. The process of digitizing involves paying someone to select the photos. Everyone forgets about that aspect.

Woods: So you have to have some model for value.

Broadwater: Absolutely.

Woods: Do you let the value proposition drive the choice about where you practice asset management?

Broadwater: Part of it is around a value proposition, and part of it is about a basic need, which is the other thing that drives a lot of projects. When we started our website, we were talking about asset management, and everyone was trying to do asset management and sell it into the organization, en masse.

Then the website came along, which I was majorly involved in, and I said, “We need asset management for this project.” So there’s a value proposition right there. We’re going to do this.

The reason why we’re focusing on audio was that we were doing an outreach project, and a person from outreach said, “I need these songs, and I can’t find them. And I need these sorts of sound effects.” The person who runs our digital asset management group had to say, “Well, we don’t have any of that cataloged. That’s a good task, and a small enough group that we can focus on it, and it has immediate value.”

Some of it’s a creative decision. Our production team and marketing team are going to have some creative decisions in saying, “These are the things we think are valuable. We’re working on a book; we need this.”

Woods: Once you create a system and have done the work of cataloging it, how do you design and refine the process of making sure it’s up to date and that new assets are incorporated?

Broadwater: There are two components to that. One is another common mistake, which is, “tag it once, and it’s done, and you’ll never have to look at it again.” Things change. We base our system on the Dublin Core system, which is like a Dewey Decimal system for asset management. It is refreshed every few years.

There are certain specific things that Sesame does—ways we catalog our assets that no one else ever will. How many people care about our characters and the way we use them? Once we create new characters, we have to go back and integrate them. It’s a living, breathing system. You have to understand that and put in place ways to modify it with the least amount of resistance. A lot of it’s programmatic.

This is where the backend piece needs to be really great. If you have a system that’s really strong, you can programmatically make adjustments to the whole system.

The other part is, you have to make it worthwhile for people to do it from the get-go. So now, we’re about to go into production for the next season of Sesame Street, and as we do, we’re thinking about how we are tagging those audio assets as we create them.

Woods: I understand how you’ve created awareness that you need to tag, but what was the reward there?

Broadwater: There are multiple rewards. One is rights management. We’re a media company. We have to pay royalties. For any actor, we have to pay guilds, actors, musicians, writers, composers, and directors. You have to know everyone who’s attached to any given segment, whether it’s audio or video. You have to know which photographer took that shot. That’s all asset management. The immediate reward is: “If you don’t do it, you can’t use it.”

We’re also creating what’s called a metered system, which identifies the easiest things to reuse that cost the least or are easiest to get clearances on, or have the widest variety of usage.

Let’s say there’s a video clip, and you can use it on the Web, in marketing material, and in domestic or international broadcasts. It doesn’t have that many guild payments or a lot of extra royalties. So, when you’re are trying to do something mission-based, which for us means trying to help the most underserved, and you don’t have a lot funding, you want to find those assets quickly.

There’s a huge reward for having everything tagged. You can find what you need. When we’re building outreach around resiliency, you can find what assets meet that need.

Woods: So it sounds like the value in the system drives the change in the process to keep it up, and if people don’t see that it’s worthwhile, maybe it’s not working properly.

Broadwater: Correct.

Woods: The value drives the impetus to maintain the system.

Broadwater:The funny thing is, we don’t often think of finance in [IT] asset management, other than to imagine them saying, “Well, that’s going to be too expensive to digitize.” Our finance department is very aware of asset management, and they’re often very much behind it because they know they have to make these payments. It helps us make a better determination about what we should pay for.

Our legal department is very much integrated. For any project, I stress integrating legal early. They’re your best friends. The legal department can say what the clearances are and show us where we can use assets.

If they do it early, they don’t have to be asked later. So my other piece of advice is, “do it now so you don’t have to stop and do it later.”

Woods: Gil Elbaz, founder of Factual, previously created a company called Applied Semantics, which was what became AdSense for Google. His insight is that data is a liability, not just an asset. So, when you create an asset management system, you’ve also created a liability, which is the cost required to keep it up to date.

I’d be interested to hear, what are the signs that your costs—the liabilities—for asset management are exceeding the value it’s providing? Do you analyze the value being created and whether you’re creating a liability?

Broadwater: We actually do. We do an analysis that says, “Are we cataloging the right things? How much is too much?” This goes back to the issue of having all these photographs. We don’t want to tag them all, because it’s going to cost too much. Our CFO is part of the steering committee for asset management, and he’s always got an eye on total cost of ownership.

So far, we have asset management that allows us to create relationships and work with other systems that are often consumer-facing and help our mission—because we’re a mission-based organization. So, if it costs a little extra to provide better resources for underprivileged children—actually, for all children—it’s worth it.

So far, we’ve found that we’ve made up the money we’ve spent at least one and a half times. I’m not going to say that there are massive amounts of revenue coming in from it, but you can look at it as “speed to market.”

Woods: Benefits aren’t always financial, and you have to take that into account.

Broadwater: We tag all of our content educationally. If a parent is looking for all the videos or games on sesamestreet.org about economic stability, we have videos and games on that. If we didn’t tag it, they might not find anything, and they’ll go somewhere else. In that case, we’re not accomplishing our mission of being able to provide them with resources, and we’re losing valuable customers. So, there are definitely benefits beyond financial benefits.

See Part II of this interview with Noah Broadwater, which focuses on portfolio management.

I’m told by people who would know that Jeff Bezos thinks of Amazon as a software company, not a retailer or an e-commerce company.

There comes a time in the life of every important technology where the CEO gets interested and wants to know what the fuss is all about. That time has come for Spark.

How is your organization going to harness the power of mobile devices? In many ways, the answer to this question is a proxy for your approach to technology in general.

Creation of VMs from a golden image can be automated, but the requirements are not easy to discover.