Okay. Good morning. I'm Paul Hennessy, CEO of Shutterstock. Welcome to Investor Day. That's where you applaud if everyone could start. Thank you very much. Thank you very much. I can see this is gonna be a slow crowd, I'm gonna have to do extra energy up here. For those of you that came to the Empire State Building, thank you very much for coming. Welcome to the Empire State Building, home of full-time home of Shutterstock's worldwide headquarters. If you look out the window to your right, part-time home of the mighty King Kong. Be careful today as you engage in our Investor Day. For those of you that are dialed into our call remotely, thanks for joining. It's great to have all of you here and all of you connected.
I think we're gonna give you a good show today, so thanks again for coming. Can't have this particular event without referring you to our forward-looking statements, and if you've got other questions or issues, refer to all the reports that we file with the SEC. I've said that we can move forward. We've assembled our leaders and subject matter experts to give you a much better insight into Shutterstock. One of the things I'm absolutely confident and certain about is by the end of our presentation, you will be as excited about Shutterstock as we are all excited about Shutterstock. We live in an amazing world and an amazing time. The world is moving incredibly fast. We now live in a world where words like machine learning and artificial intelligence are in our vernacular.
Acronyms like AR and VR and XR and AI, these are words that we throw around both at cocktail parties and little league fields. With the speed of change that's going on and the pace that the world is moving, there's only one way for us to stay ahead of that speed of change, and that's to lean into the technology, lean into the world called the Metaverse. In the spirit of that, you'll notice that our good-looking leaders here that are distinguished in black and white have been transformed and are boldly leaping into that Metaverse with our own avatars. We think that that's critical. If you think about the old world and the new world, it's a much better world, isn't it?
I know it's not lost on you that my avatar had a lot more time than I have to go to the gym, and we're gonna rightsize that. Yet another reason for me to spend my full time in the Metaverse. I've got a much better look there. You'll notice that Jarrod's in a suit and tie, and I think if you've seen him walking around, you'll see him in a suit and tie today. It's hard to know what's real and what's the Metaverse. We have assembled a really great agenda. We were really thoughtful in what we wanted to make sure that you took away from our session, and I think you'll also find it informative and a lot of fun and maybe even surprising. I look forward to walking you through this agenda.
The team's excited to share this agenda with you, and we're gonna open up with our business model. You should understand our business model well. We're gonna make sure you've got that. We're gonna walk you through our content engine, our Creative Engine, and our now emerging and very important Data Engine. We'll wrap with, you know, a really robust discussion of our financials, and we'll leave enough time for Jarrod and myself to answer any and all questions that you may have. Sound good? Great. 20 years ago, our business was purpose-built when our founder, Jon Oringer, tried to do a simple task. He wanted to license content for commercial use, for use in digital advertising, and he found that that was massively complicated and very expensive. He built a business to fix that problem, and he did. He fixed it.
I'm honored to say that Jon Oringer is in the room today with all of us. Jon, if you're around, could you raise your hand. Jon's in the back there. Let's hear it for Jon. Thank you for joining us. Jon, it's not lost on me that I would not be standing on this stage and none of the people in this building here would be here, and people on the call would not be here if not for you. Thank you for that and you've built an amazing company. He started with a content engine, a simple two-sided marketplace, serving contributors and customers in the e-commerce space. The flywheel started to turn from the day that he started this. Contributors gave us content. Customers liked that content and downloaded that content. That creates a royalty.
We paid our royalties to our to our contributors, and they liked the royalties. More contributors came and brought us more content, and then more customers came and downloaded. You start to see how this flywheel starts. By the time we IPO'd the business in 2012, we had about 35,000 contributors, we had about 500,000 customers, we had 19 million some odd images, and we had about 500,000 videos. The business was basically profitable right out of the gate. It's impressive. Today, it's much more impressive. The flywheel continues to spin in our content engine, and today, things like our 19 million images in 2012 now stand at over 600 million images. Our 500,000 videos now stand at over 45 million videos.
By the way, twice the closest competitor. We've got two million music tracks. That's really constructive. We've got sound effects, millions of sound effects. We've got 1.5 million 3D models. Those 35,000 contributors jumped to 750,000 contributors in 2019. By the time we got to today, 2023, we stand at 2.3 million contributors giving us content. It's incredible. That's a 3x in four years. Super impressive. They've been well compensated. They've enjoyed $1.5 billion in royalties over the life of this company. We've got a very, very strong ecosystem with our contributors, but also our customer numbers have grown.
They've grown from about half a million at the time of our IPO to today stand also at $2.3 million. You can see that our customers are deeply engaged, like our contributors are deeply engaged. What's interesting is the volume of our business is extraordinary. We're downloading six pieces of content, six images per second. If I just count to five, another 30 images or another 30 pieces of content have been downloaded. It's an extraordinary business, it's showing up in our results, isn't it? Right, we had a very strong year last year, $828 million in revenue, $218 million in adjusted EBITDA, a very, very compelling margin story. Over 26% margins, that margin has been expanding year-over-year.
The content engine is cranking, and it's one of the growth pillars of our business. We've organized our business around two fundamental channels so that we can cover the entire creative market. We started the business, as I mentioned, with an e-commerce business. Think of that as our self-serve business that serves our individual creatives, our freelancers, our small customers, even medium-sized customers, in a self-serve way. We've, over the past couple of years, launched our enterprise channel. The enterprise channel handles the largest Fortune 500 companies in the world. Global brands engage deeply with Shutterstock. Here's what's really interesting about this. When the business was in 2012, about 10% of our business was from direct sales to enterprise customers. Now, as you can see by this slide, 40% and scaling.
Another interesting couple of facts about this. We're moving our business from a downloading business, a one-off business, to a subscription business in both the e-commerce channel and in the enterprise channel. Our average order value is also increasing. As we grow into our large enterprise customers, average order value is increasing. You can see the delta between our e-commerce average order value and our enterprise. It's about a 30x. The business is moving in all the right directions. It's naturally diversified. It's diversified in customer segments, as I've discussed. It's diversified in the channel and our approach. It's diversified in all of the products that we offer, custom-tailored products for the smallest of businesses to the largest of businesses. What I'd say is it's geographically dispersed now. Jon started the business in North America.
The majority of our business happens outside of North America. The business is strong, it's durable, it's scaled, and it, you know, is able to withstand variations of the world that we live in. When we started the business, our TAM was $4 billion, that's an attractive TAM for the image business. We've been really successful in moving from one single content type, expanding into more content types, and we've been super strategic in the M&A that we've conducted. As we've moved in content types and purchased companies that are in alternate content types, not only have we expanded the TAM from $4 billion to $16 billion, we're also moving into faster-growing TAMs so that our flywheel continues to spin. There are strong tailwinds that are also behind this business, we get to enjoy those tailwinds. What are the tailwinds?
There was at once a time when you would use stock imagery for, I don't know, the newspaper, right? The early advent of digital advertising. Now, customers need our content in all of their executions. Think about the things that didn't exist not even 10 years ago. Podcasts. Now we're being used in film, we're being used in television, we're being used in Metaverse-ready assets so that we can participate in whatever the hell the Metaverse is going to be. I'm not sure what that is, but we're gonna be there. As we've expanded the TAM, as we've moved into growing TAMs, and we've got these tailwinds, we're well-positioned, right? Well-positioned for the journey ahead. Here's something you should consider. The finish line for creative content does not exist.
We're seeing every day greater and greater use cases from our small, medium, and large customers that want what we have to offer them in a variety of ways. Now we're well-positioned. I'd like to say we're positioned at the intersection of creativity and productivity to be able to serve our customers well. Not surprisingly, on our 20th birthday, we need to realign our entire company around a bigger and broader mission to satisfy our customers. I, and I'll read it to you. Shutterstock is the end-to-end creative partner that empowers customers to tell the best versions of their stories. Our teams capture the moments, develop the tools, deliver the services, and provide the ingredients needed to simplify and supercharge creative storytelling. We have a great purpose.
We bridge the gap between idea, design, and execution, inspiring the world to create with confidence. These are important missions. You'll find them on the walls and on the TV sets of all of our offices because it's critical that all of our employees understand our mission, truly embrace our purpose, and move forward, aligned and executing. How do we execute against our mission and our purpose? I've mentioned our content engine. It's an important flywheel. It's the majority of our business, and you understand how that works. As we get more content, we get more customers. Our customers convert better. That creates royalties, which then begets more contributors and more content. That's a strong, powerful flywheel in our business. You're gonna hear more today, but we've also built a critical piece to add on to our content engine.
It's called our Create Engine. Our Creative Engine provides customers the tools they need to create, design, store, publish, do everything they need to do. That interlocks and spins the content flywheel even faster because they're also creating things like templates that ultimately become new content for other people to use. It's a powerful insight when you add the Creative Engine because guess what? Our customers want to use the tools that are easy to use and get content in the same place. By bringing those two together, the two flywheels work well together. Think you're all now very well aware that our Data Engine is yet another flywheel that's starting to spin. By having the content that I described, by having the amount of content that I described, not only do we have the content, but underlying that content is metadata.
That metadata has incredible value. It has incredible value to us because we can understand what that piece of content is, so that when someone searches for that content, we can deliver it to them. That content metadata also has great value to some of the largest brands in the world to train models for AI purposes. What's happening now is we're forming long-term strategic data relationships with the largest brands in the world, and this didn't even exist a few years ago for us. What's interesting is now when you bring those three flywheels together, the business scales, we serve our contributors better, we serve our customers better. That drives incremental growth, becomes a very, very powerful tool for our customers and for our company, which in turn then for our shareholders.
Interesting thought, I mentioned the content engine, it was the majority of our business and has been for quite some time. I'm here to tell you that in short order, in the coming years, we believe that the content side of our business will be less than half of our revenue compared to then the Creative Engine and our, and our Data Engine. These are emerging channels, they're growing fast, and they're very profitable. It's just the beginning. As a 30-year marketer, I know that there's a holy grail in advertising, and it's being able to predict which advertising works and which advertising doesn't work.
We now have the infrastructure, we have the underlying data, and we have the mechanism to start to help our customers, our advertisers, whether it's the smallest or the largest, to have insights on which creative will work better for them. You'll hear more about that in short order. Not surprisingly, customers like us. They really like us. We've got the major brands in the space, who trust Shutterstock to be their creative partner. They're deeply engaged with us, and we're creating new work, bespoke work, together to help them with their creative storytelling. I'll leave you with this thought.
Whether you're a florist or whether you're a Fortune 500 company, Shutterstock has the content, has the creative tools, ultimately has the underlying data, and has a path towards insights for your performance, to make you more successful. Thank you very much. What I'd like to do at this time is bring up Dade to tell you more about our content engine and give you some insight into our 3D business. Dade.
Thank you, Paul. All right. Thanks, everyone, for coming. Thanks, Paul, for the intro. You know, Paul talked a lot about the content engine and how important it was to our customers to make sure that they had the types of media that they needed in order to tell better stories. For them, that's music, sound, image, video, and 3D. Sorry. What I'm here to talk about today is 3D. A lot of people kind of wonder what 3D is all about, right? It can be a little technical, but really what 3D is, 3D has been at the leading edge of telling stories for about four decades now. It's come a long way, right? In the last I mean, in the last 10 years alone, we've seen it grow in all kinds of different ways.
I myself have been creating 3D for about 25 years, so I'm very passionate about it. You'll know 3D when you start to think about movies. Think about your favorite animated films, visual effects extravaganzas, all of those are brought to life through 3D. Architects, product designers, they use 3D to show spaces and buildings and things that don't even exist yet, right? Really, honestly, 3D can be very stylized, or as you can see in these images, very realistic. Even maybe sometimes better than life itself. Finally, 3D enables content creators who are making games to immerse users in truly visual experiences, right? What's amazing is that most every AAA game that's made uses 3D as a foundation, and it helps immerse viewers in that experience. To be honest with you, this is only the beginning.
As I was saying earlier, in the last 10 years alone, we've seen an entirely new use of 3D emerge in front of us. 3D has gone well beyond visual effects, gaming, architecture and your typical formats. Now 3D is being used for machine learning, where entire worlds are being recreated in 3D to run simulations. Thousands of simulations in minutes, whereas normally that might take years. It's amazing to watch. Along with that, we also have Metaverse. Metaverse, yeah, I'm using the M-word. You know what? I'm gonna continue to use it until we have something better. For now we'll use Metaverse. Metaverse is really changing a lot of things, right? It's really giving people an immersive experience to really experience brands in a whole new way. Everyone's lining up to get their brands into the Metaverse.
Finally, AR, VR, MR, whichever R you wanna use, is being used by medical professionals to reimagine the human body in ways that was never possible before. Did you know that right now, kids who have access to AR and VR medical experiences in high school sometimes know more about human anatomy than people coming out of college because they're able to experience it in a entirely new way? It's really, it's an amazing way to visualize the world around us, whether, again, it's very stylized, this is a very stylized human body, or very realistic. Now, here's a question: How do we meet this demand? Think about it. Digital twins, we call digital twins, basically a 3D replication of anything real in the world is a digital twin. Digital twins of everything in the world. That's a moving target, right? Like, that's something that's continuing...
Every day it's changing. How do we do that? Well, we have our ways. We are replicating everything in the world, as well as stylistic things that are not necessarily real, right? The amount of content is overwhelming. I'll give you a great example. We've just recently partnered with a contributor who is now building one of the most detailed anatomically correct human anatomies, male and female. What's amazing about this is that this model was took eight years to build. Eight years of 3D professionals, medical professionals, texturing artists, photographers, to manage every single system in the body and represent that in 3D. This is amazing because every single layer is You can go through those things, you can turn things off, you can look at them in whole different ways.
This is how people begin to understand human anatomy. You can't do this with photography. You can only do this with 3D. You can move around it. You can see it 3Dimensionally. It's really an amazing experience, and it's changing the way healthcare professionals will see the world, will see the human body, will see everything around us. One thing I always say is that for a long time, 3D was living in a walled garden. It was difficult to use. It required both a very technical and a creative mind at the same time. That's a unicorn. That's not something that you usually see out in the wild, right? Easily. What we've done is we've sort of breaking down those walls. I came along with the acquisition of TurboSquid about two years ago.
If you're not familiar with TurboSquid is the leading 3D marketplace on the planet. TurboSquid has been around for 25 years, as long as I have in the 3D industry. In those 25 years, we've unleashed 3D content to everyone. We've come up with ways that allow 2D designers, photographers to utilize 3D in ways that they've never been able to do that before. We have the expertise to understand what it means to utilize 3D content, to simplify it, to organize it, manage it, to distribute it. This is an important part of managing 3D and getting 3D out to different users. It's something we're really proud of. It's something we'll continue to grow. Now I come back to how do we meet that demand. Really, honestly, generative AI is the way to do it.
This is why we're so heavily invested in generative AI. We are currently right now on the leading edge of enabling generative AI for the average user. Look, I'm a nerd. I'll be honest with you. I'm a 3D guy. I can go into Discord and get an image of anything I want in the world. It's like coding almost, right? What we've done is we've broken down that wall and we made it easier for anyone to be able to utilize generative. We're gonna continue to do that. We're gonna do that with 3D because let's face it, there's only one way that we're gonna be able to match that demand, and that's gonna be through generative. We're gonna continue to push the envelope.
We're working with some of the biggest companies in the world to match that need. It's really an amazing time. We're completely changing the way people create. We're completely changing the way people research. We're changing the way people see the world around them. It's really an exciting time. I'm happy to be here enjoying it with everyone else here at Shutterstock. With that, I'd like to pass it on to Candice Murray, our VP of Editorial, and she's gonna take you towards another adventure in content at Shutterstock. Thank you everyone.
Hey everyone. Thank you for joining us today. My name is Candice Murray and I'd like to tell you a little bit about our editorial business today. First, why does our editorial business matter? Why does this business matter to Shutterstock? Well, in addition to being a significant part of the content engine Paul talked with us about, we are covering the live events around the world, and it's a crucial part of being in this space. It allows us to really seize what we believe is an emerging opportunity for Shutterstock as a brand. Today we're covering 10 times more events than we did just a few years ago, and we plan to do this and continue to grow the number of events we cover every year.
Additionally, through our acquisition of Splash News last year, we are at the forefront of trending stories and breaking news to our customers in the entertainment space. Today I even brought friends. We have Miles Diggs, AKA Diggzy, who is here somewhere. Thank you. Diggzy will be with myself in the other room representing our Splash News and breaking entertainment business. We also have Andrew Walker, one of our staff photographers who will be in the other room with me to show you how we set up a portrait studio at live events. Bottom line, the impact to the Shutterstock brand through these breaking news and live event coverage is positive. The impact is super positive for this brand. Our photo credits appearing on every global media outlet around the world means Shutterstock becomes an everyday brand.
Our business is really a global for full service editorial offering. We talked a little bit on the left here you see the event coverage, our breaking news operation. In addition, it's important to note that as we capture these live events, we are also creating an archive of content that will be valuable for years to come to our clients. We'll look at a couple of examples of this in just a couple of slides. To ensure we're meeting our customers' needs, we offer an archive of over 73 million assets consisting of mostly exclusive content our customers cannot get anywhere else. This evergreen content is key as we partner closer with our clients to effectively tell their stories. Whether you are a documentary producer, scripted studio or streamer, or you're a breaking news outlet, access to an archive is important.
It allows you to illustrate a story that's happening today and/or couple archival content with that. Finally, we offer a number of services through our editorial business to ensure we are partnering with our customers, making it as easy as possible to work with us. We offer live assignments. We set up portrait studios as we have in the other room here today. We offer our content for commercial use, allowing brands and advertisers to use editorial content to promote their product or service, and we'll clear the third party rights for them to do so. It's also important to note every day there are billions of eyeballs on our content, depending on what you are consuming as a consumer of media. We're working with the biggest media, film and production companies in the world.
We'll take a look at some of the editorial content as it is published. I also wanted to call out the partnerships that we have with production companies, studios and streamers. Shutterstock is the creative force behind all things entertainment, including popular shows like House of Cards, House of the Dragon, excuse me, Ozark, The Crown, and Bridgerton. With millions of people reading, engaging, and watching, Shutterstock is connecting with those billions of people every single day. We know editorial matters to our business. We know it contributes to both our content engine and suite of services to support our clients' storytelling. We know it is a branding opportunity through photo credits globally. How do we continue to grow this business? How do we win in this space?
We'll continue to expand our contributor network all over the world with folks like Miles and Andrew. We will differentiate ourselves through partnerships with studios, specifically around live events, elevating our production capabilities. We will do more partnerships directly with talent and/or their publicists and agents, and we'll continue to scale our live assignment business. Now I'd like to take you through a few examples of how our Shutterstock Editorial business really does show up in the wild. Our brand name really does power the global headlines. We work with global brands to enhance and capture their biggest branded moments of the year. This is an example of our activation at Sundance. We partnered with multiple brands like Audible, Canon, to capture their bespoke activations for marketing and PR.
Through our strong partnership with Dotdash Meredith, we were able to secure a partnership with People and Entertainment Weekly to execute a portrait studio in the Canon space. The Shutterstock portrait studio at Sundance resulted in absolutely stunning photos that were published first exclusively by People and Entertainment Weekly, but then globally distributed to our media partners around the world. The Audible event you see here on the left is just one of many live assignments that we executed on during this week. We partnered again with People and Entertainment Weekly at the Golden Globes. In 2023, the Golden Globes were back and broadcast for the first time since 2020. We partnered with the HFPA, who hired Shutterstock Editorial as their house photographer and their PR team, with whom we executed a stunning portrait studio in partnership with People Magazine.
Working with Hunter Abrams, a renowned photographer, we captured portraits like this of every single celebrity that attended the show. Lastly, Super Bowl LVII. As we all know, the Super Bowl conversation goes well beyond the game. This is really an event where entertainment and sports collide. We were there and had access to it all. From Rihanna to the celebrities in the stands, we partnered with our clients beforehand to ensure they had all the content they needed to participate in those conversations. Our Splash News team are always tracking the stories that are trending, stories we know that our media partners want to talk about. We position ourselves as an extension to our clients' newsrooms. We are in regular communication with them around what really resonates with their clients. From breakups to makeups, new babies and living single again, we capture the moments and break this news.
Here is just one example of Machine Gun Kelly and Megan Fox exiting a counseling office supporting rumors of a potential split. The public's interest in relationships between public figures and how they cope once single again are breaking stories that people just love. Our clients love them. Their clients love them. The stories that are published here, they have legs. People have been talking about these two stories for weeks and for months, so we continue to partner with our clients to make sure they're telling that story as well. We partner closely with studios and streamers. I talked a little bit about that earlier. This is a really cool example, though, and I think speaks to our royal coverage and our royal archive.
Through the acquisition of Rex Features a number of years ago, exclusive partnerships in our continued royals coverage, we have one of the largest royal archives globally. We have one exclusive royal photographer, Tim Rooke. Providing our clients access to some of the biggest milestone moments throughout modern royal history enables them to bring their documentary storytelling to life. These are just some of the images that were used in the now notorious Harry and Meghan documentary that we were able to provide. Lastly, I wanted to share with you an example of how our archive can be used in a commercial way. We partnered with Dior on their new ad campaign, and their ask was simple: Find an iconic portrait of Christian Dior to highlight the heritage of their brand.
Within our exclusive The LIFE Picture Collection, we were able to find this stunning photo that you see here on the left. It is from a Paris fashion show that took place in March 1947. The end use on the right, billboard is an understatement. This appeared on the side of the Louvre for six months as part of the new ad campaign. To close, I thought I'd show you a reel that I believe brings our editorial story and our business to life. Thank you so much for your time today.
Good morning. I'm Meghan Schoen. I'm the Chief Product Officer here at Shutterstock, I'm really excited today to talk about our Creative Engine. You've heard from my other presenters today all about our content engine, this amazing global marketplace that we've built over the last 20 years with the world's best creative ingredients. It should be no surprise to you that when someone has a story to tell, when they have a piece of marketing collateral to create, they often start at Shutterstock. In the last few years, we've been very focused on how we can extend our relationships with these customers to not only be the place where they come to find their creative ingredients, but to be the place where they come to do their creative work.
Aiden is gonna come up here in a little bit and wow you with how we're doing this for some of the biggest brands in the world, and I'm gonna spend some time right now talking about how we do it for everyone else with our suite of intuitive creative design tools. The point is, at Shutterstock, whether you're Amy or you're Amazon, we have the ingredients you need, the tools you need, and the recipes you need to tell the best version of your story. When I say people start here, I mean 30 million, to be exact. 30 million people every single month visit Shutterstock with a creative intent. You heard from Dade that that creative intent may be to create a virtual world. You heard from Candice that it might be to tell a breaking news story.
It might be a florist who's coming to advertise their business on Instagram or a larger company who needs to put together a multi-channel, multimedia marketing campaign. The point is that they start at Shutterstock. The problem historically has been once they find that creative ingredient that they think they need, they immediately download it, and then they leave our platform to go do their work. We're changing this. We have a huge opportunity here to extend those relationships and keep them on our platform longer. We know that this is a huge opportunity because when they leave us to actually go do their work, doing that work isn't easy. Some would say it requires a Photoshop PhD in order to get this work done. They have to make hundreds of creative decisions. They have to choose colors, templates, fonts.
Even after all this time spent, they're not sure that what they've designed is actually going to work, it's actually going to resonate with their audiences. They finish the job, they've done their design, they download that design. They have to upload, download, redistribute across a variety of channels. You get the point. It's an arduous process. It's a lot of file sharing, time spent, and platform switching. Creative Flow provides a faster and more intuitive way to do your creative work. I'd like to introduce you to Creative Flow. When you're ready to take new content live, it's not unusual to feel a bit of anxiety. Self-doubt is a common roadblock to creativity. You can replace that fear with confidence with Creative Flow from Shutterstock. Imagine you want to promote your new blog on social media.
First, you need to find images that'll resonate with your audience. How do you know which one is going to be most successful? That's where the Predict app comes in, showing you which visuals are statistically predicted to help you reach your goals based on your audience and industry. If you want to add copy or other edits, the Create app allows you to customize text, remove backgrounds, and make other touch-ups directly on Creative Flow. Whatever visual you're imagining, you can make it into reality with just a few clicks. If you're not sure where to begin, we've got that covered too, with professionally designed templates to help you start fast directly on the Create app. Once your content is looking pixel perfect, you can organize and access the assets you find with the Catalog app, keeping track of all your content in just one place.
That means less time searching and more time creating, blogging, taking a victory lap. Whenever you want your social post to go live, just add it to your Plan app for a better sense of where and when it'll live compared to other campaigns. Kick your feet up and watch visits to your blog blow up. That's a quick A to Z look at Creative Flow's suite of powerful apps and how the entire creative journey can be completed in one place with 100% confidence. That sizzle did the hard work of showing you all the power of these tools. Hopefully you saw that in just a few clicks, anyone with any level of design experience can create something that looks professional.
Just to put this in perspective, these tools did not exist on the Shutterstock platform a year ago. Today, a quarter of our new customers who are purchasing products from us are experimenting with these tools before they make a purchasing decision. 10% of customers who once only came to us to find their creative ingredients are now experimenting and using these tools as a part of the work that they're doing on our platform. Creative Flow has the potential to bring new users to Shutterstock who have creative intent, but maybe never thought of stock first, and we can show them stock along the way. It has the ability to retain our existing customers because now, for the price of content, we've also given them the power to go create. Anyone? There we go.
The really important thing is that these creative tools don't just live in our applications. They're not a box software product that you have to go buy. The reality is every single person in this room can pick up their phone, take a photo, edit it within two minutes, maybe one minute if you're really good, and upload it to your social channels. We're taking that same intuition and applying that to how we're bringing these design features and functionality to our creative ingredient selection process. When someone is looking at an image on Shutterstock, we're going to allow them to crop it, understand what it might look like zoomed in or zoomed out, add filters, so before they do creative work in Creative Flow or elsewhere, they can rest assured that they found the right ingredient that's going to help them tell their story.
We're taking these design features and baking them into every element of the experience on Shutterstock. You heard Paul say at the very beginning of this presentation today that performance is the only thing that matters as a 30-year marketer. I've talked a lot about ease of use. These tools are simple. Anyone can do it, but it doesn't matter how simple it is if it doesn't work. This is the power of Creative Flow. Not only is it simple, but we combine that with a constellation of 800 billion data points that help us predict what creative elements are going to work. Before, during, and after the creative process, we can tell our customers, our creators, that they should use purple instead of blue if they're trying to drive top-of-funnel awareness.
We can tell them that the model in the picture should have both eyes open instead of her eyes closed if they're interested in driving click-through rate. We can even go as far as to tell our customers that they should show two hands in a photo instead of one if they want to more effectively resonate with women on Instagram. This is the power of Creative Flow. It's not just about ease of use. It's about marrying that with data and insights to drive better outcomes. We're not stopping there. You remember that awesome little spinning gear, flywheel visual that Paul showed you? This Creative Engine accelerates our content engine.
We are focused on bringing to life a reality where templates and campaign kits can become sellable ingredients in our creative marketplace, where we can leverage our community of 2.3 million contributors to help scale this product. That scale, coupled with our ability to predict what creative elements will work, can allow us to help customers scale their marketing campaigns. Just by telling us what they want to achieve, we'll be able to automatically recommend colors, templates, and accelerate, if not completely automate all of their marketing activities. This is our vision. This is Creative Flow. With that, I'm gonna relinquish the stage to Aiden Darné, who's gonna tell you how we're providing creative services to some of the biggest brands in the world. Thank you.
Thank you. Thank you, Meghan. Good morning. Good morning. Hello, everyone. My name's Aiden Darné. I am the VP and Global Head of Shutterstock Studios, a studio I'm super excited to talk to you all about today. Before we kick off and talk about how we've been really transforming and changing the production landscape, we're gonna take a look at a reel of some of our recent work. That is a banging tune, even if I do say so myself. You can license that on PremiumBeat if you'd like. The nature of Shutterstock and the core business here has enabled us to create a truly unique global production studio.
It's unique in the way that we can service our customers and go further and more beyond our competitors in servicing those customers, and it's unique in the way that we can scale global production. In the past two years since we've launched, we've produced global campaigns, tentpole moments, branded entertainment, product launches for the biggest customers in the world. Those customers have been coming into us from our brand direct, from our agency and our media and entertainment channels. One of the ways that we kill it, that we win, is via our global production footprint. We've designed a network and an infrastructure that means that we can turn production on quickly at scale and at an efficiency that's really hard to compete with across APAC, EMEA, North America and LatAm.
Of course, we have uniquely a real fantastic global creator network, a network of creators and contributors that's driving the studio forward, that's the driving engine behind it all, and that enables us to have boots on the ground anywhere. Some 2022 highlights. It was a big year. We shot in over 35 countries. We shot over 500 shoot days. We delivered over 30,000 unique assets to our clients and customers and lived to tell the tale surprisingly. But our AOV really, I think, is really the wonderful story here. We've now exceeded $100,000 a deal, $120,000 per contract. Which means we're going toe-to-toe with the established production community, and we're winning.
It also means that we're bringing in these massive whales of customers, these biggest brands in the world, in through studios, and then we're setting them loose in the rest of the Shutterstock ecosystem. It's a very complementary flow between Shutterstock to studios and studios to Shutterstock. Not just the global nature of the business, it's the versatility of the studio that we've designed. We have best in class award-winning work being delivered every day across all of these multi-discipline forms of output and creative. Photography, video, of course, we've produced all types of photography, high fashion, product photography, lifestyle for the biggest brands. Video, documentary, original documentaries, TV spots, scripted, multicam, live, TV commercials, you name it, we've done it, and experiential event and illustration as well.
It's this bottom row that's really exciting as Dade really eloquently spoke around the 3D and virtual production. We could spend a lot of time just talking about 3D in this space, but we have been absolutely killing it in the space of virtual production in 3D, which I'm gonna give a little demo through later on. Rather than me blabbering on, let's look at some work that we've made. CPG and beauty is a really fantastic vertical for us. We've produced global campaigns, global projects and programs for the biggest CPG brands in the world. They come in small, and they start big, and they just do not leave. We continue working across creative development strategy and custom production as well. Our network of top-tier directors and photographers is how we've continued to win within this vertical.
Auto, such a fun vertical to be producing all different types of branded content, branded entertainment. We have a documentary series on NBC Sports at the moment that's also been picked up by Sky in the U.K. around famed racing car driver Jimmie Johnson. We've produced social campaigns, branded content and TV spots for all different product launches for the Ford Bronco for Ford as well. Social media and tech, not only are we producing content for brands that are distributing on these channels, we're producing content for the channels themselves, utilizing and leveraging this amazing global network that we have. The network is also giving us amazing insights into what local consumers and audience members are consuming and reading and watching every day. We often know that more than anyone because we are more often than not that target demographic.
Healthcare and pharma, very similarly, we've produced all different types of campaigns for the biggest pharmaceutical brands. We can do everything from the creative direction all the way through to post-production as a complete end-to-end solution. Lastly, 3D. It's such an amazing new playground. Not that new. Dade's been around for 25 years doing it, but we've been harnessing all of these amazing in-house capabilities, and it's really Shutterstock Studios matched with the acquisition of TurboSquid, 25 years of 3D experience. That means we've been completely unique in the space across 3D strategy, development and fulfillment, 3D commercials, 3D animation, 3D digital twins for major brands. To the extent that if a brand doesn't have a 3D strategy for a campaign launch or a product launch or a tentpole moment, they might be missing a beat.
We're very far up the funnel and upstream in supporting in that. How are we winning just before we get into the XR demo? Global production expertise. We have an absolutely unrivaled global creator network, unmatched 3D, as I've been talking about, the most versatile production studio on the planet in-house. And in-house everything means that we can put more on screen, more ROI, and it maximizes the budget for the client. Of course, having the broader Shutterstock offerings available to us as a studio is also really, really unique. Now we're going to nicely segue into our XR demo and virtual production. If you will kindly take me join me on a flight through the streets of New York. Firstly, what is virtual production and XR? XR is this wonderful volume stage.
It is an LED wall powered by all different types of 3D content, 3D environments, 3D landscapes. We have a camera here shooting on this background. You can imagine if there was a prop or a car, anything in the foreground here, it means that at the click of a button, we can move through all different types of landscapes without having to travel, without the logistics, without the cost associated with all of that. This technology has been completely transforming how content marketers approach content marketing production. It means no flights, no tech scouts. Everything's built in pre-production. Post-production is done in two to three days as well. It is the most creative and economical version of production today. Oh, let's go to our loft. Pardon me, missed that cue already. Let's go to the loft. Great. Interior here at the loft.
To speak more specifically on a challenge that brands are facing today. Brands have huge amounts of content to make, very lofty content goals and very lofty content objectives, and they have equally high and lofty creative expectations, and rightly so. We do too. Expectations that they put on themselves or expectations that come from their bosses or the brand guidelines, whatever it is. Within today's economic environment, with budgets often being a little bit tight, those two can feel at odds. We need to make a crap ton of stuff, we need it to be great all the time. That's difficult. Through this technology, we can absolutely solve that solution. I'm gonna now start decorating this space just to show the flexibility of what we can do here.
In a click of a button, we can change backgrounds, we can add 3D digital twins. All of this is powered by our content, by our 3D content. We're now gonna pan round through to the kitchen and hold for a second. Yep, hello there. Just to speak to one more real-life example, a wonderful client of ours, a brand, a big brand, CPG brand, shoots thousands of shoot days a day. Pardon me. Thousands of shoot days a year, and they're making TVCs, lifestyle photography, product photography, product launches, and everything in between. They had a huge facility full of all different types of studios, bedrooms, bathrooms, living rooms, kitchens, and dozens of iterations within those different types of rooms.
What we did was created 3D environments of all of those studios and sets. They were able to close up that huge facility, and at the click of a button, we can move through a kitchen, we can move through a bathroom, a bedroom, whatever it is, scaling much, much quicker, way more economical. We're now just gonna show the flexibility of what I wanna do. Let's change the kitchen appliances. Let's change the facade, the cabinetry. Also to the extent, by the way, in serving global content objectives, you can see that we're gonna be looking out the window soon. We're gonna be, at the moment, in New York, but we're gonna go and go to Paris, to London, to Tokyo, wherever you wanna go. Being able to satisfy and serve the global audiences that every brand needs to today.
Now, of course, though, if you can be anywhere, why would you wanna be in a loft? I mean, I would love to live in this loft, but if you can be anywhere, why would you wanna be here? Let's move through to a desert. Here we're gonna showcase the power of controlling time, controlling the sun. Again, think of as a, as an example here, there's a car, talent, a prop, product, whatever it is, and we can click through these locations in the background, and we can affect the light, affect the sun, and control time. How often in production, I don't know if how familiar you are with production, time and travel and locations, the seasons, the weather, it's all a pain in the ass if it doesn't go well. This solves all of those issues.
I think, though, for the sake of perspective, let's add in a product, I think. Okay. A fridge. Okay, so we've got a fridge in the desert because why not if you can add a fridge in the desert? And you can see the light nicely reflecting off here. We can add more props and products in the foreground here to elevate and create the shot. So the power to control the sun. Now let's move, though, to controlling the weather. Let's go to a wintry storm. I don't love the landscape. Let's change the mountains a little bit. Cool. And a product of something. Fricking fridge. Okay. Someone really wants me to buy a SMEG fridge, which is fine. So again, the ability to completely tailor, adjust the landscape creatively is a dream solution.
It's a dream scenario to be able to go to creatives and say, "Ignore all the practical realities of travel, of shipping, of tech scouts, of reccees, science, gravity. Forget it all. We can do whatever you need within this shot physically, within the space, with these types of backgrounds." It's the most versatile economic version of production today, and it's powered by all of our content. Now to wrap up, let's go to space because why the hell not? It's pretty cool. yes, just to wrap up, you know, the unique nature of Shutterstock Studios, TurboSquid, the biggest library of 3D content, 25 years of 3D experience in-house, has been revolutionizing how production is done, and this is being completely adopted and used by the biggest brands in the world. With that, I'm gonna be handing over now to the wonderful Sejal. Thank you so much. Appreciate it.
Thank you so much, Aiden. It's super nice to see everyone here today. I am Sejal Amin, Shutterstock's Chief Technology Officer. And I'm gonna continue the exciting journey that the team has started today, but we're gonna take a bit of a left turn now and talk about data, a topic that I'm super passionate about and a topic that's important to Shutterstock. Just at the start of this, you already heard about our first flywheel, our content engine. Paul started with that when he opened our session together. We manage a large library across a variety of content types, including images, video, music and sound effects, and 3D. We have a content library that reflects 20 years of investment in content collection, asset curation and annotation, including a video library that is two times the size of our next largest competitor.
I'll be speaking today about how that content engine ingests, curates, annotates, and stores not only content, but a deep set of metadata around that content. We're finding that that deep metadata set is increasingly valuable to various customers in various use cases, specifically in the realm of computer vision model training. In short, our content and metadata is allowing us to develop new products and capabilities. You may not know what metadata is just yet. Very simply put, it's small data about the bigger data content or asset. By the time I'm done here in a few minutes, I promise you'll understand what metadata is. With that, let's dive in. While we have a massive store of assets, that content alone typically isn't enough. Over the years, metadata around our content has been collected and curated and has traditionally been used to classify content.
While we have a massive store of assets, that content alone typically isn't enough. Over the years, examples of metadata are information about the image, like a keyword description, or descriptive data about the asset. In other cases, it could well be custom curated data as well. In our current state, we have 185 metadata points about every image in our content library. That data is either automatically or manually generated, and I'll talk more about that on the next slide. This is super important because metadata has become increasingly important in recent years. For machine learning models to be useful, they need the context about an image or video, and keywords and descriptions provide just that. This is because machines understand visual data by learning from tagged keywords or labeled examples.
It acts as a ground truth about the image, and it's also super directly useful for text-to-image generative models as well, which you will hear about in just a few minutes. How does a piece of content get information rich? With that, let's embark on a journey together and follow a piece of content as it enters our ecosystem and how our management of it expands the depth and breadth of our content store. As I do this, I will take you through our key powerful ingredients, which can be described as the four Ps, but I'm gonna start with just the three. First is the people. Our contributor network is 2.3 million strong. In 2022, we added 300,000 new contributors, and they added 38 million new assets.
I mentioned on the last slide that we have up to about 185 metadata points. Our contributors provide about 50 of those metadata tags on average per asset. Now, they do it through technology we provide them. The technology provides them keyword recommendations which could describe the asset, such as location, color, or some descriptive keywords to go along with it. It assists them in their process in addition to the freeform content that they also add. This amazing contributor base is an ongoing global source of data. Next is our proprietary platform technology, which we've been investing in year in and year out, largely for search, but now it has become much more meaningful for us because it's creating a dual benefit. This tool automatically annotates content with additional context and metadata. Examples of this include technical quality, aesthetic scoring, and color analysis.
In addition to this, we also store EXIF data about an image, which is the information about the make and model of a camera, location, lens, and other various image metrics. This is all data that is automatically added, making up the other 135 metadata points. The final step in this is our human review process, which vets for legal compliance, alignment to our diversity and inclusion standards, as well as tagging consistency and accuracy. This is a team of over 150 people who have been doing this for 10 years. If not for automation, the size of this team would have to be many multiples of what it is today. Bringing it all together, this is our story on how we produce data about the content.
We have a depth and breadth advantage that no one else has, and with these three things put together, it means we have 110 metadata points about our assets. This only covers the three Ps, and I'm gonna leave you hanging on the fourth one for just a few more minutes while I take the time to connect the dots on the unique opportunity that people, platform, and process have created for us. Why does an augmented data content store matter to us? This content has created significant data feed and computer vision opportunities, opening up new monetization channels. In addition, we're able to commercially license this content by providing customers with the legal guarantees that they need. In 2022, we saw engagement with technology companies such as Meta, OpenAI, and LG.
As we turn the corner into 2023, the size and shape of our customers has evolved to a diverse array of companies representing technology, financial services, consultancies, and other small businesses. Our most immediate use cases are generative AI, content moderation, public safety, and interior design use cases. If you think bigger and broader and beyond this, computer vision, as an example, enables self-driving cars to understand their surroundings. It plays an important role in augmented and mixed reality, helping to determine locations on a display to place objects, and healthcare to automate the detection of cancerous moles in skin images or finding symptoms in X-ray or MRI scans. The point is that the possibilities are endless. We also have another secret weapon under our hood, the elusive fourth P that I mentioned at the start. Okay. Take a look at this.
As a result of our acquisitions, we have a constellation of 800 billion data points. Sit with that for a minute. 800 billion. Meghan and Paul mentioned this earlier in their sections as well. This data tells us how our content performs in the wild. With this proprietary data, we know what elements in a piece of content lead to higher awareness and higher click-through rates. The huge unlock for us is when we can create the connective tissue between this performance data and our metadata to not only share what is in an image to help train a model, but what's more likely to resonate in that image. That will enable the development of much more sophisticated and targeted algorithms.
As an example, it's one thing to know if there's a dog in a picture, but it's an even more powerful thing to know if a French Bulldog or a silver Boston Terrier in the picture will resonate with women aged 18-34 living in the U.S. in an Instagram post. Think targeted content, think more engaging, high-performing content, think creative storytelling, but most importantly, think higher impact. As these massive content stores come together, it will enable us to continue to differentiate our data in only ways we can. Now, this is just one data monetization example of many. I'm gonna invite Meghan to come back on the stage to continue the excitement that we've started here, and she's gonna take us into the world of generative and some of the new partnerships that we've built there. Thanks for listening.
Hi, it's me. I'm back again. I get the distinct honor to talk about this little tiny thing called generative. If you haven't heard about it's a dead giveaway that you probably actually shouldn't be in this room 'cause you haven't been paying attention to our earnings calls, and you've definitely zoned out during all of the Q&A for the last six months. Just you can leave. No, I'm just kidding. In all seriousness, why is there so much buzz around generative? One of the reasons is that fundamentally it allows small businesses with small budgets to punch above their weight class. That fancy guy, Aiden, showed you what we could do with six-figure budgets for some of the biggest brands in the world. We can put a SMEG refrigerator in the desert. We can literally take them to the moon.
With generative, anyone can go to outer space just by describing it. Just by typing something like "astronauts eating breakfast on Mars," you can get an image that's as high quality as the one that you see behind me. This is incredibly powerful because it means regardless of budget, you can come to Shutterstock and find all of the creative ingredients you need to tell your story. Need a picture of Rihanna? We got it. Need a 3D model of a human heart that's anatomically correct? Got that too. Need a picture of astronauts eating breakfast on Mars? Just describe it. At Shutterstock, we are offering creativity at the speed of your imagination for every budget, for every customer, for every story. This is generative. Need to get creative quickly, but have no time, no budget, no designer? No problem.
With Shutterstock's AI image generator, just type in a few keywords and you're done. Generate high-quality custom visuals in a matter of seconds. Perfect for all your marketing needs, from invites to social media posts, websites, and more. Shutterstock's AI image generator allows you to meet your business goals without needing extra resources. It's part of the Creative Flow platform, saving you time and money and delivering creativity at the speed of your imagination. Our content and metadata is the core of this generative business. We license properly to a variety of different companies and industries our metadata so that they can create products, experiences, technology. These aren't just transactional deals. What's very different about this business is that when we license our data, we are forming deep technology partnerships.
Our metadata helped train DALL-E 2. We're now the commercial arm of this technology. It's been fully integrated into our platform. We're not just integrating. We're differentiating and customizing, leveraging what we know about our customers, the creators, to make this a product that can actually help them tell their stories. We know that sometimes it can be hard to describe the thing that you actually wanna create. We've built in a style picker that allows these visual creators to actually look at something and just select it to communicate with the model what they want their outcome to look like. They can use a combination of these visual inputs and keywords to actually generate new content. I think you heard from Paul, hopefully it resonated, that we are a global company.
One of the things that we've done is make sure that this technology is available around the world to all of our customers. We've translated it into the 21 languages that we support, so anyone in their native language can use this technology on Shutterstock. We also know that prompts can be hard to come up with. There's actually a whole industry, a whole new profession of people called prompt engineers. We don't want our customers to have to be prompt engineers to get creative ingredients that work for them. We've put ourselves as a Wizard of Oz, the magical person behind the screen, who can take something as simple as a single word, like the word cat, and actually generate creative outcomes, leveraging our proprietary technology to translate that into something that these models understand.
Last, but certainly not least, we've leveraged our 20 years of experience in building search algorithms to build bias mitigation into our returned results so that we can show diverse results even with generative content. We're following our proven playbook. For 20 years, we've made the connection between content and contributors. We're doing the same thing here. When somebody generates and licenses a piece of content on our site, we pay a royalty to a contributor fund so that our contributors get compensated. Our customers come to us because they want the creative ingredients they need to tell their story. They expect us to innovate, and we're doing just that. Our contributors come to us because they want ways to make money. They expect us to continue to find new avenues for them to sell their content. With this product, we're doing just that.
How's it going? We launched this product four weeks ago. You've probably heard Paul say that this is very much the first pitch of the first inning. Here's what we do know. We have 30% more people giving us their email addresses than before we had this product. That's 30% more opportunities to tell the Shutterstock story. 30% more opportunities to take someone who's interested and convert them into a paying customer. 10% of our customers who have active products with us, they have a subscription, they have a pack, have used this technology. We've just added value for no incremental charge for our existing customers. What we know after 20 years of experience is when you add value, it's highly correlated to retention. Amazingly, users are generating almost two million images on Shutterstock every single week.
When we talk about the size of our library today, we talk about it in the hundreds of millions. This can very quickly become a conversation in the billions. Billions of pieces of content for our customers to discover, new monetization channels for our contributors. The content engine accelerates. We happen to be moving very fast in this space. We launched this product at the end of January. About a week later, we opened it up to our enterprise customers for opt-in. On February 9th, we did this little thing where we integrated generative search results into our regular image search. This means that customers don't have to decide what platform they're gonna go to before they actually generate or find creative ingredients.
It means that they can come to a single destination, simply tell us what they're looking for, and we will show them results direct from our 2.3 million contributors, as well as all of the content that has been generated on our site. We are the only business in the world doing this today. We've also integrated feedback loops. You see this little like smiley, happy face guy right here? Seems like a really small user experience enhancement, but what it allows us to do is actually have our customers give us feedback. They can tell us when the content that they're generating is actually worthwhile and useful for them. This data directly feeds into our models so that we can continue to improve what is generated and make sure that it suits the needs of people who actually want to work with this content.
Coming very soon, I think any day now, we're gonna be adding additional functionality like zoom out and variations. When someone generates a piece of content, it might be almost right, but not quite perfect. Now on Shutterstock, with a single click, someone can find something that's almost right. Click a single button and either change the view of it and/or create additional versions. I mentioned in the style picker, we have about 30 styles that we support today. By the end of March, we'll have over 100 supported styles on shutterstock.com. This isn't where we're stopping. I spent a lot of time today talking about the power of generative to create content from nothing, but this technology also has the power to augment existing content.
Imagine a world where someone can come to Shutterstock and find this really happy looking image of this dog that one of our contributors uploaded. Simply by selecting a portion of the image and adding text as simple as add a dog bone, they can transform the image before they download it into this perfect little dog with that lovely dog bone in its mouth. This technology can not only grow our content library, it can also bring new life to the powerful content engine that we've built over the last 20 years. With that, I'm gonna hand it over to our esteemed CFO, Jarrod Yahes, to take us through an equally exhilarating conversation about our financials. Thank you.
Hello. Incredible stuff. Generative AI, metadata, data partnerships. Generative AI is getting really, really good. Good that it can even generate a replica of a tie from the mid-90s in purple Brooks Brothers on a CFO on a stage. Incredible stuff this generative AI is capable of. Fantastic. I know two things that are even cooler than generative AI and cooler than metadata, and that's revenue and EBITDA, and that's what we're gonna speak a little bit about. Today I'm gonna be speaking to you about four elements of Shutterstock's financial model, taken together that make for a powerful combination, and we believe ultimately the potential for outsized returns for our investors. It all starts with our content engine, a revenue base that's increasingly subscription, leading to higher levels of net revenue retention, and two powerful channels, our e-commerce channel and our enterprise channel.
We continue to operate profitably. We've paved the road with expanding EBITDA margins through G&A leverage as well as disciplined cost management as an executive team. We're strong capital allocators, and we're gonna talk to you about how we return value to shareholders through dividends, buybacks, and mergers and acquisitions. Lastly, we're positioning ourselves for significant increases in scale in both our Data Engine and our Creative Engine. We think these two new flywheels have the potential to significantly change our company over the course of the next five years and fundamentally transform our business. Taking a little bit of a look back at Shutterstock over the past 10 years, we've continued to steadily grow our revenues. Over the course of the past three years, our revenues have grown at 8% on an annualized basis.
In 2021 and 2022, we experienced accelerated growth in our enterprise channel, with enterprise channel growth experiencing 15% growth last year alone. Strong new product introductions like SMB Flex and 40% growth in strategic solutions, including our data partnerships, our studio solutions that Aiden spoke about, as well as our creative solutions and editorial solutions are what drove that 43% growth. In 2021, we experienced a strong rebounding growth post the pandemic in our e-commerce business, where we rolled out Flex products, which allow our customers to consume images, music, and video all in one simple subscription. We also rolled out our Creative Flow product. As Meghan mentioned, these products didn't exist two years ago. Today, they form the bedrock of the value proposition for Shutterstock's customers.
We think about our TAM, most everyone here knows that the traditional stock TAM is about a $6 billion end market, growing between 5%-7%. We believe that our industry will return to those growth rates. As we think about Shutterstock and the way we're building up our business, the evolution of our content engine into faster growing modalities such as video, 3D, and music, and our focus on our creative and our Data Engines, we believe that we are positioning ourselves for above average industry growth over the course of the next several years. How are we gonna get there? We are evolving our business surely, slowly, deliberately towards a subscription model. Our subscription revenues are growing at three times the growth rate of the company as a whole.
Our subscription revenues have increased from 34% of revenues to 42% of revenues. We now have four times the number of subscribers that we had just three years ago. We expect these trends to continue. We expect them to continue in both of our channels, enterprise and e-commerce. What's driving the growth in subscription? Why is this taking off? Why are we deliberately focusing in this area? Number one, we are encouraging our clients to sign up through subscriptions through marketing, merchandising, and pricing. We're putting in place the right incentives. Number two, in enterprise, we're seeing the customer take-up of our multi-asset subscription products being exceptional. As we disclosed recently, over 20% growth in our subscription bookings in enterprise. A tremendous outcome.
We know product market fit when we see it, and we're seeing that in enterprise for some of our SMB subscriptions that we brought to market. There's two opportunities that we see in front of us vis-a-vis our subscription revenues. Number one, with our recent acquisitions of Pond5 as well as TurboSquid, while garnering us a leadership position in video and 3D, those businesses are largely transaction-based businesses. Were it not for that, our subscription revenues would be 4% higher as a percentage of revenue than they even are today. There are tremendous opportunities to drive subscription product SKUs in both TurboSquid as well as in Pond5 that we look forward to executing on in the years to come. Secondly, we recently introduced a lower-priced $10 Creative Flow+ tools for a subscription. This was rolled out just this past summer.
Today, we have approximately 300,000 Creative Flow+ subscribers. We think the opportunity to sell customers $10 a month Creative Flow+ tools for subscriptions is into the millions. Getting customers to our site who do not think stock first, but think tools first is critical and part of the path to developing our Creative Engine as well as our Data Engine in the years to come. Why does subscription matter? Why are we deliberately leading with it, and why are our customers adopting it? Ultimately, we want to establish deeper and more sustainable customer relationships. Our subscription revenues today have 7% better net revenue retention in enterprise and 3% better net revenue retention in e-commerce. Today, our higher retention rates in enterprise are having a compounded impact on our revenue growth.
When you see the 15% revenue growth, it's because our enterprise subscription revenues are growing more than 1.5 times faster than overall enterprise revenue growth. Our e-commerce subscription revenue growth is growing faster by three times. Just as a point of reference, if we're able to take our subscription revenues and move them from 42% of revenues to 50% of revenues, or an 8% improvement, that's going to improve revenue retention at Shutterstock by 4% and ultimately improve our revenue growth by the same amount. You understand now why we think subscription is so powerful and why we're continually moving in that direction. I'd like to take a minute and speak to you a little bit more about our enterprise channel.
When you think about our enterprise channel, you can think about the strength in terms of the four Ps: people, product innovation, packaging, and predictability. In terms of people, we have a global sales team of 250 individuals adept at having strategic conversations with our customers to identify their pain points. We're incentivizing that team to upsell and cross-sell the range of strategic solutions that we offer, including editorial and studios. It's obviously resounding. Our strategic solutions at enterprise are growing in excess of 40%. Product innovation. We've launched a series of new products in enterprise over the course of the past several years, including studios and our data sales businesses, which have been hugely successful. We recently doubled down on editorial with our acquisition of Splash News, which is now our flagship brand for candid celebrity entertainment content.
Strategic solutions in enterprise, which includes Studios, Editorial, and data sales, were up 43% this past year and now almost represent 20% of enterprise revenues. These are the oars that are in the water that are driving the enterprise channel to the 15% growth that we experienced last year. From a packaging perspective, we flexified our products to increase our customers' consumption of different asset types like video and music. This has been particularly successful. Three years ago in enterprise, video and music were 18% of enterprise revenues. Today, this past year, they were 25% of revenues. Lastly, there's the predictability of our enterprise channel. Our revenue retention in enterprise is already in the mid-90s, even though subscription revenue is one-third of the total revenues in enterprise.
We believe there's a tremendous opportunity to take enterprise revenue retention above 100% in the years to come. Show me a business with revenue retention in the mid-90s, focused on product innovation, with a strong motion for upsell and cross-sell across content types and services, and you quickly understand why for us, enterprise is a crown jewel asset at Shutterstock. I'd now like to turn and speak a little bit about our e-commerce channel. E-commerce is 60% of the revenues at Shutterstock, and we have a four-point plan to get it back to growth.
We're adding value to the customer with new and innovative content types like generative AI that Meghan spoke to, and PixelSquid, which is a new 2D, 3D format that we'll be rolling out to our customers in the next several months, as well as providing them our Creative Engine and suite of self-serve tools to allow our customers to create with confidence. Our product marketing and engineering teams are working together to ensure the best product market fit on a global and localized basis. We're increasing experimentation globally, broadly, and at scale in order to test new product SKUs, new user experiences. Lastly, our marketing team is working hand in glove with our new e-commerce head, Jon Kane, to ensure we have a world-class acquisition and retention engine.
Investing and experimenting across multiple channels, including SEO, CRM, affiliate, and social, we think we have a tremendous opportunity in front of us to attract more customers and retain them with better levels of customer satisfaction. I'd now like to pivot away from revenue and talk to you a little bit more about EBITDA. This executive team has a strong track record of achieving EBITDA margin expansion, having more than doubled our EBITDA in the past three years. We are in an amazing spot with our margins today, with EBITDA and free cash flow margins that would be the envy of many companies. As an executive team, we have our hand on the dials. We deeply understand our cost structure. In pursuit of EBITDA margin expansion, we've pursued several initiatives over the past years that have taken our EBITDA margins from 15% to above 26%.
We've optimized spend on product and technology, setting us up for a long-term success and investing in the most important initiatives. We've optimized marketing spend, including calibrating our search marketing spend, our affiliate marketing spend, and diversifying into newer channels with higher ROI. We've restructured our royalty payouts. We further automated our content ingestion process that Sejal spoke about, improving our gross margins significantly. I'm proud of this team and all we've accomplished in terms of driving an exceptionally high margin business. We think there's more to come. Going forward, there is tremendous opportunity for further margin expansion at Shutterstock. Our G&A at 16% of revenues is high for a technology company approaching $1 billion in annual revenues. We enjoy natural operating leverage as our business grows and scales.
More importantly, we have discrete and ongoing cost reduction initiatives that add up to over $20 million of annual cost reduction potential. As a team, we are programmatic and deliberate about cost reduction. We are not reactive. Discrete cost reduction initiatives that are in focus for us this year in 2023 include lease footprint reductions, software consolidations, vendor and professional fee reductions, and extracting synergies from our acquisitions of Pond5, TurboSquid, and PicMonkey. Ongoing cost reduction initiatives each and every year include content acquisition, royalty optimization, location strategy, finance and operations process automation, reducing our indirect tax leakage, cloud optimization, and content ingestion process improvement. As you can see, we keep extremely busy in terms of driving our profitability and working towards margin expansion. Margins are important, ultimately, our goal and our objective as a management team is to allocate capital and deliver returns for shareholders.
Today, we believe we've established a track record of prudently allocating capital internally to research and development and capital expenditures and content acquisitions, but also returning capital to shareholders. Our target, our goal, is to combine the growth in our business, margin expansion, returning capital to shareholders through dividends and share repurchases and mergers and acquisitions to deliver mid to high teens compounded return for shareholders on a long-term basis. We believe that we're extremely well set up to deliver on these objectives. We have zero debt, and we have the financial flexibility with significant access to capital to invest. We just increased our dividend by 13%, and we're on the hunt for mergers and acquisitions with valuations coming into line and great opportunities that we're seeing in the marketplace. Let's turn to M&A and think about that for a minute. Shutterstock is going to be acquisitive.
We've developed a tremendous track record over the course of the past several years. Our goal is to get our content engine into faster-growing content types, to accelerate automation for our customers in our Creative Engine, and to enhance the metadata in our Data Engine. Look at the work we've done so far. It's amazing. We acquired Pond5 and TurboSquid and Splash to make ourselves the leader in video and 3D and entertainment and trending content. We acquired PicMonkey to pull forward our Creative Engine product roadmap by years. We acquired Pattern89, Datasine, and Shotzr to create a massive performance data constellation that we believe is gonna be a major mover for us in our vision of making creative more effective for marketers. We've built a systematic approach to M&A. We look at market consolidation opportunities, acqui-hires, and technology bolt-ons to accelerate our product roadmap.
Going forward, we're gonna be looking at new content types that are fast-growing and scale. We're gonna be looking at enhancing our metadata. We're gonna be looking at tools for creative automation that make the journeys of our customers quicker and ultimately more fun. Let's talk for a little bit about ultimately data and some of the large data partnerships that we've been announcing. Lately, our commercial strategy has been focused on getting the data flywheel turning. We've been landing large partnerships. We've been using the revenue from those large partnerships to pay to our contributors into the contributor pool, which will in turn facilitate us getting more data and then landing larger and larger partnerships.
We're scaling our Data Engine, leveraging our two decades worth of content and data and existing customer relationships with some of the largest players that are out there, including Meta, LG, OpenAI, and others. The unit economics of the data partnerships that we've established are comparable to the unit economics in our content engine, with 65% gross margins with payments into our contributor pool. The data partnerships also have the potential for significantly greater profitability due to lower SG&A costs. We believe that there's the potential for 15% higher EBITDA margins in our data partnerships as a result of the lower overhead and selling costs associated with these partnerships. Two big opportunities that are in front of us with respect to data partnerships. Number one is access to proprietary technology. We realize this is going to be a fast changing landscape. We don't have the R&D.
We don't have the capital to build all the AI that is gonna be needed to deliver to our customers, yet we have a truly unique data asset that we can use and leverage in order to get access to that AI. Secondly, there's a tremendous opportunity to make our data partnerships and the revenues associated with those data partnerships and the revenue stream more SaaS-like in nature. Today, the revenue streams are more like the old perpetual license model, but we're seeing more demand for other content types from our customers, custom datasets, data curation and hygiene services. Our customers are basically telling us that fresh, ongoing data is mission critical to accomplishing their objectives, and they want us to supply that data for the long term.
Let's take a little bit of a deeper dive into some of the data partnerships and some of the trends that we're seeing in these deals. We are excited. The data partnerships are getting larger every year. They're getting longer dated, and the pipeline is broadening. Let's look at the evolution of the demand environment. In the beginning, and by the beginning, I mean a short 36 months ago, these were one-time transactions. Now, our partnerships are more typically three to five-year long-term contracts, and they're becoming much more multifaceted. Why are the partnerships getting larger? Ultimately, the budgets are there on the side of clients. They're looking for different types of content. They're no longer just looking for image metadata. They're looking for metadata associated with video and 3D and music.
The use cases are broadening and now critically proof points exist that there are AI products being brought to market that are actually commercializable. The customers know what they want to do with Shutterstock's metadata when they procure it. There's an applied use case and hence the budgets are opening up. Why are we winning? As Sejal mentioned, there's massive scale in the content and the quality of the metadata. There's variety in the content type. We passed the legal due diligence test. It's extensive with respect to our ability to legally convey metadata and then compensate our contributors for the sale of that metadata. Lastly, Shutterstock is approaching these deals in a true spirit of partnership with respect to marketing, commercialization, and product investment.
We're gonna be a great partner for some of these large global technology companies to help them commercialize these nascent offerings and bring them to market through our distribution channels. I'd now like to turn from metadata and our data partnerships and talk a little bit about our commercial strategy around generative AI. Our commercial strategy with generative AI is to make it simple and pervasive. We fully integrated generative AI into all of our existing products. Our goal is to accelerate consumption with our pack customers, improving their net revenue retention as they buy the next pack, as well as improving engagement with our subscription customers, again, improving net revenue retention. We want to encourage broad experimentation by our customers, we don't wanna put a ton of restraints in the way that we can use our generative AI.
Today, the unit economics for generative AI are consistent with the remainder of our content engine because we're taking the cost of our AI partners and our contributor royalty pool, and we're centering it on the current gross margin profile of the business. There's a few big generative AI opportunities that are out there that we're focused in on. Number one, hand-holding our enterprise clients through this journey. There's a tremendous opportunity for asset assurance services and for enterprise clients to leverage our studios and our Creative Engine in conjunction with generative AI. There's also an opportunity to reduce our content supply and our ingestion costs over time. Once content is previously generated by a customer, it enters into our library. We don't have to pay for that content ingestion, and we don't have to pay that AI partner again for that to actually expand our library.
As Meghan mentioned, we now have in front of us the ability to have an enormous content library of 650 million assets expand and start having conversations around billions of content assets. It inspires us each and every day. To close out, as the data and Creative Engines ramp up at Shutterstock, we are going to evolve such that in five years' time, the creative and Data Engines in Shutterstock will represent more than half of our business. Over the next five years, when customers come to Shutterstock, and they pay for a subscription or a packed product to Shutterstock, what they receive and what they value is going to completely change from the premium content stack that they receive today to a set of creative and workflow tools facilitated by AI.
This is a massive evolutionary change for our company, and myself and this management team could not be any more excited for what's to come. I'd now like to welcome Paul back onto the stage for some closing remarks. Thank you so much.
Thanks, Jarrod.
Thank you.
What a journey we've been on. We started with breakfast in New York City, and we took you to have breakfast on Mars, didn't we? We've taken you to the movies, and we've dropped you in the Metaverse. You've suffered through our hijinks of dropping refrigerators in the desert, but we showed you that we can control the sun. Now at Shutterstock, we can move mountains. You've seen that we're friends with both Rihanna and the royals, two people you wouldn't think in the same sentence would go together. You now understand how all of those things power our business model. You've learned about our content engine and how that flywheel is spinning and growing and scaling. You've now also learned about the power of our Creative Engine and how that flywheel is spinning and adding value to our customers.
You've also seen now how the Data Engine is the next horizon for our business, and it's spinning on its own, and we've hinted at the power of performance and what we're able to deliver for our customers. I think you now know what we know. This business has a powerful history and a very bright future. Thank you all for all of your attendance, for your engagement, for listening for the past two hours, and thanks to all of our speakers. We've got time now to flip to Q&A, and I'd love to have Jarrod come back up on stage and we can flip to Q&A. We've got a hand up already, Chris.
Hi, all. Nat Schindler from Bank of America. A couple questions. One, if you look at what generative AI is, it's really just tools. It's making what was difficult easier for and thus opening up to more people. As you guys brought up, it's like giving the power of big budgets to small budgets. Have you done any work on what the impact that would be? Looking at any other industries where tools like this have been developed, maybe not to this extent, but similar steps in this direction, making it easier, thus giving some sort of sense of how much more market expansion this creates. I have a second follow-up.
You know, I have some thoughts on that, Nat, and then I'll pass it over to Paul. I think for us, when you kind of think about what has made Shutterstock special over the last several years, it is massive scale, and making things easier to consume has generally been a very good thing for our business. Breaking down barriers to people consuming stock and making it easy and simple with a straightforward commercial contract and broad-based use cases, this is the bedrock of our company. When you think about generative AI, and you think about it as a tool, make it easier for people to create content seamlessly in two seconds, our hope is that does increase volumes, it increases engagement, it increases experimentation.
Anything for us that ultimately gets people out there consuming content, experimenting with content, using content, we believe is a good thing for our business long term, even if the unit price may actually decline over time. We think we're extremely well-positioned to benefit from a world where volumes in the queue continues to go up, even if the frictional input costs or the P continues to go down. We think we're the winner in that ecosystem. For us, as these tools emerge that make it easier for people to consume, easier for people to create, lower costs with lower barriers of entry, I think we like that world. I think it makes it more fun to be a creative, and it makes it better for us to be a partner for those creatives.
We're, you know, we're excited to support it, even if it lowers the barriers to entry. We think it takes up volume ultimately in the long term as well.
Great. Thinking about the scale on that, would you scale this similar, the creation of generative AI, being able to make whatever you want, instantly by using large language models and just typing in some text string? Is that similar to the transformation that just simply the whole idea of online marketplaces for stock photos created for promotional and marketing materials 20+ years ago when they suddenly were able to sell you not anything you wanted, but a whole lot of what you might have wanted? Is that scaling the transformation correctly? One more follow-up.
Maybe I'll jump in there. On that, I think it's an accelerant, right? If you think about something Meghan said is very important, AI as a standalone generative is critical. I think there's a massive opportunity in just in actually being able to create new content. Two, you mentioned tools. I think that's an accelerant for tools. We're literally living our mission, which is simplifying and supercharging. Then 3rd, I think there's a combination of existing content and generative AI, and putting those two together unlocks the value that we already have in our existing infrastructure, and it accelerates the entire business going forward. We see it as, yeah, I think this is the next thing.
This is the kind of gasoline on the fire to drive our business forward 'cause we're already synonymous with the creative place where people come to get all of their assets and tools. I think it's got a powerful next step for our business.
Finally, one last one.
All right. Sneak it in.
Sorry.
Yeah.
This is a jab at Jarrod.
Fire away.
Hold the mic then for as long as you'd like.
Is Jarrod letting you do what you need to do with all this push on margin growth and all this push on 2% to 4% returns on share repurchases and dividends? Is that giving the opportunity that this represents enough juice? Are you getting the fuel you need?
Yeah. That's a great question, and there's a bit of a trick question, right? 'Cause some of my past comes from growth engines, and I won't take the bait completely. Look, we're here to grow the business for our shareholders and for our customers, so we're gonna be opportunistic about what we do. Jarrod doesn't have handcuffs on me. We're actually great partners in deciding what we ought to invest in. He mentioned just moments ago that we're always going to be acquisitive. We've added great value to the business by buying things. We also invest in our core business. We're marketers. I'm a marketer in my DNA. We'll grow and scale the business, and we're gonna be really prudent about doing that. Do I see a world where margins, you know, go to 50%? No.
We're not driving that business. Do I think we're on a, you know, a tear where we continue to add margin value to our business while we're growing our business? Sure. I feel like we've got all the dry powder to drive our business forward and grow it. Please.
Thanks. Lauren Schenk, Morgan Stanley. Maybe just on the computer vision deals. You mentioned on the earnings call that they're, you know, at 20+ sort of in your, in your pipeline already. Are those signed on the dotted line deals? Are those, you know, on the come? Maybe give a little bit more color on that. Could you give us some examples or case studies of how you think you can further monetize deals that have already been signed? The ones that you've already booked the revenue for, you know, where's the opportunity on the upside there?
Yeah, absolutely. This, you know, I think we stand here before you today really excited about this opportunity. Hopefully that kind of came out. The 20 deals are signed deals that we've executed on over the course of the past three years. At the onset, some of these were small experimental deals. Think an AI lab that had some excess budget of $100,000, wanted to get some metadata to enhance some of the work that they were doing without a discretely defined commercializable product that they were working towards. That has completely turned on its head over the course of the past year.
Ultimately, you know, today, the groups that are coming to us have defined products that they're working towards that are commercializable products, and because of that, they have real budgets associated with it. They know that they may have $5 million-$10 million to spend on metadata. You know, there couldn't be any more of a sea change between the world that we were living in 24-36 months ago as compared to today. Feel very good about the pipeline. We also, quite frankly, feel very good to the second part of your question about our ability to expand existing relationships that are there.
Many of the deals that we've announced were for select parts of our content library. What we've seen, we've seen this now in multiple instances with very large technology partners, is once they have some of our image data, they're looking for it all. They also want video, music, and 3D data. It's extraordinarily important. What we're seeing is, we may sign up for one, and then they derive the value from it, and then they move on, and they move to others. There's sort of two discrete opportunities that are there to land and expand. One is the content type and the scale of the content that we're selling them. Also, we're being asked about what I would call custom curated data sets.
Ultimately, our customers want model outputs that are brand safe. They want images and content from their models that are representative of the world around us. We understand what DEI looks like from a model training perspective. We understand the geographic nature of some of the data that we purvey. Ultimately, a generative AI model and what the outputs are may look different in Asia than it does in Europe, than it does in the United States. Being able to segment that data by location is also extremely important. The curation, customization, of that data is something that we're seeing customer demand for.
Great. Maybe just one follow-up. Is there an opportunity to monetize or license some of the tagging technology and sort of helping others build on the metadata side?
Mm. I would say the answer is yes, there's a possibility in doing that. We view that as, you know, a differentiator in our own business, and that's something I think that's very powerful. We've been doing, as you heard from Sejal, for a long time. We add value to every single piece of content. I think we're gonna keep that proprietary for the near future.
Thank you. Youssef Squali with Truist Securities. Couple questions. I guess just following up on the generative AI data side. There's obviously a lot of hype for good reason. From the outside looking in, we can see kind of the progression, but we don't really have great visibility into the impact of it on your numbers. It's all in within the enterprise, right? Number one, are there any plans as this grows fast or grows bigger for you guys to kind of provide maybe more visibility into that? The other question we often get is around the lumpiness potentially of it.
My understanding is that the way, you know, these deals get recognized, that you recognize the 70%, 80% of it up front, and then there's a three to five-year continuum of revenues, but at a much lower, you know, rate. One of the fears is at some point we're gonna be done with recognizing the lump sum, revenues may kind of roll over. Maybe can provide a little bit of insight into that, and I have follow-up.
Sure. To your point, Youssef, our data partnerships revenue is disclosed in our enterprise channel for external disclosure purposes. You know, one of the things we spoke about on the presentation today was a quarter of that business is growing in excess of 40%, right? The data partnerships are an intrinsic part of that. Studios is growing exceptionally fast, editorial is growing very rapidly, and data partnerships is explosive this past year. Right now, from a guidance perspective, because this is a new offering for us, unless deals are landed and in the bag, we're not putting that into our guidance. These are large, lumpy deals. As and when they sign, we'll update our guidance accordingly.
You know, ultimately, as this business grows, if this business becomes 10% of our revenues, which we think it does have the potential to become 10% of our revenues over time, it is highly likely that we'll call that out for investors and sort of break that out separately and regularly. You know, we feel excited about the pipeline. We're excited about the demand signals from the existing clients we've signed. Nothing says that a business has more opportunity than a satisfied existing client who wants more of what you're selling. To your point on the lumpiness of the revenue, you know, ultimately, we wanna do what's gonna provide value to our clients.
Fitting a square peg into a round hole to try and even out the revenue and make it more SaaS-like in nature, that's ultimately not the starting point for the way we're thinking about the go-to-market. It's solving the customer's pain point, making sure that we're delivering them something of value, and that we're using it as a land and expand opportunity. My own view is that while the initial burst of revenue is helpful, building up a maintenance stream of revenues over time is also a highly sustainable, durable business model that we've seen work very well for software companies for many years in the past. While I'd prefer it if it were a SaaS-based revenue stream, building up a highly profitable long-term maintenance stream of revenue is another way to achieve many of the same objectives.
I think we'd be excited regardless of the turn that it kinda takes.
Just on the margin implications of that. I think you guys are committing to showing progression in margin, year on year. Is there any risk that if that revenue stream were to kind of flatline or even go down, that those 50 basis points of margin improvement annually could be at risk?
You know, we are not predicating our margin expansion on a rapid acceleration of our data partnerships business. Ultimately, as I mentioned, there's $20 million of potential cost opportunity. We attack this very regularly, very systematically across multiple functions in the company. When you take a step back and you think about a technology company approaching $1 billion of revenue with 16% G&A as a percentage of revenues, it's not that difficult for us to get that 50 basis points of margin expansion. We don't necessarily need to cut to be able to get there. We need to grow, and we need to cap the growth of our G&A costs in order to achieve those objectives. You know, I'm not saying that it's not easy for us to continue to do this.
It is hard for us to be able to continue to do this, but we think we have the tools in our toolkit with the natural leverage built into the business model, which doesn't necessitate that we need the data sales business to be able to achieve those objectives.
That's great. Thank you, and thank you guys for organizing this event.
Thank you for coming.
Hey, guys. Thanks for taking the question. Bernie McTernan from Needham. I thought the comment that eventually the data or in Creative Engines would be over 50% of revenue, in pretty short order was a really interesting comment. Just any other, you know, further detail you can share on that, like what the world looks like? I'm assuming that's while the content engine is also getting back into growth as well.
Yeah, maybe I'll start. This dovetails off of the, the comments from Youssef a moment ago. He referred a couple of times to when the, when the data revenue dries up. While certainly that's possible in our distant future, we really believe it's early days in our long-term, you know, data training, relationships with some of the largest companies. Jarrod mentioned the value of the pipeline, and we see that the value is increasing. The conversations are happening now. They're not ending now, they're happening now. We're getting, you know, we're getting very, very positive traction in this space, and I think we're enthusiastic about what the data business, the data relationships, the long-term relationships will do for our company. I just wanna be really, really clear on how this works.
No models work where we're all trained up and now we're done. It doesn't work that way. Models have an insatiable appetite for more data. If you think about the way this world is going, why are we not all driving autonomous cars? It's because the last mile is freaking hard to get right, and that last mile requires continuous fresh data and more modeling so that, for instance, cars can tell the difference between a stroller, a dog, and a manhole cover. That's really hard for models to understand that, and that's just one of the verticals. Sejal talked about the multiple verticals that we're enabling, and there are a lot more than that.
You know, it's a broad answer to, yeah, we see this data, long-term data relationship business, the leveraging of our metadata, working with our large-scale partners on long-term relationships as the first pitch of the first inning, not the end of this.
That's helpful. Then a follow-up question. You guys have provided some really nice qualitative comments on the pipeline, but we get a lot of investor questions on what the quantitative aspects of that pipeline is. I was just wondering if there's anything further you'd be willing to share or even directionally if that pipeline for computer vision deals is bigger than the 20 deals you've already signed.
I mean, I would say something at the risk of really getting on top of my skis, which is to say that there are individual deals in the pipeline which are larger than all of the deals cumulatively combined from 2020 and 2021, right? 2022 saw a significant step up. The types of deals that are out there today, we are having conversations around $5 million and $10 million deals. These are deals where not only are they around data, but they're also around commercial partnerships. That's one thing I wanna be crystal clear.
We view this as a monetization and a commercial opportunity, but we also view this as a way to build sustainable competitive advantage for our business by getting proprietary or advanced access to technologies that we would not be able to build ourselves with our current capital and R&D budget. When you have something really rare and really valuable, there's two ways to monetize it. One, you can sell it for the highest possible price. Two, you can compromise a little bit on price and get access to something that you view as even more valuable long term. I think we are taking that second approach. We view this as really the potential to expand into some really exciting areas of AI. Today's conversations were about generational AI and image because that's really what's taken up the imagination and the fore.
Meghan started to really point to this. That is, you know, the end of the beginning, not the beginning of the end. We genuinely believe that in our heart of hearts. There's a lot more to come, and I think, you know, when this crowd gathers several years from now, we're all gonna be shocked at sort of the trajectory that this thing has taken and how the pace of technology change has only accelerated. This is on the minds of every major global technology platform CEO in the world today. The conversation that we're having today and the focus around AI and generative AI, this is the same conversation happening in boardrooms of every $500 billio n and trillion-dollar market cap company in the world today. It's staggering, I think what people are seeing.
Yeah. Something to add, you can imagine if people are interested in the business that we've just discussed, you know, a long-term data relationship. What companies are waking up, and maybe this has happened to you as well, you suddenly understand that Shutterstock has a lot more to offer than stock photography. The conversations that we're having run the breadth of their advertising budgets, and the people that we're dealing with have large scale advertising budgets to suddenly say, "Oh, we didn't even know that you could drop a fridge in the desert." Right? "We didn't even know you could control the sun." Each one of these deals opens the door for other sales in our business, for other opportunities in our business to further partner, and then you've seen the retention levels that Jarrod talked about in our enterprise business.
When we land one of our enterprise businesses, they tend to stay because we have a broad product suite to offer them. It's not... I would hate for you to leave with, you know, with a headline, "The only thing that Shutterstock has now is, you know, long-term data relationships." That's just one arrow in our very, very deep, broad quiver that we're using in the space, and I think that's really interesting.
Just, Andrew Boone from JMP. Just on that last point in terms of the broadening of the quiver and as you guys now have more go-to-market across enterprise, you know, editorial, everything else you guys just brought today, can you guys just talk about how the go-to-market has to change, though, as there is more competition for your core marketplace, right? I mean, Jarrod, you brought up the fact that this lowers barriers to entry. What does that mean for Shutterstock, I guess, from a marketing or a CAC perspective as there is just easier access to content? Thanks so much.
Yes. Sure. I'll start. I think a couple things. Number one, and we sort of touched on this a little bit. From an enterprise perspective, handholding is gonna be mission critical. As you think about 250 feet on the street, never has that been so valuable as a time of significant perilous technology change for an enterprise marketer. It truly is a crown jewel asset on a standalone basis, but at a time of rapid technology change, having those 250 professionals who are in sales or customer management to really walk clients through what is generational AI? How do I leverage it? What are the legal pitfalls associated with it? How can I use it for areas such as marketing or websites? What's appropriate? What's not appropriate?
We are ideally positioned to really be that handholder for our enterprise clients and be their initial entree and foray into generational AI. I think we're excited about that. From a CAC perspective and from an e-commerce perspective, I think Meghan mentioned a little bit about this during her talk. I mean, we are extraordinarily excited by the top-of-funnel engagement. You know, to say that we are surprised to the upside by the visits, by the sign-ups, we are enthusiastic. Ultimately, it's about monetization. It's about conversion. We know that we have work to do in order to make sure that those people who are coming and signing up are ultimately putting in their credit cards and staying with us for many years to come.
That's where the battle is going to be fought. I don't think we're thinking about doing anything different with respect to CPA. We're not looking at massively spiking up our marketing budget. What I will tell you is there are reallocations going on, there is a tremendous amount of pent-up interest in generative AI among creatives, and we have a highly performant SEM apparatus, SEO apparatus, and marketing machine that is going to get those eyeballs and that engagement over to Shutterstock.com. We're very focused on leveraging the excitement, using our existing marketing channels, you better believe that we're looking at reallocating spends and rethinking about what's going to be most effective for us in this context.
Any other questions? I think we're actually right on time. Chris, are we?
We have time for a few questions.
Okay. Anything else from the floor? Yep.
Toby, hold on. Excuse me.
Hey, Toby Haselberger, Greenlight. Thanks for the presentation and everything. Pretty fascinating stuff. As you guys license out, you know, some of this data content, you know, I know one of the big things here is, like, you know, enterprise customers wanna kinda do things the right way and, you know, not have issues with sort of rights entanglement or potential future legal issues for, you know, acquiring content or using content from engines that acquired it the wrong way. When an OpenAI kind of licenses this content from you guys, and builds their model with it, does that then mean that the content, or the images that are created by users of OpenAI is all sort of, you know, free and clear from a, from a rights perspective?
I wanna make sure that I've got it. The content that we've licensed through OpenAI that now is on our website, everything that customers get from our website is for use. They have-
Right. Okay. I was actually-
You're just saying if I go directly to-
If you go directly to OpenAI, since they have paid for that, paid you guys for that and have that partnership-
Yeah.
As opposed to some of these other Stable Diffusion and whatnot that have not.
Is Meghan in the back? Can you just yeah, give a quick holler. I wanna make sure we answer this properly on behalf of another company.
Hi. Fundamentally, we can't speak on behalf of OpenAI, but here's what we can say. We properly license content to them. When that's now on our site, and it's generated on our site, we are taking the additional step to compensate contributors again. Obviously, if you go direct to DALL-E, that's not happening there. We can't comment on whether that's the right thing to do, but we're following our playbook, which is we wanna make sure that we're giving our contributors every opportunity to continue to make money and monetize with us. That's the full circle that we're creating exclusively at the Shutterstock platform.
Meghan, OpenAI, do you know if someone goes directly to OpenAI, if in fact, there is for use rules or license? No. There are not.
There are not.
Yeah. Okay, great. Thank you. Great. Something else?
We have time for one more question. Yep.
I'm a cinematographer kinda by technical trade, I would say when I look at Shutterstock, I'm like, "Hey, it's you're only as good as the content," right, "contributors are giving you." Obviously, I'm biased. I look at, you know, we're trying to serve the customer, investors, and contributors, that's in my experience, that's hard. That's a hard trifecta, right? To serve all those people well. How do you then I look at the pressure on the subscription, which I don't know how ultimately the payouts are for contributors, it seems like it's probably a little bit of downward pressure. The AI stuff, like, how do you sustain happy contributors?
Yeah, it's a great question. We're passionate about it. What I can tell you is, I mentioned if you think all the way back to 9:30 when I had my opening remarks, we've never had more contributors in our business than right now, and they've never given us more content than right now. Striking that balance is actually something we're very, very good at. We are contributor-centric. We think about, like, no content, no business. We understand that. No content, no business. No customers, no business, right?
This idea of customers and contributors coming together for a two-sided marketplace is exactly what Jon built when he started this business, and we're continuing to build that with an equal and balanced approach to making sure that we get it right for our customers and get it right for our contributors. So I don't actually see that as full-blown tension, but I see it as something symbiotic that actually creates a moat around our business when you take really good care of our contributors so that they go out and bring us more, more content, which we desperately need, and our customers desperately need that content to buy to purchase so they can make their creative stories. I think it works really, really well.
Over the years, right, our contributors have been compensated and keep coming back and are enthusiastically leaning into our business. We're very friendly in that if some contributors say, "You know, this whole data thing freaks me out, and I don't wanna participate," we get that. Opt out. No harm, no foul. We're in the monetization business. Contributors come to us, say, "Hey, Shutterstock, could you help me make money?" Sure. We're pretty good at that. We've got broad distribution from the smallest of customers to the largest of customers. Our job, right, our oath, is to monetize on behalf of our contributors. Our oath to our customers is to help them get the right content so they can better tell their story. I really see this flywheel working well and will continue to work well in the future.
You know, we wouldn't have a business without our contributors. Hopefully that answered that. Great. Great. Something else from the floor? No? Chris? Well, great. Thank you. We're right on time. I've got just a bit of housekeeping for you. Thank you again for all of your attendance and engagement. It's been awesome. On the back of your badge, don't look at it. No, I'm just kidding. You can look at it. There's a map and a color, and what's going to happen now is we've set up some amazing activations. I'd urge you, if you've got a few more minutes, make the rounds to our activations. We've got a lot of smart folks that are gonna show you a lot of cool things. If you follow the map and follow your color scheme, it will tell you where to go first.
Then from there, you can actually rotate around. If you have any questions, just grab any one of the Shutterstock employees, and we can help you find your way. Thank you all for coming. For all those online, thank you for engaging with us and dialing in remote. We will see you soon. Thank you very much.