As the models evolve and issues resolved, we may choose to offer our customers generative tools that enhance their creativity and efficiency. And we may choose to accept generative content consistent with other computer-enabled content on our platform. So we will evolve as the landscape evolves. But it needs to substantively meet the needs of our customers. You need to elevate their creative outcomes. It needs to save them time, save them money and mitigate their risk. So that’s kind of how we think through that third dimension about offering new services. And then lastly, the fourth layer is respect with our editorial business. This technology can be used for nefarious purposes. — And Getty Images needs to continue to stand as a source of truth with respect to our editorial imagery.
And so to do that, we’re really focused on continuing to invest in the people, the partners, the processes and the technology that makes us a trusted source. So it’s a long-winded answer but hopefully, that gives you the kind of the 4 dimensions that we’re really thinking through AI, how we think about it as an opportunity for — to grow our new revenue stream, how we think about it relative to our existing revenue streams, how we think about it from a product and services standpoint and how we think about it in terms of editorial integrity and deep page technologies.
Operator: Our next question is from Tim Nollen with Macquarie.
Tim Nollen: I’ve got a couple of questions about your KPIs related questions interrelated questions, I should say. So your customer count was up nicely year-over-year. It looks like it was down quarter-over-quarter. Can we assume there might have been some macro impact that affected that? And then also, you had some nice — very nice rise, especially in your active annual subscriber count, also your video attachment rate. It looks like maybe not fully layering into revenue growth yet but maybe I’m not clear on that. If it looks like it’s not factoring in yet, is it a matter of it just takes time to build the ARPU, if that’s the metric you use in terms of the revenue that you get on those customers? Or is it maybe some Unsplash customers that are just lower cost? Maybe if you could just help clarify how that works.
Craig Peters: Sure. I’ll take this and then Jen, you can layer on anything over the top. So on the account side of things, we are seeing a slight macro impact, as we mentioned, kind of in certain geographies and within certain parts of our business. But I would say the bigger thing that you’re really seeing there, Tim, is we are pushing to drive that annual subscription number more aggressively. And we do trade off some level of purchasing as a result of that but it’s a very net positive trade-off to the business overall. With respect to video and subs, we do see attachment rate there. We’re not growth in annual sub growth. Again, back to my comments, we’re pushing that more aggressively. Unsplash layers into the annual subside of things, it doesn’t layer into the video.
But even its impact on the annual sub figure is kind of minimal. It’s not overly material at this point. We only launched that subscription in October of last year. So what you’re really seeing there is we are seeing solid bookings and it takes a while for those bookings to flow through into recognized revenue. We have been introducing smaller subscriptions, smaller subscriptions that include video, smaller subscriptions that include the transition customers on the e-commerce side of things to annual. So maybe the ARPU of some of these subscriptions isn’t quite what it’s historically been but it’s highly additive to the business. And again, we kind of book it up front and then we flow that through over the next 12 months. So that is some of the impact that we expect to see coming into this year.
Jen, anything that you would add?
Jennifer Leyden: No.
Operator: Our next question is from Ron Josey with Citi.
Ron Josey: I have two. And maybe, Craig, starting off on the AI question or topic following some of your comments on the 4 dimensions. Can you just talk a little — talk a little bit more about the new revenue opportunity you talked about that might go far beyond licensing as companies need to license the content. Where are we — I’m assuming we’re very early days here but talk to us about what needs to happen for that to just start implementing or seeing that in the model? And then, Jen, you talked about softness in agency base and e-commerce side of things. I wanted to hear a little bit more on the corporate side of the business. And as it relates to revenue retention, I think that declined in the quarter and the year, I think we heard earlier COVID comps but any insights on current cohort trends as we think about maybe retention rates going forward would be helpful.
Craig Peters: You got it. So Jen, I’ll take the first and then you can take the second. So on the revenue side of things in terms of licensing, again, I think we have litigation with stability under the search and that we believe fundamentally people should pay for the assets that we possess as they apply them into generative models or other AI systems. And we think our assets are quite unique. We have been doing some licensing for those that have chosen to approach us and have that conversation. We have reached certain license agreements. But when we look at the space and we look at the potential for computer vision and generative AI, it’s massive. And it’s much broader than simply image licensing. I mean, this has to do with the creation of any creative editing platforms across the board.
This has to do with self-driving vehicles. The application of this technology and the need for these technologies to train themselves up on a significant corpus of imagery with highly relevant metadata. And ongoing flows that kind of reflect the contemporary and where the world is moving, right? So think about how technology changes over time. Think about how fashion changes over time. Think about how simple things like landscapes and Cityscapes change over time. So what we are focused on is trying to establish clear parameters around that in terms of the need to license and then ultimately, applying our data out into this field to create a new revenue stream for the business, again, or at least a much larger revenue stream than it’s historically been given recent kind of activities.
Does that make sense, Ron?
Ron Josey: It does. Craig. Appreciate it.
Craig Peters: All right. Jen, do you want to pick up the other?
Jennifer Leyden: Yes, I’ll pop in. So on corporate, corporate actually was really strong for us in 2022. We had another year of double-digit year-on-year growth in the corporate space. That continues to be an opportunity for us. So it’s definitely one of our growth levers. We think we can drive even harder into the corporate space but a strong finish relatively speaking, with double-digit growth year-on-year. With respect to revenue retention, so we finished just over 100%, 100.1% to be exact. The — I think step down you’re mentioning that, that rate dipped to about 88% in 2020 that purely being driven by COVID impact. We saw the rate jump back up in 2021 to just under 105%. So that jump 2020 to 2021, 88% to 105%. That is the benefit of a year-on-year compare coming off of a COVID impacted 2020.
That said, historically, we have seen this rate average 100% plus. That is where we finished 2022 and that is where we would expect to see this continue to be around that 100% plus range.
Ron Josey: That’s helpful, Jen. And then on corporate growing double digits, any insights on percentage of revenue or things along those lines. And thanks again for the answers here.
Jennifer Leyden: You are welcome. So yes, we — our corporate now makes up north of 50% of our revenue with the agency making up well under 20% of our revenue.
Operator: Our next question is from Brett Feldman with Goldman Sachs.
Brett Feldman: The first one just has to do with getting a little more insight into your guidance for the year. So I look at your revenue guidance range on a currency-neutral basis, the 1.9% to 4.9% growth. How do we think about the swing factors between the high and the low end of that range? Or maybe another way of thinking about the question would be if the current operating conditions were to persist for the remainder of the year, where would that put you in the range? And then I’m curious, the range you have for EBITDA, is that really driven exclusively by the low and the high end of revenue? Or is there something else further down the P&L that would create that variability? And then I have one more follow-up question after we get to this one.
Craig Peters: Jen, do you want to take that?
