Participants

Matt Pfau; Senior Vice President, Investor Relations; Zeta Global Holdings Corp

David Steinberg; Chairman of the Board, Chief Executive Officer, Co-Founder; Zeta Global Holdings Corp

Christopher Greiner; Chief Financial Officer; Zeta Global Holdings Corp

Terrell Tillman; Analyst; Truist Securities

Jason Kreyer; Analyst; Craig-Hallum Capital Group

David Hynes; Analyst; Canaccord Genuity

Arjun Bhatia; Analyst; William Blair & Co

Elizabeth Elliott; Analyst; Morgan Stanley

Ryan MacDonald; Analyst; Needham & Company

Richard Baldry; Analyst; ROTH Capital Partners

Matthew Swanson; Analyst; RBC Capital Markets

Jackson Ader; Analyst; KeyBanc Capital Markets

Koji Ikeda; Analyst; BofA Global Research

Brian Schwartz; Analyst; Oppenheimer & Co Inc

Zach Cummins; Analyst; B. Riley Securities

Gabriela Borges; Analyst; Goldman Sachs

Clark Wright; Analyst; D.A. Davidson & Co

Presentation

Operator

Greetings, and welcome to the Zeta first-quarter 2025 earnings conference call. (Operator Instructions) As a reminder, this conference is being recorded.
It is now my pleasure to introduce your host, Matt Pfau, Senior Vice President of Investor Relations. Thank you, sir. You may begin.

Matt Pfau

Thank you, operator. Hello, everyone, and thank you for joining us for Zeta's First Quarter 2025 Conference Call. Today's presentation and earnings release are available on Zeta's Investor Relations website at investors.zetaglobal.com where you will also find links to our SEC filings, along with other information about Zeta. Joining me on the call today are David Steinberg, Zeta's Co-Founder, Chairman and Chief Executive Officer; and Chris Greiner, Zeta's Chief Financial Officer.
Before we begin, I'd like to remind everyone that statements made on this call as well as in the presentation and earnings release contain forward-looking statements regarding our financial outlook, business plans and objectives, and other future events and developments, including statements about the market potential of our products, potential competition, revenues of our products, and our goals and strategies.
These statements are subject to risks and uncertainties that may cause actual results to differ materially from those projected. These risks and uncertainties include those described in the company's earnings release and other filings with the SEC and speak only as of today's date. In addition, our discussion today will include references to certain supplemental non-GAAP financial measures which should be considered in addition to and not as a substitute for our GAAP results.
We use these non-GAAP measures in managing our business and believe they provide useful information for our investors. Reconciliations of the non-GAAP measures to the corresponding GAAP measures, where appropriate, can be found in the earnings presentation available on our website as well as our earnings release and our other filings with the SEC.
With that, I will now turn the call over to David.

David Steinberg

Thank you, Matt. Good afternoon, everyone, and thank you for joining us today. 2025 began much like 2024 ended with Zeta exceeding expectations and gaining share in a dynamic market. This marks our 15th consecutive quarter of beating and raising our guidance, reflecting the strength of our innovation, execution, and consistent delivery of customer value.
While macro uncertainty has become more pronounced at the start of the second quarter, Zeta's value proposition continues to stand out, particularly in environments where marketers are under pressure to deliver measurable results and do more with less.
Our omnichannel people-based artificial intelligence platform delivers both scale and precision, enabling marketers to target, personalize, measure and optimize across owned and paid media more efficiently and effectively than alternative solutions. We proved the strength of our model during COVID-19 and the tech reset of 2022, growing revenue by 20% in 2020 and 29% in 2022, despite challenging macro conditions.
Our focus on delivering predictable, profitable and measurable ROI is driving continued market share gains and has contributed to our strong first quarter results. In the first quarter of 2025, we generated revenue of $264 million, up 6% year over year with adjusted EBITDA of $47 million, up 53% year over year, both well ahead of our guidance. Given our first quarter's outperformance, strong pipeline and second quarter visibility, we are raising our second quarter and full year 2025 guidance.
While our momentum supports a larger raise, we're taking a disciplined and conservative approach in light of the ongoing macro uncertainty. We believe one of the key reasons our business has remained resilient is our focus on performance-based outcomes and exposure primarily to lower funnel marketing. Let me share a few examples of existing super scaled customers that illustrate this. For a large telecommunications customer, we have consistently performed their benchmark for new customer acquisition cost, most recently beating their target by 53%. Our ability to be a repeatable, scalable solution has led to an increase in commitments culminating in the recent signing of a two-year agreement that more than doubles their annual investment with data.
For a major insurance provider, we helped drive 7% lower new customer acquisition cost versus their 2024 target. We expect this success to drive a 50% increase in spend with us this year. For a financial institution, we beat their target new customer acquisition cost by 14% in 2024, and the customer is expected to increase their spend in 2025 by close to 20% after nearly 150% increase in 2024. Importantly, these results are not self scored. They're validated either directly by our customer or through third-party attribution methods.
Now let's talk about one of our biggest long-term drivers, artificial intelligence. This quarter, we launched AI Agent Studios, a suite of generative AI tools that enable users to select and activate built agents create custom agents and link them together to execute complex marketing tasks. As a part of AI Agent studios, we also announced Agentic Workflows, which were available in beta.
Agentic Workflows give marketers the ability to build and customize AI agents that work together to complete complex tasks across the entire customer journey. Unlike generic tools, Agentic Workflows are flexible and adaptive, making AI not only actionable but also potentially transformational to a marketer's productivity.
Here is a real world example of how a customer uses our Agentic Workflows. The customer needed a multi-agent workflow to identify high-value audiences and activate to them across multiple channels. With one prompt, this workflow coordinates intelligent agents to speed up planning and execution. Through this workflow, our insights agent binds target audience segments. The strategy agent crafts the messaging and personas and the activation agent deploys the marketing through multiple channels.
This process streamlines collaboration, reduces time from insight to execution, enhances precision and creates meaningful return on investment all within a single AI-powered workflow. Now I'd like to provide an update on our agency business, which has seen great momentum to the start of the year. I'll begin with an update on our efforts with independent agencies. We executed and platformed two new independent agencies in the first quarter and are finalizing agreements with two additional ones, proving early traction in a market with over 1,000 potential prospects. We also experienced solid year-over-year growth with large holding companies in the first quarter.
For our holding company customers, we believe Zeta is by far one of their most profitable partners, which is important in good times but even more important during challenging times. Agencies choose to partner and expand with Zeta, not only because we drive efficiencies in their businesses, but also because our platform offers advantages that assist in securing new business. For example, one of the independent agencies we platformed in the first quarter used our retail audience intelligence to support its pitch to a prospective new customer, Zeta's deterministic attribution capabilities were a key differentiator, providing a level of measurement that the agency had not previously been able to offer. These capabilities help the agency secure the customers' business, which, in turn, creates additional revenue opportunities for Zeta. A true win, win, win.
At Zeta, we have faced challenging macro times before and emerged stronger by controlling the uncontrollable, how we think how we act, how we deliver value for our customers. In the end, execution is the difference between surviving uncertainty and driving through it. As always, I would like to sincerely thank our customers, our partners, team Zeta and all our shareholders for the ongoing support of our vision.
Now let me turn it over to Chris to discuss our results in greater detail. Chris?

Christopher Greiner

Thank you, David, and good afternoon, everyone. Before diving into the details of our quarterly results and updated 2025 guidance, I want to highlight four key takeaways. First, despite global macro uncertainty, we reported a strong quarter with broad-based contribution across all areas of our business. Second, our increased 2025 projections are reflective of our momentum exiting the first quarter and the good visibility we have into the second quarter, while being thoughtfully conservative in the second half given macroeconomic and policy uncertainty. Third, we've looked at our capital allocation strategy and are taking significant steps to reduce dilution and stock-based compensation expense.
And fourth, our increased focus on free cash flow generation is evident in our first quarter results, with free cash flow growth significantly ahead of revenue and adjusted EBITDA and a substantial improvement in free cash flow conversion. Now let's get into the details of our first quarter results. In Q1, we delivered revenue of $264 million up 36% year over year or 26% excluding the contribution from LiveIntent. Total scale and customer count grew to 548, up 19% year over year and an addition of 21 customers sequentially. We had 159 superscale customers at the end of the first quarter, an increase of 10% year over year and 11% quarter-to-quarter.
Scaled customer quarterly ARPU of $467,000 increased 12% year over year and superscale customer quarterly ARPU of $1.4 million increased 23% year over year. From an industry perspective, on a trailing 12-month basis, 6 of our top 10 verticals grew faster than 20% year over year. We ended the quarter with 173 quota carriers, up 22% year over year and down seven heads sequentially. The sequential decline was in line with our expectations and driven by the completion of the LiveIntent integration. Our direct mix in the first quarter was 73%, down from 74% in the fourth quarter of 2024, but up from 67% a year ago, resulting in direct revenue growth of 48% year over year.
Our GAAP cost of revenue in the quarter was 39.1%, a 90 basis point improvement sequentially and a 30 basis point improvement from the first quarter of 2024. This improvement in cost of revenue as well as leverage in other areas of our operating expenses resulted in our 17th straight quarter of expanding adjusted EBITDA margins year over year. In the first quarter, we generated $46.7 million of adjusted EBITDA at a margin of 17.7%, 200 basis points higher year over year and $2.2 million better than the midpoint of our guidance. One of the facts to adjusted EBITDA was $3 million of restructuring expenses, which were anticipated and primarily related to the integration of LiveIntent. Our GAAP net loss for the first quarter was $22 million, an improvement from $40 million in the first quarter of 2024.
First quarter net cash provided by operating activities was $34.8 million, up 41% year over year, with free cash flow of $28.2 million, up 87% and representing a margin of 10.7%. This translated to a free cash flow conversion of 60%, a significant improvement from 45% in the fourth quarter of 2024 and 50% in the first quarter of 2024. We also repurchased 1.6 million shares for $25 million, accounting for 89% of our free cash flow generated in the quarter. We continue to repurchase shares after the close of the quarter, acquiring an additional 1.8 million shares for $21 million between April 1 and April 25. We have approximately $38 million remaining under our current share repurchase authorization and upon completion of this authorization, we plan to initiate a new one.
Before diving into our 2025 guidance, I want to highlight a few key elements of our business model that are especially relevant given the current macro environment and informed how we approached guidance. First, as David discussed earlier, we deliver a clear, measurable ROI to our customers. This value proposition has helped us maintain an annual net revenue retention rate of 111% or higher per year since our IPO in 2021. Second, nearly all of the marketing and advertising spend we support is tied to measurable KPIs, particularly lower funnel conversion metrics. We believe this makes our platform more resilient in volatile macro environments as lower funnel spend tends to be less discretionary than top-of-funnel brand awareness efforts.
Third, we primarily address large enterprises whose marketing spend tends to be more stable than small businesses in our view. And fourth, Typically, more than 90% of our annual revenue comes from customers that have been with us more than a year. These dynamics have contributed to our relative outperformance during challenging periods. For example, in 2022, a year when many tech companies experienced a sharp slowdown in growth amid rising inflation and broader budget rationalization, our revenue growth accelerated by 4 percentage points, reaching 29%. Lastly, and relevant in the context of the current macro environment, our direct exposure to federal government and China is minimal.
Following a strong quarter and considering our current data points including a robust sales pipeline, positive customer interactions and good visibility into the second quarter, we would typically revise our full year guidance upwards by the amount we exceeded the first quarter's expectations as well as increase our second quarter forecast. However, we'd all agree these are not typical circumstances. Because of this, we prefer to adopt a thoughtful approach that blends short-term momentum and visibility with second half conservatism. We are increasing our second quarter revenue outlook by $2 million, while adjusting our growth expectations for the second half of the year. The conservatism in our guidance can be seen through three lenses.
First, we are one month into our second quarter and have good visibility for this period. Second, although we continue to see healthy demand, our guidance assumes a softer macro in the second half of the year. Third, to meet our guidance, our four fastest-growing industries in the first half of the year only need to grow at half that rate in the second half. To be clear, our customers have not expressed intentions to decrease their investments with data due to macro uncertainty, nor has there been pull forward in spending. Our intent is to create buffer by exercising prudent conservatism regarding expectations for the remainder of 2025.
Our 2025 revenue guidance is now $1.242 billion at the midpoint, an increase of $2 million versus the midpoint of our prior guidance. This represents reported growth of 23%, a growth of 21% when adjusting for LiveIntent and political candidate revenue in the year-over-year comps. For the second quarter, we now expect revenue of $297 million at the midpoint, $2 million higher than our previous guidance. For adjusted EBITDA, we're increasing the midpoint of our 2025 guidance to $258.5 million, up $2 million from our prior guidance and representing a year-over-year increase of 34% at a margin of 21%. In alignment with our conservative approach to revenue projections, we're confident we can achieve our adjusted EBITDA margin target for 2025, just through leverage in sales and marketing, R&D and G&A.
We have many levers we can pull within operating expenses and are confident we can protect our margins in the event our top line is adversely affected due to macro weakness. For the second quarter of 2025, we now expect adjusted EBITDA of $54.9 million at the midpoint, up from our previous expectation of $54.4 million and representing growth of 42% at a margin of 18.5%. We are increasing the midpoint of our 2025 free cash flow guidance to $131.5 million, up $2 million from the midpoint of our previous guidance and representing year-over-year growth of 43%. As I mentioned upfront, we revisited our capital allocation strategy. In addition to aggressively buying back shares, we have also decided to significantly reduce dilution and stock-based compensation.
As such, we're introducing a new guidance item this quarter, stock-based compensation expense, which we expect to be $190 million for 2025, lower than the $195 million in 2024. In addition, for 2025, we expect our normal core share count dilution to be 4% to 6% and for 2026, we expect to improve further to 3% to 4%. This is a significant reduction from 15% total dilution in 2024 which included 8% normal course equity grant dilution.
Slide 20 in our earnings supplemental outlines how we plan on achieving these targets. Part of the improvement will be driven by David, Steve Gerber, and myself not participating in the annual equity grant process in 2025 and instead, we'll have our compensation tied to longer-term goals.
We recognize that stock-based compensation expense and dilution are important topics for our investors, and we're taking significant steps to improve both. Lastly, we remain confident in our Zeta 2028 plan and are reaffirming our long-term targets, which project over $2 billion in annual revenue, at least 25% adjusted EBITDA margin and 16% plus free cash flow margin in 2028.
Now let me hand the call back over to the operator for David and me to take your questions. Operator?

Question and Answer Session

Operator

(Operator Instructions) Terry Tillman, Truist Securities.

Terrell Tillman

Yeah, David, Chris and Matt, bear with me, I've got my own preamble here. First of all, congrats on the 15th beat and raise quarter. And then also on the color on the second half guidance velocity. That's actually very helpful. My first question relates to like progress on One Zeta in terms of accelerating the cross-sell and use case expansion.
I know you all brought in some key leaders recently some talent acquisition, but just maybe an update on One Zeta and how that's tracking. And then I have a follow-up. Thank you.

David Steinberg

Well, let me start by saying thank you, Terry. I'm losing my voice here a little bit. But One Zeta -- I get that. I get that. What I would tell you is the One Zeta data strategy is ahead of schedule.
Even I just got back from the possible conference where during the course of the contract, we signed a deal which will be one of our biggest deals in history around this very same stuff. The concept of wrapping it together where instead of using one use case in multiple channels, they have multiple use cases and multiple channels continues to be one of the biggest drivers of the business.
In the first quarter, obviously, you saw a 36% year-over-year growth with 26% of that being organic. A big piece of that was One Zeta and how we're progressing on it. I was joking with someone the other day.
I remember a few years ago when our goal was how do we get five or six customers to 5 million to 10 million a year. We're now looking at how do we get five or six customers to 100 million a year. And we're actually well on our way to that exiting the first quarter.

Christopher Greiner

One just bringing the data points to like there, Terry, for you. All three of our use cases, the acquire, grow and retain. All three grew double digits year over year and our fastest-growing cohort in terms of channel expansion were those scaled customers that are using three or more, and that's up 42% year over year. So just to bring the data to David's point.

Terrell Tillman

That's great. Just one question as we think about the growth algorithm, excluding live intent, which I'm sure you'll get some questions on. I'm just curious, how are you thinking about the rest of the year as it relates to scale customers and ARPU from skilled customers?

Christopher Greiner

I think in line with our model, we did ARPU growth in the quarter was 12% year over year. Our model is 12% to 16%. Scaled customer count grew 19% year over year. Interestingly, the scale of customer count growth is obviously aided by LiveIntent. But the ARPU is impacted negatively by LiveIntent, just given what their average customer size is.
But I would expect us to continue to operate in a similar way as we did in the first quarter through the remainder of the scale of the customer count and ARPU growth side.

Operator

Jason Kreyer, Craig-Hallum.

Jason Kreyer

David, you had mentioned in your prepared remarks, just maybe more pronounced macro uncertainty at the start of Q2. Maybe you can break that down a little bit, what Zeta is seeing or what Zeta customers are seeing versus the broader macro?

David Steinberg

Thanks, Jason. I appreciate it. At this point, we're not seeing any turbulence inside of Zeta either for Q2 or for the back half of the year. We've not had one client pause, and we have not had one client leave. So we continue to see that.
Now at the same time, we're trying to be thoughtful in a world where there's a lot of uncertainty. We just wanted to make sure that we were being conservative in the back half. But as it relates to the Zeta platform itself, at this time, we have seen zero changes.

Christopher Greiner

In fact, just in April, ended just as strong as March did. So we're -- the quarter itself is even off to a great start, which gives us the type of confidence we did we had to have to raise the quarter.

Jason Kreyer

Perfect. Good to hear. I wanted to also just ask about this independent agent opportunity. curious there if the go-to-market like how the go-to-market with independents is different than holdcos? And then, Chris, in the past, we've dealt with some headwinds to margin and cash flow as you've seen greater opportunities with hold post.
So just curious if there's anything we need to think about as you deepen that opportunity with independent agents.

David Steinberg

So interestingly enough, you don't have those headwinds with independent agencies. Traditionally, we are platforming the entire independent agency at once, Jason. So you're seeing long-term contracts where you're -- you've got a lot more visibility almost all of it's on platform, and we're getting paid more in line with our normal DSOs than the DSOs from the agency holdcos. Although I will say our largest agency holdco in the first quarter did execute and move to a multiyear contract, and we expect that client to continue to be very, very solid and grow in the years to come.

Christopher Greiner

Jason, on the cash flow point, I'm glad you raised it. Throughout the course of the quarter as we were meeting with investors, there were three things that were important to them that we delivered on. The desire to have a clean quarter with clean comps, we clearly delivered that. To address stock-based compensation and dilution, I think we've definitively addressed that as well. And then the third was to continue to focus on more and more free cash flow conversion and generation.
So to your point, this quarter at a 60% conversion from adjusted EBITDA was a record for us on an as-reported basis. That still included a headwind in terms of working capital by virtue of the growth we're having with the agencies. But I feel like we've more than covered for that in our guidance conservatism for free cash flow conversion, which is right around 50%, again, compared to the 60% that we just generated this quarter. David?

David Steinberg

I would think about it, Jason, that the independent agency space is more like just signing a very large multinational enterprise, where you're looking at comparable margins and comparable term of contracts to those versus with the agency holdcos.

Christopher Greiner

In fact, this week, in addition to promise, we just signed another independent agency this week.

David Steinberg

Yeah. We're -- I mean we will -- I mean the independent agency ecosystem has over 1,000 potential prospects in it, and we doubled it in the first quarter over last year. And I think that's a trend -- I think you can't probably double it every quarter from a cyclical perspective. But I would say -- I would expect that strategy to continue to bear fruit in the form of expanded on-platform revenue and expanded conversion of EBITDA to free cash flow from where we originally projected.

Operator

DJ Hynes, Canaccord Annuity.

David Hynes

Congrats on the nice quarter. I'm glad to hear April closed equally as strong. David, you gave us some great expansion examples in telco and insurance and finance. If we think about some of the verticals that maybe you're keeping a closer eye on today in this current environment, what are those? And where would the impact show up first, right?
Is it in ARPU? Like what are you paying attention to and what are the verticals?

David Steinberg

Yes. Sorry. Thank you, DJ. I would say, listen, I came into April most worried about automotive. We have a bunch of automotive manufacturing clients who are international in addition to manufacturing in the US, but as we all know, I think even organizations in the US that manufacturer here are bringing parts and they have supply chain issues in there.
We saw the opposite. We actually saw the automotive space grow a little faster than we started looking with that pull forward, and it does not appear to be because most of the increase came from new contracts and new agreements that were signed for the calendar year or longer. But I would have been most worried about automotive. We're not seeing anything there yet.
And then retail. Right, as retailers look at their businesses, how did they perform and very similar. Through April, we've seen no disruption in the retail space. In fact, it's interesting. One retailer who I met with at the possible conference made a comment that because we are, by far, their most efficient partner, we might see additional growth out of them that we might not have expected as they shift budget from partners that don't have the type of return on investment and, quite frankly, the ability to prove it that we currently have an evidence.
Chris?

Christopher Greiner

Yes. DJ, a few points of color on the industry vertical question because it's obviously really important as you start to break down your guidance. For us, and I mentioned in the prepared remarks, on a trailing 12-month basis, 6 out of our top 10 verticals grew over 20% and it was well over 20%.
If you look at it on a first quarter basis alone, there were 7 out of the 10 that grew north of 20% and it was well now north of 20%. As I think about those verticals, and we put in our guidance commentary our four fastest-growing verticals can grow half that rate in the second half, and we still get to our guide.
And again, we're not seeing any data-driven reasons as to why that would be the case in our sales pipeline, et cetera. It's just our conservatism and the buffer that we're putting in place. But we've even gone to the lanes in verticals like consumer and retail where we've gone customer by customer, and we've looked at it through two dimensions, what is their business model sensitivity to China and where are they on the curve of discretionary spending. And we walked away from that analysis in consumer retail and other verticals, not having -- being very comfortable with where we've set guidance in terms of our relative exposure and their exposure.

David Hynes

Yes. Yes. Okay. That's all very helpful color. Chris, maybe this is probably not a fair question, given the way you're talking about the business and the optimism going forward.
But if growth were to slow to sub-20%, is it right to think that the trade-off would be showing more operating leverage. Maybe you could just talk philosophically with respect to the internal scenario planning that you're doing?

Christopher Greiner

Yes. I mean we have -- our business model is we don't have to sacrifice one for the other. As we've been growing fast, we just talked about our 17th straight quarter of adjusted EBITDA margin expansion. To your point, if we were to see a slowdown we have the levers in our business. When you think about it from a sales and marketing perspective, the amount of variable spend there, obviously, if revenue were to slow, that has an element of commissions expense that you have flexibility on.
And at the same time, within G&A and R&D, which we've done for a while, we continue to move work where it could be performed best and most effectively. So continuing to use our global distribution would be a lever that we have.
But I will point out even though we didn't fully roll through the revenue beat, we did fully roll through the adjusted EBITDA beat in the first quarter through the full year as well as free cash flow for the full year. That was an important signal to -- the Street that there's more leverage in our business for us to get for the rest of the year as well.

Operator

Arjun Bhatia, William Blair.

Arjun Bhatia

Congrats on a strong start to the year. David, I think you were right to point out earlier in the call that you grew through 2020, you grew through the post-COVID, post slowdown. But if I think about the business, it seems like you're in a very different situation now than you were two years ago or five years ago, just the business has grown, you have a lot more growth avenues.
So if there is -- in the back half of the year, some macro slowdown. I'm curious what strategies you have at your disposal now that you can maybe undertake to protect the growth rate somewhat, not suggesting that you'd be immune, but I'm sure there's maybe new products or different use cases or different customer types that you can lean into at this point.
But I'm curious what that looks like.

David Steinberg

Arjun, as usual, you're totally right. I mean when you look at our business today, the incremental use cases and the incremental channels that we have available to us allow us to move things around in a way that we even couldn't in 2020 or 2022.
The other thing that has changed is our attribution capabilities are a step function better than they were even two years ago. When we look at return on investment, the return we give clients today is superior to at any point we've ever run this business. So what we believe would happen even if the -- if macro uncertainty went to macro headwinds, which we're not seeing today, we don't believe we will be as affected as others because of that return on investment.
Now we are already out speaking with clients showing them clear return on investment, wrapping our arms around them, and we're seeing upsells currently that we might not have expected because we're being more proactive in wrapping our arms around our clients than maybe others would be. But I do think, as Chris said, in addition to the levers we have in the EBITDA and free cash flow, we have a number of levers in growth as well.
The other thing that I think people might not understand are our relationships with the holdcos are really helped there because as they see turbulence we are so much more profitable than the other partners they work with they're already bringing us new brands to get started in case they need to really scale those in the back half of the year. So once again, we don't want anybody to confuse conservatism with any weakness in the business.

Arjun Bhatia

All right. Yes, perfect. Very helpful. And then one for you, Chris. It's encouraging to see the cash flow conversion get better, obviously something investors have been focused on, but what are you doing actually to drive cash flow conversion higher?
And maybe what's still available to get that metric higher through the year?

Christopher Greiner

Yes. Look, something that we have talked about with investors and -- the Street is we fully expected our -- if you look at CapEx, it's just drilling on that as an example, not only do we expect that to continue to be more efficient from an EDR perspective. But you saw a really sizable tick down in CapEx, both software capitalization and data capitalization from the fourth quarter. So it's areas like that, that we talked about. We expected to see that we're executing on.
And just more and more of the business is dropping to the bottom line. It's -- the headwind that we have is just pure timing of when the agencies pay us right now. And we're optimistic as to where we ended the first quarter free cash flow for sure.

Operator

Elizabeth Porter, Morgan Stanley.

Elizabeth Elliott

I first just wanted to ask on the mix between integrated and direct. The direct growth remained really strong and elevated -- well, it's like integrated just slowed a bit. So I was wondering if there was anything to call out there as it relates to customer changing their mix or agencies that typically spend more on that integrated platform.

David Steinberg

No, Elizabeth. We're actually going through the process we expected to go through, right? So we've been saying for a while that we expected the early agencies to follow the same pattern as our first agency client, which started very, very highly on integrated and then move to direct. We're already seeing the agencies that are scaling very, very rapidly begin to move to direct a little faster than we expected them to.

Christopher Greiner

In fact, Elizabeth -- are amongst our fastest-growing channel, CTV is still the fastest, but social grew really nicely. So nothing there at all that alarmed us or caused us any concern.

Elizabeth Elliott

Great. And then just as my follow-up, I wanted to just get more perspective on how you're thinking about the growth opportunity with agencies longer term. I think it was about 20% of the business last year. And given the move more into the independent side is to -- could we start to think about the agency business as being a majority piece of the business in the not-too-distant future? And then any more in term changes to be thinking about as it relates to the cohort in 2025.

David Steinberg

Yes. So let me say that in order for that agency business to get 50%, it would have to grow substantially faster than we're currently growing and what we expect. So I think it's going to continue to grow. It will continue to inch up as a percentage of revenue. as they move from indirect to direct, I think we'll continue to see on platform and cost of goods sold continue to get better.
We're just -- right now, we're seeing that as very, very solid. But our direct-to-enterprise business is still growing very rapidly. So we're not seeing a slowdown there. And I will also point out that I think it's important to note that the independent agencies, of which we doubled that business in Q1 over Q4, and we expect to continue to grow that at an accelerated pace, really acts like the enterprise business. You're platforming there.
Cost of goods sold is very, very low. And we're able to work with them as it relates to cash flow. So I think right now, we've got a really good mix, and I think it's going to continue at its current pace.

Operator

Ryan MacDonald, Needham & Co.

Ryan MacDonald

Congrats on a nice quarter. David, curious to know about how the generative AI adoption continues to trend throughout Q1, particularly with the launch of [agent Studio]. And as you think about within this environment, are you seeing maybe a tighter budgetary environment if that happens, being a catalyst for faster adoption of AI agents doing more with less? Or do you think this causes maybe more of a pausing on some AI initiatives? What's the sense you're getting from your customers?

David Steinberg

We are not seeing any pausing whatsoever, Ryan, on AI adoption. In fact, if anything, I think the uncertainty is accelerating it. We saw a meaningful step-up in adoption in Q1. I think that's one of the reasons you saw the numbers flow through. Our -- our AI adoption leads to a substantially higher revenue from customers who adopt it.
And we continue to invest in it. As Chris said, our percentage of capital investment lowered as a percentage of revenue. We expect that to continue, but that's because revenue is growing so fast. We're still spending more money on innovation almost exclusively in artificial intelligence than we've ever spent on innovation in our entire corporate life. So we continue to invest there.
We continue to see adoption there. And I believe that's one of the reasons we saw the numbers come in where they did and why we feel very comfortable raising the year.

Ryan MacDonald

Super helpful color. Chris, maybe for you as a follow-up. I noticed that in the updated guidance that the expectation for LiveIntent didn't really change much. I know it's obviously a small portion of the business relative to the overall size, but is there any reason to believe that, that business acts differently or would be more or less resilient than the core business to the extent you have that visibility, given how new it is?

Christopher Greiner

Look, it performed right in line with what we expected it to do, to get to like $19.5 million compared to the $20 million guide that we had in place. So it's right on the trajectory for the year that we expect it to be on. What you're not necessarily seeing show up in LiveIntent because that's their stand-alone businesses performance are all of the synergies and the products that we're creating that, that is actually -- that's benefiting the organic side of the business to some degree.
But combined, the stand-alone capabilities in the business model of LiveIntent has added new channels for us to sell, which is additive to the business model. And then the products that we're jointly developing with them that are in market are actively being sold and contributing to the overall performance of the company. So very, very happy with how LiveIntent is performing out of the gate.

Operator

Richard Baldry, ROTH Capital Partners.

Richard Baldry

So you're seeing the agencies mix move overall from indirect to direct. I'm curious, is that because some of their brands are maturing moving along that or as they're onboarding new brands, are they bringing them on also more direct given they've the experiences they've had with initial clients?

David Steinberg

Thanks, Rich. The answer is yes to both. Because we're able to evidence that the on-platform business delivers a higher return on investment than the integrated component of the business, we're just seeing the agency holdcos move clients over faster. Some are starting there. A big chunk of them are migrating from integrated to direct, but both of those are tailwinds right now for that part of the business.

Richard Baldry

Right. And given the strength strategically of the LiveIntent acquisition, curious your thoughts overall on M&A as you go forward. You've been using a lot of free cash flow to do buybacks. But are there other tuck-in technologies or channels that you'd still like to be adding to the suite of offerings?

David Steinberg

Yes. I mean, Rich, as you know, we'll continue to look at our M&A pillars, and we have a very, very strong balance sheet, and we're currently spending free cash flow on share buybacks. But quite frankly, I've been spending a disproportionate percentage of my time taking calls from people trying to buy us over the last few weeks. And that's been keeping us busy.

Operator

Matt Swanson, RBC Capital Markets.

Matthew Swanson

Yes, my congrats on the quarter. It's not too much trouble, another macro one. I guess typically in a macro downturn, we see the expand motion come easier than land, right, because your existing customers already get the ROI joke and story. Throughout those 2020 and 2022 period, did you see any difference in mix where maybe it was easier to focus on ARPU than the new lands?

Christopher Greiner

It's interesting, Matt. If you look at our net revenue retention range, it's been like $1.11 to $1.14 since -- from 2021 to 2024. And if you look at our growth rate, it's more or less half what the growth rate has been over that period as well.
So half from existing, half from new, and it's been very consistent. And that's also what our expectation is going to be for 2025. It's about half the growth comes from the base, half from new. I think it's an interesting point.
David, you were going to add?

David Steinberg

Yes. I was going to -- I'll answer the question a little more towards what I think you were asking. No, we didn't see an increase in retain or grow versus new customer acquisition from a use case perspective as it related to '22 or '20. But what I would say, and Chris is, of course, right, when you've got 111% to 114% of your revenue becomes your net retention rate, that often is how do we move from one use case to 2.
And I would say that customers that are traditionally doing new acquisition, which is what I think Chris was trying to get across, we'll often expand into retention and monetization in a downturn faster then they will when things are not uncertain.
So listen, it's good to have all three use cases. It's good that they all grew really, really well in the first quarter. It's really good that they all grew really well through April. And we think that's a trend that's going to continue.

Matthew Swanson

That's really helpful. And then I know previously, when you've talked about guidance, you've mentioned the Zeta Economic Index is something that you guys look at as a barometer of health. Is that still a good way to think about for us to keep an eye on to think about the overall, I guess, macro environment as it relates to your business throughout the year?

David Steinberg

Yes. So we definitely look at the ZEI and I would tell you, even in the most recent ZEI, we're seeing more concern from customers than slowing spend. And I think that's a trend that continues, which is one of the reasons I think that enterprises that we work with at least have not cut back. I would not, if there were to be an unexpected downturn in the ZEI, look at that as an unexpected downturn in Zeta.
I would simply say it is a good indicator of what is happening behind the scenes with the economy. I will further say and I'll point out that what we often see is in times of uncertainty Zeta has grown faster than we expected it to because of our ability to drive return on investment that other enterprises we compete with do not do -- and that, to me, would be the better thing to look at.
And listen, there's not a lot of companies that are raising annual guidance right now, raising the quarter. We wouldn't be doing that if we didn't feel we were in a position to do it after obviously beating guidance and raising guidance 15 consecutive quarters as a public company.

Operator

Jackson Ader, KeyBanc.

Jackson Ader

David, can I just confirm, did you say that you've been fielding calls from people interested in buying Zeta the company or buying Zeta's software?

David Steinberg

No. It was the prior, but I probably shouldn't have said it. It was -- yeah, but it's neither here nor there. The reality is, right now, we are executing the company. We -- as we continue to grow and take massive market share, we're just getting a lot of attention on that side that we hadn't gotten in the past.

Jackson Ader

Understood. Second question, a little bit maybe capital allocation was, right? So the idea that you're looking to buy back more shares, $25 million a quarter, you have $30 million --or $30 million, $40 million left. And coupled with the comment that you continue to spend money on innovation, but why spend the $25 million, $30 million, $40 million buying back a relatively small number of your shares each quarter? If you are seeing a direct correlation to the amount of money you were spending on innovation, driving revenue growth.

David Steinberg

Listen, first of all, that's a great question, Jackson. If we could spend more money on innovation faster, we would. The challenge is we are very, very selective about the engineers we bring into this company, the architects we bring into this company.
And we have a large number of open job racks currently. So we are, quite frankly, in a position right now that we're able to deliver on the innovation and quite frankly, sales rep goals that we are looking to deliver on while simultaneously retiring shares.
I think the buyback goes back to our M&A thesis around our five pillars of M&A. And as I look out on the landscape at the multiples that we've been trading at with the profitability we have I don't think I could buy any company at that rate. So by retiring our own stock, it benefits all of our existing shareholders who we want to reward as a part of sticking with us.

Operator

Koji Ikeda, Bank of America.

Koji Ikeda

I wanted to go back to the commentary on the demand environment and the guidance and the conservatism in the guidance. And so loud and clear, you guys aren't seeing any change in the demand environment. And it sounds like the conversations you're having with customers are really good. And I totally get the conservatism. I really do appreciate all the color that you gave on the call in the prepared remarks about the guidance.
But I'm just curious, it feels like something outside of just the high-level macro uncertainty is informing the more conservative second half. And so if you could if you could share what are you seeing? Or maybe the better question is, what are you not seeing out there that is tilting you more conservative in the second half?

David Steinberg

Koji, thank you for your question. Let me be 100% clear. This is conservatism. That's it. We are not seeing any change to our business.
I'll remind you, 90%-plus of our customers have been with us for greater than one year. We have seen no pullbacks. We've seen no pauses in this environment. We thought that it spoke volumes to the quality of our business to raise the second quarter and, quite frankly, to raise the year at all in an environment where most companies are lucky to be reaffirming.
So we want to very clearly send a message, the business is executing as we would hope it would, and we are seeing nothing in the back half, but at the same time, we want to show we're being conservative and thoughtful as it relates to the uncertain to the environment.

Christopher Greiner

Yes. Koji, just this week, we signed two very, very large contracts. So -- and I was very specific in my comments to say every metric we're looking at right now, just as consistent with what David said, would say that we would do our normal much bigger race. We just, as David noted, what is happening by other public companies right now is we wanted to follow suit.

David Steinberg

Yes. Listen, we listened to or read the transcripts of, call it, 100 different earnings calls before getting to ours. And we wanted to make sure that we were showing shareholders and potential shareholders, not just that the business is executing incredibly well. But that we were meaningfully de-risking our back half in a situation where we could then show that we would be able to continue to beat and raise as a company and continue to execute.

Koji Ikeda

Got it. And maybe just a quick follow-up here. On the ARPU front, we did notice that sequential decline. I heard you on the prepared remarks, LiveIntent is playing a factor there. Presumably, there's some political -- postpolitical cycle deflation in ARPU happening there, but I just wanted to be sure there isn't anything else to call out in that ARPU number for the first quarter.

Christopher Greiner

No. And that ARPU sequential decline is pretty typical from the fourth quarter to a first quarter, Koji, if you look at it on a year-over-year basis, whether it's in total, up 12% or even the superscale ARPU, which is up almost 25% year over year. Yes, we're -- and by the way, we added 11 super-scaled customers sequentially quarter-to-quarter. It's one of our bigger jumps we've had. So now ARPU looked great to us.
And I mentioned the fastest-growing cohort are those scale customers using now three or more channels is up 42% year over year.

David Steinberg

We were particularly proud Koji to get to 159 super-scaled customers, and that doesn't include a number of superscale clients we've signed in the last 30 days.

Operator

Brian Schwartz, Oppenheimer.

Brian Schwartz

David, I wanted to circle back on the AI Agent Studio product. Just wanted to ask you what you're seeing in terms of consumption trends. Are the early customer adopters? Are they fully scaled in production with Agentic AI and they're coming back to buy more from you? Or is it still early in most arts and piloting the deployments?

David Steinberg

No, Brian. We have a meaningful percentage of our customers who have adopted the Agentic AI. And I believe the number is customers who have adopted it are growing at greater than 40% from a consumption revenue perspective year over year. So you're seeing meaningfully faster growth from organizations that adopt the Agentic AI. The big leap, which is in beta is stringing them together, which is an Agentic Workflow.
And I want to be clear, when you have four agents strong together, each agent that you add to the equation is a step function better than having just one. So one agent would be regular two would be a step function better three would be two step functions better and four would be three step functions better. So we're seeing massive opportunities around that beta that I think will really bear fruit in the back half of the year.

Brian Schwartz

David. And then the follow-up I have for Chris, what can you share with us from what you're seeing in the chips in terms of the leading indicators? Just thinking about your pipeline, maybe the rate of RFPs, the cadence of conversions, the size of your lands, how have they been trending in Q1 and year-to-date?

Christopher Greiner

Thanks, Brian. Trending similar to how they ended last year, sales pipeline growing faster than the overall revenue of the company, which has always been our model, and that continues to be the case. Very healthy RFP backdrop. And then sales productivity where at all really the rubber hits the road, we're seeing very strong sales productivity from all of our cohorts. Those sellers that are in their first year, that 12- to 24-month period and then our most experienced sellers continue to be our most productive.

David Steinberg

Yes. I mean, Brian, I'll just make an off-the-cuff comment that I've never seen our pipeline RFPs or conversions at a better place than they are right now. We've signed three of the biggest deals we've ever done in the last 90 days.

Operator

Zack Cummins, B. Riley Securities.

Zach Cummins

David and Chris. Chris, I wanted to start off. I don't think I saw it in your script or within the presentation, but in terms of all the new scaled customers added here in Q1, is it fair to assume essentially all of those are from the organic business rather than from any live intent customers?

Christopher Greiner

It's all fully integrated now. There are some LiveIntent customers that are contributing to it, but it's all fully blended now. That obviously has the -- if it's helping skilled customer count, it's a drag on scaled customer ARPU because they tend to have smaller ARPUs to begin with. But really, I mean, could not be happier with how skilled customer count in total start of the year.

David Steinberg

And by the way, they didn't have many large customers. So even in the case where somebody might have been attributed to them, we would have had to grow them in many cases to that rate to get them to there.

Christopher Greiner

Yes. Incidentally, we talked about earlier, but it just came to mind, one of the verticals that contributed the most scaled customers quarter-over-quarter was actually consumer retail.

Zach Cummins

Got it. That's great to hear. And my one follow-up question, maybe geared towards David. I know most of the major holding companies, at least on their commentary in mid-April, with the macro uncertainty, we're talking about leaning into their more durable businesses, which tend to be the data-driven marketing initiatives. So just curious if that's potentially accelerating any opportunity that you see with these agencies?
Or just any update around transits conversations with major holding companies?

David Steinberg

Yes. No, it's meaningfully accelerate there. We've had -- we opened -- we talked about the fact that we've had one agency holdco that we've worked for a long time that continues to grow nicely. We had one agency holdco that continues to scale at rates that we've not seen before. We now have a third one scaling at the same rates as the second.
And I think a lot of that goes back to our ability to provide them with data-driven marketing. I think it's also really important to note that even the three of the -- we've got five of the eight large agency holdcos, you're starting to see number two, three, and eight scale meaningfully right now.
And I think part of that goes back to two things. One, we are often their most profitable partner. So as they're looking at uncertainty, they're looking to drive greater margin opportunity for themselves by using the Zeta marketing platform and our data cloud; and two, back to your point, the ability to show true return on investment to their clients.

Operator

Gabriela Borges, Goldman Sachs.

Gabriela Borges

David, I wanted to ask you a little bit about what you're seeing with replacement cycles for some of the marketing cloud competitors that have talked about having their RNA strategies. Just curious if you're seeing a continuation in the acceleration trend that you've talked about previously and if there's any other color you had out here.

David Steinberg

Yes. Thank you, Gabriel. That is a great question, and the answer is yes. We continue to see the replacement cycle, which we believe is still in early innings accelerating. And it's interesting because when you think about our platform, because we rearchitected and relaunched in 2021 by putting AI agents and data as native to the application layer, there's no latency in the processing inside of our Marketing Cloud, whereas all of our competitors currently are operating with AI outside of the core cloud.
So you leave the cloud to the algorithm to acquire algorithm then has to do a data dip to go back to the algorithm to make an answer and then go back and tell the marketing cloud what to do. That tech debt is destroying return on investment for enterprises. And we are seeing, as a part of that marketing cloud replacement cycles are continuing to grow, and we expect that trend to continue.

Operator

Clark Wright, D.A. Davidson.

Clark Wright

Awesome. There's been some recent events within the walled garden. I was wondering if you could talk about maybe the opportunity that a fragmented ecosystem might have for Zeta, especially with global agencies since they typically start with social media at their first use case.

David Steinberg

Yes. I mean I wouldn't -- I mean, I think people are still talking about how Microsoft was supposed to be getting broken up. What I will caveat Clark, is that if there is a meaningful breakup in the ecosystem of the large walled gardens, that would be a massive opportunity for Zeta to work with those new assets by some interesting assets or partner with interesting assets. And we're already looking at what that could look like. But once again, I don't want to put the cart before the horse here.
I'm not sure we'll see anything happen anytime soon. Although if it does, that would be a meaningful accelerator to our business.

Clark Wright

Got it. And then as a follow-up here, you talked about multiple multiyear deals. Can you talk about maybe the change in duration of contracts that you've seen over the last few years? And what do you expect and which is implied by the guide I said you first?

David Steinberg

So interestingly enough, I would say most of our early agency holdco deals had no contract. It was a monthly agreement that we would just agree to operate together and go from there. And we now have multiyear deals on the three largest and scaling the fastest and at least one-year deals with everybody else. So it's really giving us a lot of long-term visibility. Our largest agency holdco client, where, quite frankly, we're working across dozens of brands, and we're working across mobile agencies that are a subset of it.
It rolls up to one enterprise. But in that case, they just signed a meaningful multiyear deal, which we were very pleased about.

Clark Wright

Awesome. Thank you.

Operator

We have reached the end of our question-and-answer session, which concludes today's teleconference. You may disconnect your lines at this time. Thank you for your participation.