Follow Up Call Transcript

Operator

Good afternoon. My name is Krista, and I’ll be your conference operator today.

At this time, I would like to welcome everyone to Meta’s First Quarter 2026

Results Follow-Up Q&A Call.

All lines have been placed on mute to prevent any background noise. After the speaker’s remarks, there will be a question-and-answer session. If you would like to ask a question, please press star one on your telephone keypad. To withdraw your question, again press star one. We ask that you limit yourself to one question, and this call will be recorded. Thank you very much. Kenneth

Dorell, Meta’s Director of Investor Relations, you may begin.

Kenneth Dorell

Thank you. Good afternoon and welcome to the Follow-up Q&A call. With me on today’s call are Susan Li, CFO, and Chad Heaton, VP of Finance. Our remarks today will include forward-looking statements, which are based on assumptions as of today.

Actual results may differ materially as a result of various factors, including those set forth in today’s earnings press release and in our annual report on

Form 10-K filed with the SEC. We undertake no obligation to update any forward-looking statement.

During this call, we will present both GAAP and certain non-GAAP financial measures. A reconciliation of GAAP to non-GAAP measures is included in today’s earnings press release. The earnings press release and an accompanying investor presentation are available on our website at investor.atmeta.com. And now, I’d like to turn the call back over to Krista for the first question.

Operator

Thank you. Your first question comes from Mark Mahaney with Evercore.

Please go ahead.

Mark Mahaney

Okay, thanks. I just wanted to ask about two property questions, Threads and

WhatsApp. And maybe you provided some of the details, but any more color on the ramp of monetization, particularly on those two assets, and any particular new products or features within those that have really driven greater engagement? Thank you.

Susan Li

Thanks, Mark. We’ll start with Threads. So, Threads, the number of Threads’ monthly and daily active that continue to grow in Q1. I think the last update we shared was that we are now at over 150 million daily actives.

And we continue to see growth in the priority markets in particular that we’re focused on, like the U.S. and Japan, and priority verticals as well. So, growth in creator production in key areas like the NBA, K-pop, reality TV, and we’ve done this in part by investing in partnerships and product activations with creators.

1

We don’t have a specific update to share this quarter on Threads’ time spent, but I think Mark mentioned that Threads is making good gains and continuing on its trajectory to become the leading app in its category. And our focus remains on establishing Threads as the place people go to, to keep up with what they care about, and we’re prioritizing the quality and timeliness of our content recommendations to drive cultural relevance for the app.

You asked, I think, about monetization explicitly in Q1. We continued expanding ads on Threads to over 200 countries. We’re looking forward to reaching global availability over the coming months. We’re working to grow both demand and supply for Threads ads as we continue to optimize the ads shown on Threads, and that’s our typical playbook when ramping monetization on a new service.

So, we’re excited about the long-term opportunity with Threads. We don’t expect it to be a meaningful driver of overall impression or revenue growth this year. And then I think the other app you asked about was WhatsApp.

We continue to be optimistic about the longer-term opportunity with ads in

Status. Given the strong base of engagement, Status posts are extremely popular in countries where WhatsApp is used, particularly in Latin America,

Asia, and Africa, where it often outpaces Instagram in terms of posting.

Today, Status ads are now being viewed by hundreds of millions of people each day, and we expect to complete the global rollout of ads in Status throughout this year.

We’re also expanding the types of ads supported. So, that includes ads that click to a website and campaigns that are optimized for reach impressions or video views. But we don’t expect ads in Status to be a meaningful contributor to total impressions or revenue growth for the next few years.

That’s in part because we are in the middle of a slow rollout, which I think is a playbook that we understand well, starting at a low level of ads for an extended period of time before we gradually ramp supply as we optimize ad formats and performance.

And then some structural reasons why we would expect Status ads to monetize well below Instagram Stories longer term. The first is geo mix of

Status usage that skews more towards markets that have lower ad spend per capita compared to Instagram Stories, for instance.

And then we also expect ad targeting to be less robust for people who don’t have their WhatsApp account linked to the Meta Account Center, since we have limited information on WhatsApp.

Operator

Your next question comes from the line of Shweta Khajuria with Wolfe

Research. Please go ahead.

Shweta Khajuria

Thanks a lot for taking my questions. Can I please try two? First is you mentioned less personalized ads. How much of a headwind could that

2 potentially be? And also, the regulatory concerns around youth-related issues.

When the L.A. court ruling first came out, there was a couple days of growing concern that back half could see some rising costs. So, did you set some budget aside? How should we be thinking about that?

And then second is on just the overall within advertising growth rate and the durability of ad growth. So, what do you see as the biggest areas for runway for growth and durability across engagement where it’s recommendations or across monetization from optimized ad load to performance or new products and creatives or business messaging? So, you’ve got a lot of drivers, but how do you see in terms of the most impact from some of these? Thanks a lot.

Susan Li

Shweta, I’m having a little bit of trouble hearing the question. So, I’m going to do my best here, and let me know if I miss the mark. I think the first question was about the regulatory landscape. And I think I heard you ask about LPA specifically and maybe youth. So, I’ll touch on those two first.

The LPA question, just to set a little context, so we had aligned in December

‘25 with the EC on further changes to our consent model for personalized ads in Europe. In Q1 of 2026, users in Europe began to see the changes to the less personalized ads flow.

In general, we believe that the changes that we’ve had to make to address the

EC concerns go far above and beyond what’s required by the DMA itself, and our appeal of the commission’s decision will continue in the courts.

But with that said, we are in dialogue with regulators, and we believe that the revenue impact from the updated user flows will be larger in Q2 than it was in

Q1 since we were rolling out the updated offering through Q1. So, Q2 and quarters going forward will have the full quarter impact. And that expected impact is factored into our Q2 ‘26 outlook.

I think there was also just a question in there, maybe on the youth litigation landscape. I don’t know that we have a lot to say here, but I would just say that we remain very committed to building safe, supportive environments for young people, and we will defend our record here vigorously. And I think it’s premature to comment on any ultimate litigation outcomes. And then your second question, I had some trouble understanding.

Chad Heaton

I think there was something in there about the ad load, so -- and the durability of ad load in the future. So, maybe I’ll start with that.

So, I would say we continue to optimize ad load across our apps by determining the best time, place, and person to show an ad. This is enabling us to drive strong incremental conversions without having to rely solely on increasing ad load.

In some cases, optimizations do result in overall ad load increasing. So, for instance, ad load increases on Instagram Feed and Reels contributed to year-

3 over-year impression growth in Q1, but the larger driver of impression growth continues to be growth in users and engagement.

I would also note that we’re very measured about our approach with ad load to ensure that any changes that result in higher levels of ad load don’t trade off against the long-term health of our business or engagement.

Operator

Your next question comes from the line of Deepak Mathivanan with Cantor

Fitzgerald. Please go ahead.

Deepak Mathivanan

Great. Susan, first, I just wanted to ask, was there any signals of macro weakness you would point out either in 1Q or perhaps that you have factored into your 2Q guide, maybe in certain verticals? Anything that you can add there, that would be great.

And then second one, also wanted to ask about the compute capacity that you’re getting between the first party and third party. Beyond near-term availability, are there any specific aspects of the third-party deals that’s helping with perhaps maybe giving it for the internal workload while you use some of your own data centers coming live from model training?

Can you help us understand how you’re utilizing different types of compute, particularly in the next few quarters when some of these third-party deals come live?

Susan Li

Yes. So, on your first question on the macro landscape, we did see reduction in advertiser spend coinciding with the beginning of the conflict in Iran at the end of February, and that continued through the quarter. We saw the most pronounced impact within the Middle East user region, but we also saw some softer trends in markets outside the Middle East, including in the U.S. and

Western Europe.

I would say we’ve begun to see some signs of improvement in demand, both in the Middle East and around the world, as we’ve progressed into April. And our guidance, I would say, embeds a range of possible macro outcomes, ranging from continued improvement to macro deceleration. So, where we ultimately land within the range will depend on how demand progresses throughout the quarter, among other factors.

Chad Heaton

So, on your second question about the cloud capacity, so we’re contracting capacity from a number of third-party cloud providers to meet our infrastructure needs. These are multi-year contracts that will come online progressively over the course of 2026 and 2027, and the capacity will primarily support AI workloads.

We still anticipate investing significantly in our owned and leased data center capacity. However, a lot of that is going to be for capacity that comes online in future years. So, we’ve been signing cloud deals that will enable us to bring capacity online more quickly this year and next year.

4

Operator

Your next question comes from the line of Michael Nathanson with

MoffettNathanson. Please go ahead. Michael, your line is open.

Michael Nathanson

Yes. Hey, thanks. Hey, Susan. Well, I have two for you. One, on the OpEx guide, given the headcount reduction, I know there’s higher CapEx as well. I’m surprised it wasn’t changed in the outlook. Is there anything else on the cost side that may be offsetting cost savings from the employer reduction?

And then earlier on in the first call, you said that there’s been continual underestimation of compute needs. I wonder, has there been any impact on your revenue growth because of that underestimation, and what have been the biggest impacts, let’s say, to your business from underestimating compute?

Thanks.

Susan Li

Thank you, Michael. So, on your first question, the 2026 expense outlook, as you mentioned, is unchanged. We do expect lower going forward employee compensation expense versus what we would have expected last quarter, given the planned workforce reduction, but that is offset within this year by restructuring costs that we expect to incur as part of the layoffs.

And I will say that there are still many puts and takes with our expense outlook this year. A big one, frankly, is the timing of cloud deals coming online and when the capacity gets delivered, but there are many others. And the planning process overall is pretty dynamic. So, the range, I think, reflects still that there’s a pretty wide spectrum of outcomes.

On your second question, which is referencing us being compute-constrained,

I’d say a couple of things. We generally expect to make good progress alleviating our current compute constraints through the year as we bring more capacity online. That continues to be very dynamic.

I would be the first person to say we’ve underestimated our compute demand in the past, and I think we’ll get more visibility to the compute needs that we require to support our AI products as we launch them. So, right now, we’re also, in some sense, also planning for the future and planning to give ourselves the ability to scale inference capacity as product usage necessitates it.

We’re also -- we’ve talked a little bit about the planning process, that we have every budget in that process. We are both evaluating requests for headcount and capacity. And so, we’re pretty disciplined through that process about making sure that we are funding the engagement and revenue initiatives that are critical to driving the business engine of the company. Having said that, I’m not sure there’s a team at the company that wouldn’t like to get a little more compute right now.

Operator

Your next question comes from the line of Barton Crockett with Rosenblatt.

Please go ahead.

5

Barton Crockett

Hi. I was wondering, Susan, if you could address the question of how far you would push CapEx and still feel comfortable within your free cash flow and capital structure?

So, the percentage growth and the numbers that we’re looking at in CapEx right now, if that were to continue for a couple of years at this pace, you could be maybe wanting to borrow, lever up the balance sheet, digging into negative free cash flow. What’s the guardrail? How hard would you push it to pursue the opportunity you see?

Susan Li

This is a really good question. I’d say that right now, we really believe that we have very compelling investment opportunities ahead of us and that we think we really need to pursue them. They are critical to the consumer and business experiences that we expect that we are going to build over the next few years.

And I think Mark, in his remarks, I think gave some more color about really how he’s thinking about what we’re building and the degree to which we think that we are going to be building foundational tools and technologies that will be transformational for the people who use them.

And so, we aren’t really optimizing for a specific cash flow level this year.

Having said that, we continue to be very focused on operating the business efficiently to help offset our investments.

Operator

Your next question comes from a line of Ross Walthall with Cleveland

Research. Please go ahead.

Ross Walthall

Yes, thank you. You guys had some interesting comments on generative AI creative. I was just wondering, could you give us some examples of how advertisers are using those tools and why you mentioned some performance gains how that’s actually playing out?

Susan Li

Great question. So, we are seeing continued strong growth of our gen AI ad creative tools. Over 8 million advertisers are now using at least one of our AI ad creative tools.

That’s up from 4 million at the end of 2024. The vast majority of advertisers using our AI ad creative tools are SMBs, which I think is an interesting highlight in terms of the value the tools have for businesses that wouldn’t otherwise have had resources to invest in developing and optimizing ad creatives. Video features in particular are seeing strong growth in usage that contributed to an uptick in advertiser adoption in Q1.

So, image animation is a good example. We’re seeing strong growth with image animation as we’ve expanded eligibility to more surfaces. That now includes

Instagram Reels, Facebook Reels, Facebook Feed, and Instagram Stories.

We’re also continuing to expand testing of video generation to more advertisers. We expect to complete the rollout over the coming months. This

6 feature allows advertisers to transform multiple images using gen AI models into a dynamic multi-scene video with text overlays and music.

In a large-scale test that we ran, advertisers who used our video generation feature drove a 3% gain in conversion rates compared to advertisers not using the product. We think we’ll make more progress in our video generation tool this year as we continue to expand the capabilities of our offerings to meet the different video needs of advertisers. As you can imagine, advertisers who come to us might range from having no video assets at all to ones with meaningful existing assets and are trying to find different ways to optimize them.

So, we’re building different solutions for different needs. We’re also seeing growth from the expanded set of video solutions that we rolled out in Q4. That includes video filters, which automatically transforms videos to adjust contrast ratios or add more vibrant colors to make them more engaging.

Music generation generates music based on advertisers’ ad context and tone of message. That also makes the ad more interesting. And in Q1, we continue to expand tests of our voiceover translation feature to more advertisers.

This enables advertisers to automatically adjust voiceovers in an ad to the preferred language. And we’re also expanding image overlay translation, which translates text overlays in an image ad. And we have a handful of other things sort of in earlier stages of testing, but similar features that are intended to give advertisers a wider range of tools and features to make their ads more engaging.

So, we’re seeing good traction with our creative tools. There are lots of things we’re going to do to advance the offering in ‘26. As the models get better, we think the quality of the outputs will get better.

We’re very focused on expanding the capabilities and relevance of the tools to meet advertiser needs and getting better at understanding advertiser intent and brand context when we’re generating the output, and then finding more entry points in our tools so that advertisers have the tools ready and available to them at whatever is the point in the workflow that would make the most sense.

Operator

Your next question comes from the line of Josh Beck with Raymond James.

Please go ahead.

Josh Beck

Yes, thank you so much for taking the questions. I wanted to ask a little bit about the new Meta AI experience, and obviously, product discovery is very prominent as are other areas. But just double-clicking on that and the shopping aspect of it, what are maybe some of the obstacles ahead to maybe improve the conversion within this?

It seems like a lot of Meta’s shopping is more of an impulse buy. This is a little bit more of maybe a discovery process, just like what are you working on? It

7 obviously seems like it’s in a pretty good spot, but what are some of the key milestones for shopping within Meta AI that you’re working on?

Susan Li

Yes. We’re excited to have Muse Spark powering Meta AI across the standalone app and the website and direct chat threads on the various messaging apps. You referenced shopping specifically.

We’re excited about the shopping mode within Meta AI also. So far, we’ve made it available in the standalone app and website within the U.S., and while it’s early, we’re seeing people use it for product inspiration, browsing, product comparisons. We think there is a lot of opportunity to build engaging product experiences that will make it easier for people to discover and buy products across their services.

That’s one of the focus areas within shopping. Work that we’re going to be doing there, I think, is personalization is going to be a big angle. We have a strong track record of building highly personalized experiences, and I think making Meta AI into an assistant that can understand your context and preferences based on your likes and interests will be very helpful in all contexts, including shopping and similarly building the agentic capabilities to let it do things, including engaging commerce on your behalf.

This is obviously very early, but in general, commerce has always been a significant area of focus for our teams and, of course, our largest ads vertical. I think investing in making the Meta AI shopping experience very high quality is very adjacent to that.

Operator

Your next question comes from the line of Nick Jones with BNP Paribas. Please go ahead.

Nicholas Jones

Great. Thanks for taking the questions. I guess just one on Manus, that kind of acquisition is kind of getting undone here. Should we look for maybe additional

M&A or how you kind of fill in on the product suite that was kind of on the strategic roadmap there? Thanks.

Susan Li

I would say that we have a lot of confidence in sort of our agentic roadmap, I think, period. Obviously, we don’t have a lot more to share on Manus, but I think we’re very focused on building agentic capabilities for all the reasons that Mark outlined in his original comments. I think the two pillars of personalization and greater agentic capabilities are going to be a big feature in all of the AI tools and experiences that we build going forward.

Operator

Your next question comes from the line of Rich Greenfield with LightShed

Partners. Please go ahead.

Rich Greenfield

Hi. Thanks for taking the question. I have two pieces. One, Susan, I just wanted to go back to this Meta AI and sort of the connections, if you think about the agentic world. When you announced the launch of the new model, you specifically said it was designed to integrate everything that you do across your family of apps. Obviously, it doesn’t do that now.

8

How should we think about the roadmap from when it will understand context for your Instagram or understand where you’ve been or who you’re connected to on Facebook?

If it’s built for that, what’s stopping you from enabling that? And then even with existing connections like Gmail that I think were there because of the Meta

AI glasses integration, the Gmail integration looks very rudimentary relative to what I’ve experienced with other AI platforms across the web right now. How should we think about what’s there integration-wise relative to what’s coming?

Susan Li

Yes. I mean, I would say we are very early on this journey. I think Mark alluded to the kind of breakneck pace at which we have been ramping, frankly, the lab and getting to a place where now we have what we think is extraordinarily talent-dense lab and that has released a model that has sort of caught us up to being in the pack in record time.

But that is all to say that the vast majority of the journey is ahead of us, both in terms of our ability to release future models that we hope will put us closer to the frontier, to expand the capabilities of the models across multimodal perception, across tool use and multi-agent orchestration.

And so I think there’s really a lot more work to be done both on the models themselves, as well as integrating them more deeply into product experiences and our product roadmap. So while we’re excited and very happy with these sort of early results, we’ve seen the results from Meta AI with being powered by Muse Spark as an example.

And we’re excited about some of the things we’ve launched, some of the things we’ve talked about in this call, like shopping mode. I think we are, again, very, very early. And I think I don’t have anything specific to announce in terms of product launches, but we feel good about our position to uniquely deliver on

Mark’s vision of personal superintelligence.

I think the three things I might highlight there are first, our social graph and first party signals are going to allow us to ground agents that we build into who you care about or what you follow, what you’ve engaged with. So that should get us to more personalized outcomes than generic assistance. Second, messaging is a natural action, or interaction, surface for agents. It’s where people coordinate plans, where they already interact and transact with businesses.

And third, our hardware investments and Ray-Ban Meta Glasses provide what we believe is the best form factor for agentic interactions. We think that should be hands-free, real-time, and I think over time it should become really proactive. So, again, very early. I think there will be many years of product launches and releases from us to come.

Kenneth Dorell

Krista, I think we have time for one last question.

9

Operator

Thank you. And that question comes from Stephen Ju with UBS. Please go ahead.

Stephen Ju

Okay. Thank you so much. So, Susan, I hate using the anecdotal example here to ask a question, but I am now seeing what are probably more prominent calls to action in my Instagram feed to download Meta AI and start creating AI- generated content.

So, the optimist in me wants to start believing that we’re on the cusp of getting through that, making progress toward that second pillar of AI revenue that

Mark was talking about in early 2025. So, am I being too optimistic here or, better yet, where are we on the progress there in terms of using your own models to help, I guess, influencers and users generate content? And do you still have that partnership with Midjourney? Thanks.

Susan Li

Yes. So, we alluded to some of these things in some of our earlier comments. I think some of the most straightforward versions of this, for example, are the

Gen AI ad-creative tools that we talked about.

We are releasing sort of an ever-growing and richer tool set of features to help advertisers make better and more engaging ads. That’s, I think, very straightforward in terms of how adjacent it is to our business. We’ve also introduced the Meta AI business assistant that helps advertisers optimize their campaigns.

And that, in connection with the ad-creative tools I just mentioned, should make it easier for businesses to manage, create, optimize their ad campaigns.

And as we improve performance, that should accrue to revenue. We also think the opportunity to build engaging organic experiences to reduce the barrier to producing high-quality content, ultimately to open up new, more immersive experiences, should benefit organic engagement on the platform as well.

So, again, I would say we are quite early in the journey on all of these angles.

But these are things that I think don’t even really require that you unlock something meaningfully different from what is driving today’s core business to be very valuable to our platform. And then, of course, I think over the longer term, there are new revenue streams around how do we monetize building consumer and business AIs in ways that we alluded to earlier on the call.

So, I think, in general, we feel good about the progress we’ve made here. And I think your question maybe was specifically focused on content generation, which, again, I think ad-creative is probably the most directly commercial version of that. But we feel good about the overall roadmap of opportunities here.

Kenneth Dorell

Great. Thank you, everyone, for joining us. I know it’s a busy night, so I appreciate your time. We look forward to speaking with you again soon.

Operator

Ladies and gentlemen, this does conclude today’s conference call. Thank you all for your participation and you may now disconnect.

10