Logo of jester cap with thought bubble.

Image source: The Motley Fool.

DATE

Thursday, May 7, 2026 at 5 p.m. ET

CALL PARTICIPANTS

  • President and Chief Executive Officer — Hayden Brown
  • Chief Financial Officer — Erica Gessert
  • Head of Investor Relations — Gary Fuges

TAKEAWAYS

  • Guidance—Revenue -- Full-year 2026 revenue is projected at $760 million to $790 million, reflecting both a reduction and a wider range versus previous guidance.
  • Guidance—Adjusted EBITDA -- Full-year 2026 adjusted EBITDA is forecasted at $250 million to $260 million, representing an adjusted EBITDA margin of 33%.
  • Guidance—Non-GAAP EPS -- Full-year 2026 non-GAAP diluted EPS is expected to be $1.50 to $1.55; Q2 2026 is guided to $0.35 to $0.37.
  • Marketplace Take Rate -- Reported marketplace take rate was 19.4% for the quarter.
  • AI-Related Work Growth -- GSV derived from AI-related work exceeded $300 million on an annualized basis and grew more than 40% year over year; this represented 8% of marketplace GSV in Q1, and 11% of job posts were in AI categories.
  • AI Risk Exposure -- Approximately 10% of GSV is estimated to be in the AI at-risk category, down from 11% a year ago, with the remainder comprised of net new AI work and AI-insulated contracts.
  • Cost Reduction Measures -- Announced restructuring, including a 24% reduction in total workforce and an estimated $70 million annualized cost base reduction, is expected to contribute $40 million in 2026 savings and incur one-time separation charges of $16 million to $23 million mostly recognized in Q2.
  • Business+ Product Performance -- Business+ plan posted 34% quarter-over-quarter GSV growth as Upwork expands into the SMB segment.
  • Lifted Enterprise Pipeline -- Enterprise subsidiary's pipeline for new clients grew 3x, and for existing clients 9x, with initial customer migrations scheduled in June.
  • AI Initiatives -- Human-supervised AI agents were in beta during Q2 and are planned for full rollout by year-end.

Need a quote from a Motley Fool analyst? Email [email protected]

RISKS

  • Geopolitical instability and war concerns have slowed higher-value contract volumes, while accelerated AI adoption has reduced client activity for contracts of $500 and below.
  • CFO Erica Gessert explicitly cited, "macro factors, most notably the combination of ongoing inflationary pressures from tariffs and new ones from soaring energy prices as well as persistent higher interest rates were impacting their business and spend." as directly impacting customer spend, notably among very small businesses and at the low end of contract values.
  • Management lowered revenue guidance and broadened the outlook range due to "the increased uncertainty in demand," and rapid late-quarter demand declines.
  • One-time charges of $16 million to $23 million related to workforce reductions will primarily be recognized during the current quarter.

SUMMARY

Upwork Inc. (UPWK 5.17%) reported that its AI-related work now comprises 8% of marketplace GSV, with AI job postings showing even stronger momentum at 11% of total posts. The company introduced an updated, task-level analysis to quantify AI-driven risk, identifying 10% of GSV at heightened risk, down from 11% last year, while emphasizing that net new AI work continues to substantially outpace AI-vulnerable categories. Structural cost reductions totaling an expected $70 million annually were announced through a workforce reduction of 24%, which management stated would accelerate profitability and operational agility. Product innovation, highlighted by the beta launch of human-supervised AI agents and ongoing expansion into large enterprise accounts, was positioned as key to capturing new AI and enterprise-driven demand channels. Management presented its revised revenue and profitability guidance as accounting for both cyclical macro softness and the effects from rapid changes in customer activity patterns, particularly at the smallest contract and client levels.

  • CFO Erica Gessert stated the new AI exposure analysis uses an LLM classifier informed by methodologies from Anthropic and Stanford for higher precision in risk quantification versus prior approaches.
  • The restructuring is expected to enable Upwork to reach a 35% adjusted EBITDA margin in the back half of 2026, more than two years ahead of prior schedule.
  • Management highlighted a robust AI-driven enterprise pipeline, noting that approximately 20% of net new enterprise opportunities are AI-related work.
  • Initial migrations for the new lifted enterprise product are targeted for June, and its pipeline was described as "extremely strong" by management.
  • Upwork's announced AI data strategy, leveraging its 3 million annual job posts, is in exploratory and pilot stages and positioned as potential upside to current financial model expectations.

INDUSTRY GLOSSARY

  • GSV (Gross Services Volume): The total value of payments made by clients to freelancers and agencies on Upwork’s platform before deducting service fees or take rate.
  • Take Rate: The percentage of total GSV that Upwork retains as its revenue.
  • Business+: Upwork’s fastest-growing product targeted at small and medium businesses, designed to facilitate larger and more complex client projects.
  • Lifted: Upwork’s enterprise solution and product line focused on large customers and complex engagements, referenced as a wedge into the large enterprise market.
  • AI-At-Risk: Platform work categories or tasks that management’s analysis deems most likely to be replaced by automation due to AI advancements.

Full Conference Call Transcript

Gary Fuges: Thank you, and welcome to Upwork's discussion of its first quarter 2026 financial results. Joining me today are Hayden Brown, Upwork's President and Chief Executive Officer; and Erica Gessert, Upwork's Chief Financial Officer. Following management's prepared remarks, they will be happy to take your questions. But first, I'll review the safe harbor statement. During this call, we may make statements related to our business that are forward-looking statements under federal securities laws. Forward-looking statements include all statements other than those of historical fact. These statements are not guarantees of future performance, but rather are subject to a variety of risks, uncertainties and assumptions. Our actual results could differ materially from the expectations reflected in any forward-looking statements.

For a discussion of the material risks and other important factors that could affect our actual results, please refer to our SEC filings available on the SEC website and on our Investor Relations website as well as the risks and other important factors discussed in today's earnings press release. Additional information is also available in our quarterly report on Form 10-Q for the quarter ended March 31, 2026, which was filed today. In addition, references will be made to certain non-GAAP financial measures. Adjusted EBITDA, adjusted EBITDA margin and free cash flow are non-GAAP financial measures, and all other financial measures are GAAP unless cited as non-GAAP.

Information regarding non-GAAP financial measures, including reconciliations to the most directly comparable GAAP financial measures can be found in the press release that was issued this afternoon on our Investor Relations website at investors.upwork.com. Finally, unless otherwise noted, reported figures are rounded in comparisons to the first quarter of 2026 are to the first quarter of 2025. And with that, I'll now turn the call over to Hayden.

Hayden Brown: Good afternoon, and welcome to Upwork's First Quarter 2026 Earnings Call. Q1 was a dynamic quarter that demonstrated our ability to produce value under a range of market conditions. We delivered in-line revenue and phenomenal bottom line results with a $10 million adjusted EBITDA beat. Upwork continues to be uniquely positioned to power the AI-enabled labor market of the future. The dynamism this quarter was notable, and we're going to spend today's call unpacking this moment and the opportunities in front of Upwork, which continue to be formidable. Starting in late February, we began to see two clear trends. First, geopolitical instability and concerns around the war began slowing the volume of higher-value contracts.

Secondly, accelerated AI adoption degraded the volume of client activity on the low end, impacting contracts of $500 and below. While this dynamic is not new, the pace of AI automation was faster than previously seen. We see the impact of AI in two discrete ways. On the low end, simple tasks are getting replaced with AI tools. Conversely, GSV from AI-related work exceeded $300 million on an annualized basis, growing at more than 40% year-over-year.

Through an in-depth analysis that Erica will touch on in more detail, we have determined that the portion of our GSV at high risk of being replaced by AI has continued to shrink, while we anticipate the market opportunity for AI-related work will far exceed this. While the overall impact of AI is marginally a net headwind for Upwork today, we are confident it will become a longer-term tailwind. Importantly, our strategy of growing larger complex work with bigger customers is working and will have a greater impact in future quarters. We have yet to fully tap additional growth levers, including launching agents in our marketplace and AI data opportunities.

As all these efforts begin contributing more growth, they add firepower to offset cyclical near-term labor market and AI factors in driving business outcomes. With the ongoing geopolitical uncertainty and continued softness amongst our lowest value contracts, we are adjusting our 2026 revenue guidance to account for the continuation of these trends. At the same time, we are focused on executing our cost management plans and have moved swiftly to pull key levers in the business. We launched multiple new AI-related features and exciting products in the quarter and showcased our spring Upwork updates earlier this week.

In Q1, we announced leadership adjustments to bolster our execution and pave the way for a cost management program that we are announcing today, which will further reduce our OpEx. These measures give us confidence in increasing our adjusted EBITDA outlook for this year. We have been on a journey of transforming our business as recapped in our Investor Day in November. The new Upwork is focused, nimble and AI forward. In Q1, we saw that our strategies around SMB, enterprise and AI are delivering above our expectations. First, we have great traction with SMBs and enterprises.

Our Business+ plan continues to be the fastest-growing product in our company's history, growing 34% quarter-over-quarter as we expand into the $530 billion market with SMBs. In our listed subsidiary, our enterprise pipeline grew 3x for new clients and 9x for existing clients in the quarter. The new lifted product is on track for the first customer migrations in June and is our differentiated wedge into the $650 billion enterprise market. These successes will continue to drive spend per client up and underscore our unique market positioning and success in attracting and converting larger customers.

And with our expanded OpenAI relationship, including the launch of the Upwork app in ChatGPT last month and a robust road map in this area, we are just scratching the surface on a promising new set of demand channels. Second, our SMB AI value proposition continues to expand with our investments in Ooma, our AI work agent. This quarter and in our spring 2026 Upwork updates, we launched features which make Ooma a more capable agent that synthesizes hiring insights, assists in evaluating candidates and helps move decisions forward so SMBs can hire faster and with greater confidence.

We also are rolling out a new project continuity offering in which we seamlessly source replacement talent if a hire is not working out. Third, we're also becoming the AI diffusion layer for SMBs, providing the AI skilled talent they need to create value from this exciting technology. AI-related work in our marketplace continues to grow and in Q1 was 8% of our marketplace GSV, with low-end AI-impacted jobs on the platform continuing to burn off and the net new AI work continuing to grow substantially, AI overall is an enduring massive opportunity for Upwork. And AI is adding a tailwind to our enterprise business with approximately 20% of our net new client pipeline being AI-related work.

We are in the early stages of ramping our products to tap into this large and growing AI opportunity. Additionally, our platform uniquely positions us to offer human-supervised agents at scale. The AI agent market is estimated to be $120 billion by 2028. We're building on our strong marketplace chops to extend our supply and demand sides to include AI agents. Our human-supervised agent solution was in beta in Q2 and is receiving extremely positive customer feedback. Businesses report an ongoing and increasing need to add a human layer of oversight to AI agents, and we are evolving our marketplace to meet this demand.

The size and scale of our talent asset is a tremendous differentiator and will underpin a new playbook for how SMBs can leverage AI safely and securely with humans in the loop. This exciting addition to our marketplace will roll out to all customers this year. Finally, we are just scratching the surface on our AI data opportunity. As AI data needs evolve from synthetic and static data sets toward real-time learning scenarios, we believe Upwork has a unique role to play with 3 million job posts per year across more than 125 categories. We are in early stages of exploring what this could look like in a way that enhances our platform and adds value for customers.

We're building rapidly, and we're focused in this area, which would be upside to our model. At the same time, as we drive these strategies forward, our multiyear investments in cost management and take rate levers bolster our strong financial position and give us the freedom to navigate choppy waters on our own terms. Building a more efficient, resilient organization that is best positioned for the opportunity ahead requires us to say goodbye to many talented people who have been vital to our journey. We are thankful for their impact and dedicated to supporting them through this period of change. Efficient, disciplined operations and strong profitability will remain a critical feature of how we operate.

Our control over adjusted EBITDA comes from the growth of our high-margin offerings, our ad and monetization levers that improve take rate and our internal operational efficiency. These measures bolster our confidence in increasing our adjusted EBITDA outlook for this year despite the softer demand environment. Our confidence in the positive outlook for Upwork and its essential role in the new AI world of work is stronger than ever. There is no doubt that AI is changing the future of work. Against this backdrop, we are building Upwork to power the AI-enabled labor and agentic market of the future. We're excited to drive the business through this transformation of work and the workforce.

And with that, I'll hand it over to Erica to discuss our financial results in more detail.

Erica Gessert: Thanks, Hayden. I first want to unpack the trends that we saw in Q1. Leading up to our Q4 earnings call on February 9, our GSV and active contract trends were within our expectations for the year, right on plan and within the 99th percentile of our outlook for the year. Starting in late February through early April, demand trends slowed materially. We heard directly from customers that macro factors, most notably the combination of ongoing inflationary pressures from tariffs and new ones from soaring energy prices as well as persistent higher interest rates were impacting their business and spend.

In some places, particularly at the lowest end, customers are adopting AI tools to supplement their needs as they experience financial pressures. Consistent with this, we continue to see the greatest pressures on our business at the lowest end of contracts at the sub-$500 level. These trends have since stabilized, albeit at lower growth rates and consistent with the volume trends we have seen in our business over the past 2 years rather than the heightened growth expectations we expected at our Investor Day 6 months ago. While our growth expectations are tempered for this year, signals of strength within the key growth pillars we have identified for our business give us confidence in the long term.

We know that there are questions about the impact of macro versus AI on our business. To disentangle these impacts, we have built a new in-depth analysis of our AI risk exposure. This new analysis uses an LLM classifier trained on methodologies from Anthropic and Stanford, amongst others, to assess AI automatability at the underlying task rather than the contract or job level. This is a more advanced approach than our prior job level analysis, and therefore, we believe more comprehensively assesses the true AI risks and opportunities we see in our business.

This methodology estimates that tasks representing approximately 10% of the GSV on our platform are in the AI at-risk category and tend to be concentrated in our smallest jobs. This is down from 11% a year ago and declining. The remaining 90% are net new AI work and AI insulated jobs and contracts because the majority of the underlying tasks in these jobs are considered insulated from automation by leading frontier AI companies or they are AI work itself. We expect our AI exposure to continue to diminish over time as new AI work and AI resilient jobs are the fastest growing on our platform and AI-exposed jobs concentrated in our smallest contracts are declining the fastest.

I now want to shift gears to the restructuring actions that we announced today. We have always said that we saw ongoing cost optimization opportunities in this business, and we've been preparing ourselves for the next stage of operational excellence and increased productivity. Our focus on our cost structure and internal tool development are enabling us to not just commit to our profitability goals, but to raise them. Our results in Q1 show that we have the discipline to produce strong profitability and free cash flow regardless of the vagaries of the broader macro environment.

Today, we announced the next step in this direction by reducing our cost base by an estimated $70 million on an annualized basis over the next few months. This will include an approximate 24% reduction to our total workforce. These decisions are never easy, but we will treat our colleagues with transparency, fairness and dignity as we work through our plans. We expect to incur onetime charges in the range of $16 million to $23 million, primarily related to severance and other separation costs, the majority of which we will recognize this quarter. Turning now to guidance.

While we remain confident in and committed to the growth levers for our business that Hayden outlined, we also acknowledge that we are operating in an environment of heightened volatility this year. As such, we are both lowering and widening our top line outlook for 2026. For the full year 2026, we now expect revenue growth in the range of $760 million to $790 million. This reflects the increased uncertainty in demand, as I previously discussed, and assumes heightened external pressures throughout the year. As planned, we will continue to expand take rate this year as a part of our overall growth story.

We expect full year 2026 adjusted EBITDA of approximately $250 million to $260 million, which represents an adjusted EBITDA margin of 33%. We're able to raise our guide because of our ongoing disciplined cost management and the restructuring we've announced today, which will contribute approximately $40 million in savings in 2026. With this raised guide, we now expect to achieve our 35% adjusted EBITDA margin target in the back half of this year, more than 2 years ahead of schedule. We expect full year 2026 non-GAAP diluted EPS to be between $1.50 and $1.55. Our updated guidance also assumes stock-based compensation of approximately $60 million to $65 million for the full year 2026, which reflects the impact of the restructuring.

For the second quarter of 2026, we expect to generate revenue in the range of $187 million to $193 million and adjusted EBITDA in the range of $56 million to $59 million, which represents an adjusted EBITDA margin in the range of 30% to 31%. We expect Q2 2026 non-GAAP diluted EPS to be between $0.35 and $0.37. In closing, while the trends this year continue to evolve, two things remain true and very consistent for our business. The growth levers we've identified are showing promise and our confidence in producing growing profitability and free cash flow in 2026 and beyond is unchanged.

With a clear vision for our growth levers and a firm handle on our cost base, we are in control of our destiny. We have ample cash to pay down the debt on our balance sheet if we choose to, invest in strategic M&A as opportunities arise and return money to shareholders. Our confidence in Upwork's ability to grow and take our rightful place in the future of work is steadfast, and we will continue to execute with strength and efficiency in the new world of work. And with that, we will now take your questions.

Operator: [Operator Instructions] Our first call comes from Maria Ripps from Canaccord.

Maria Ripps: Can you maybe help us understand how to think about your enterprise business sort of in the second half of the year with this new guidance? Do you still anticipate to see some benefits from the launch of lifted here in the second half? And then secondly, just broadly, given your new guidance, it feels like things have changed very rapidly over the past 90 days. I guess what gives you confidence that the current framework sort of captures the right range of outcomes for the rest of this year?

Hayden Brown: Thanks, Maria. On the lifted side, we're really pleased with the progress and everything in that part of the business is 100% on track. The pipeline is extremely strong. Customer feedback and demand continues to mount. And even though our fully integrated product following our acquisitions is not yet live, we've already received recognition in the form of a star performer or top quadrant rating from the Everest Group in a competitive landscaping study. And these ratings are a big consideration for customers in the whole kind of RFP and selection process. So all of this is really validating our approach and the acceleration that we are going to see here.

And certainly, we're doing everything we can to even pull that forward where we can. But June is when we're going to be migrating customers, and we are going to see that back half impact, which Erica can share a little bit more about.

Erica Gessert: Yes. So to be clear, Maria, there's absolutely no change in our outlook for lifted, and we still feel really confident in the kind of the 25% GSV growth that we provided for the outlook for the year. The changes in our trends we really saw on the marketplace side -- and it really was kind of distinctively a change in trajectory starting really in late February and deteriorating into March and early April. Those trends have since stabilized, and we can talk more about kind of the underlying impacts to kind of marketplace volumes, but we talked a little bit on the call about really the declines being at the lowest end of contracts.

We have prudently -- despite the kind of stabilization we're seeing right now, we have prudently forecast through the rest of the year some ongoing deterioration in marketplace trends just given the rapidity of change that we saw at the end of Q1. So we do feel that we have worked into our outlook some additional kind of deterioration and conservatism just given the volatility of the trends. I'd also just emphasize, we do continue to see strength in the growth pillars that we've identified. We're excited about it lifted the SMB growth with GSV up 34% quarter-over-quarter for Business+.

All of these also give us some very good contra indicators to some of the weakness we saw at the low end.

Operator: Our next question will come from Matt Condon of Citizens Bank.

Matthew Condon: My first one is just on -- were there any new categories that were impacted by AI in the quarter? Was there any shift in the actual category impact? And then I know that it's 10% today, but as model capabilities continue to improve, what gives you the confidence that, that 10% is not going to go to 15% and 20% here in the future?

And then a follow-up just on enterprise, along the AI lines as well, like what makes you -- gives you confidence that enterprise clients aren't going to lean more heavily on their internal workforces with AI versus needing to go outside the freelancer platforms such as Upwork for external talent, just given the efficiency and productivity gains from AI.

Hayden Brown: Sure. So on the category side, there really were no big changes in the quarter. I would say the thing that we would continue to focus on is the AI work category, which is a tremendous bright spot for us. It keeps growing above 40% year-over-year. And really, the leading indicators there are even more promising. We saw in the quarter that 11% of our job posts were in the AI category, even though GSV, which is more of the lagging metric, was only at 8% from that category. So we're going to be capturing more and more of this opportunity going forward.

Stepping back and to your question about why do we feel good about this 10% kind of boundary around where our exposure is now, there's a few reasons for that. First of all, the reason we have an updated number is really the methodologies that came out in the last few months from folks like Anthropic and Stanford gave us an opportunity to revise our approach and incorporate some factors that were not previously available. And that's where we really broke down all of the jobs in our marketplace to their subtasks that contribute to getting those jobs done. And then we classified every task as whether it could be automated or not.

And that is the exercise that led us to see that 10% of our GSV today is in jobs with tasks that theoretically can be automated depending on the capabilities of different AI tools deployed. So you're right to question like, okay, where does this go from here? And actually, the good news is that all of the trends that we're seeing and the strategies we're executing are what really reduce our exposure over time.

So first of all, we've seen in both the old methodology and this new methodology that the trend has been consistent that our exposure to AI automation has actually been diminishing, and it was down significantly this quarter versus last quarter -- last -- a year ago, even though there was some acceleration because these tasks are basically burning off the platform and the residual work is AI insulated. So that's just the reality of what's been happening and what happened in Q1. Also, we see that enterprise, as an example, and to your question, is one area where we actually have a very clear AI tailwind in the pipeline.

20% of the opportunities we're looking at are in contention for today are actually due to AI work. And so we're really not seeing any indicator that enterprises are turning more inwardly. In fact, it's quite the opposite. They're looking for skilled and capable both vendors and teams to come in and help them with expert work, and that's what we're doing for them. Then we've got helpful areas like our SMB program. That is really attracting larger, more complex projects that are AI insulated. And that's why we've been investing in this for several years now, and we're seeing great growth with 34% GSV growth in that product just quarter-over-quarter.

Finally, as the mix shift in our business continues to evolve, we just have less exposure because, again, these AI categories that are growing so quickly are taking up more and more of our mix and represent a very insulated part of the business. The final thing I'd say is that the AI strategies we're deploying like launching human supervised agents in our marketplace -- this really gives us another way to keep spend on our platform even as it gets automated because the humans on our platform are the on supervising agents doing that work and the funds flow through us in that instance. So that product is launching later this year. It's live in testing now.

So the market is absolutely going to keep evolving. We're sharing all the insights we have today and really have the confidence given our growth levers that each of them is really positioning us to minimize AI exposure and expand our market opportunity as we move forward.

Operator: Our next question comes from the line of Bernie McTernan from Needham & Company.

Bernard McTernan: Maybe to start acknowledging that 40% growth for AI work is still obviously extremely robust, but it is a slight slowdown from the 50% growth last quarter. Just was first wondering what the underlying trends that you're seeing there. And then second, you mentioned leveraging data. And I think -- but can you just talk more about that data opportunity within AI? I think you mentioned job posts and leveraging all that information that you have, but can you just maybe expound upon that? And then third, I think we normally get it in the presentation, but I don't see it. But if you could just provide the marketplace take rate, that would be helpful.

Hayden Brown: Sure. Let me start on the AI growth. We feel really good about the trajectory of this. The volume of customers looking for AI work on our platform is growing and growing. And this is an extremely dynamic space. So of course, there are going to be quarter-to-quarter fluctuations. But Q1 this year was growing faster than Q1 last year, even though that's on a bigger base. We also see these really strong leading indicators, as I mentioned, around our job post mix shift is increasingly AI-related work, and that's -- this is just a giant market that's just growing.

And frankly, businesses are still waking up to the fact that they need to deploy these new technologies and that Upwork is the place to find that talent. And that's what we see in these indicators. It's really getting traction. To your question around the AI data opportunity, -- we think this could be a very meaningful opportunity for us. As AI data needs are evolving, and we see this across the ecosystem where it's going from these kind of synthetic and static data sets towards more real-time learning scenarios. I think a good example of this is what Meta just announced in the last few weeks around tracking computer activity inside their own organization.

There is a real need for companies to have more of these environments to deploy agents within to learn and to access the data that is generated from work activity actually happening end-to-end. So we've been approached by numerous players on this topic. We know we have a unique role to play with 3 million job posts per year in more than 125 categories. And I'd say we're just in the early stages of exploring what this could look like in a way that is value additive for our customers and for our business. And so it's too soon to say what exactly that will look like, but we've got some conversations and pilots in flight.

And anything we do end up doing here will be incremental to our current outlook.

Erica Gessert: Bernie, just to answer on the marketplace, a great question of 19.4%.

Operator: Our next question will come from the line of Brian Tomahen from UBS.

Joshua Chan: This is Brian on for Josh Chan. Just a couple of questions on my end. I guess can you explain how the restructuring came about? What's changed from 3 to 6 months ago to now? And then I have a follow-up.

Hayden Brown: Sure. This is something where we have, I think, quite a track record of really disciplined cost management and efficiency in this business, and we've been on that path for several years now. And so as we've continued to see those plans come to fruition, some of the multi-quarter initiatives around automation in our own business, which I know we've shared with you guys, have been making a lot of progress. And as we saw market conditions getting a little bit more choppy, this was the right time to make these changes, both to ensure we are delivering everything we want to be delivering on the profitability side.

But frankly, also just to make sure we're leaning into every opportunity to be more nimble, more fast moving. We've seen benefits as we've made kind of similar changes in the past and have really unleashed faster innovation cycles in the business, and we have that opportunity now. So these are tough decisions because they do impact our people in a meaningful way, but we know they're the right ones to set up work out for the next chapter.

Joshua Chan: Yes. And then last question, are you seeing non-AI GSP decelerating at an accelerating pace? And if so, where may that be?

Erica Gessert: Briana, yes, look, if you really take a step back and kind of look at the dynamics on the platform, where we're seeing the deceleration is really with the very small businesses on our platform. And as we talked about with our growth pillars, enterprise pipeline is super strong. The SMB kind of work that we've done on the platform to attract larger customers above kind of 50 employees and more is really continuing to grow and improve. So where we've seen the weakness is really in very small businesses, which are most affected in this macro environment and where wallets get tight, both for them and for consumers.

And then below that, where we're seeing it is in the smallest contract types.

Hayden Brown: So I'll just add to that. We've done some really extensive work to talk with these customers, survey our landscape and understand what's going on behind these changes. And just to underscore what we've shared, the things they are telling us are things like cost of living is up, so people are buying food and gas, not the product I sell or I work in a business that's automotive related and tariffs and government decisions have created a lot of volatility, so I'm spending less. There's just a ton of these stories coming into us about how businesses are just getting more cautious. And again, these are the very smallest businesses in the economy.

So when things change, they really feel it and they have to make decisions. And I think that's what's been flowing through to what we shared in terms of the trends at the end of the quarter.

Operator: Thank you. This does now conclude the question-and-answer session. We thank everyone for your participation in today's conference, and it does conclude the program, and you may now disconnect.