Information Services Group, Inc. (III) Earnings Call Transcript

Hi. Welcome to the Third Quarter 2026 ISG Global Index Call. I am Sumeet Jain, tech analyst with CLSA. And I'd like to thank the team at ISG for their valued work on the industry and for asking us to introduce this call today. ISG has been hosting these index calls on the IT and business services industry for more than 20 years. ISG influences $200 billion of technology spending each year, giving them deep insights into the industry as well as key changes in enterprise demand. I will turn the call over to Stanton Jones, distinguished analyst at ISG. Over to you, Stanton.

Thanks, Sumeet, and welcome, everyone, to our 96th consecutive ISG next call. With me today is Steve Hall, ISG Chief AI Officer; Kathy Rudy, Partner and Chief Data and Analytics Officer; Namratha Dharshan, Chief Business Leader for ISG India; and Alex Bakker, ISG distinguished analyst and Head of Primary Research. So for those of you that may be on your first in next call, just some quick background, the ISG Index measures the overall health and growth of the technology industry, which includes both managed technology services and cloud-based software and infrastructure services, and that's what we call our as-a-service market. And we do this by tracking and analyzing annual contract value, or ACV, and that's a leading indicator of where revenues are likely to be in the future. So just think of ACV as bookings. And we combine that bookings data with our primary research on enterprise buying trends, along with our on-the-ground experience advising G2000 firms on their technology and sourcing strategies to provide you with the insights you're about to see here today. Okay. So let's go ahead and start off with a level set on the market, Steve, over to you.

Great. Thanks a lot, Stanton, and thank you, Sumeet. So let's just start right off with the big headlines. AI infrastructure boom is absolutely accelerating. I think the core message this quarter is the infrastructure cycle is still gaining momentum, ISACB reached roughly $35.5 billion in Q3, which was up 100% year-over-year and more than 30% sequentially. The hyperscalers continue to post strong growth and are still committing extraordinary levels of capital to new capacity. I think that matters to the market, though, because it still does looks like it's supply-led. It looks really demand-led. Providers are continuing to build because enterprise demand for compute, storage and AI capacity remains exceptionally strong. The implication is that the infrastructure cycle has lots of room to run, and we're not yet seeing a clear peak on where that goes. I think also the second big news is AI is reshaping the SaaS market, not killing it. I think the more extreme disruption thesis around SaaS is not playing out in the way the market expected. AI is changing the software economics. It is changing product design, user interaction, pricing and the role of the application itself, but SaaS demand remains incredibly resilient. So what appears to be happening instead is that the software platforms are becoming one of the primary distribution layers for enterprise AI. The companies that own the workflow, the data, the user relationships are embedding AI directly into those environments. And the winners in SaaS will be the platforms that use AI to deepen their control of the workflow, enterprise data not simply bolt-on and assistant to an existing product. So that kind of brings us to managed services. Technology spend is surging, but services growth is just not keeping pace. I think we're seeing this widening disconnect between the overall technology spending and traditional managed services growth. Combined market ACV is growing an extraordinary rate while managed services expanding only modestly. Enterprises are clearly spending. The issue is where that spend is going. And right now, much of that incremental investment is flowing into infrastructure platforms and software, all driven by rather than traditional outsourcing and managed service contracts. So this is in a creative challenge for the service providers. They're participating in the transformation, but they've yet to capture the same growth economics as of technology providers underneath them. So the implication is that the services firms really need to move beyond simply using AI to improve delivery or improve productivity. They really need to turn AI into differentiated offerings. New commercial models and measurable client outcomes. Then the fourth big thing is AI investments accelerating, but at the enterprise level, I think we're going to see a reckoning on the ROI. AI investments continue to accelerate, but enterprises are becoming more selective and more demanding. The conversation has really moved from experimentation to economics. Boards are asking much harder questions around productivity, cost reduction, governance, adoption and measurable business value. We are seeing evidence that the sourcing behavior is becoming more concentrated as well and award volumes in ACV are flat. That really suggests enterprises are making fewer bigger bets and focusing capital initiatives, they believe can materially change the operating model. So the next phase of AI adoption is less about the number of pilots and more about which investments can scale deliver measurable returns and justify continued funding. We showcased the disconnect between value and investments in our latest state of AI report, which was just released and Alex is also going to discuss this later to really show how we're in a process reengineering phase to help drive that. So let's go ahead and take a broader look at the market. So first, we're going to talk about the combined market. And as a reminder, the combined market is the as-a-service market, which is cloud and SaaS and the managed services market. The combined market reached $52.4 billion of ACV in the third quarter, [indiscernible] up 63% year-over-year. That's the highest year-over-year growth rate ever for the combined market. And year-to-date, the combined market is up 45% versus 2025. As-a-service continues to drive most of that growth. The as-a-service market reached $41 billion this quarter, which was up 95% year-over-year and up 68% over 2025. Managed Services was up 2.1% in the quarter which is what we forecasted in the Q2 call. So the forecast is working, but you really see this large explosion on the hyperscaler side. As you can see, the technology services demand is rapidly expanding, but most of that expansion is happening in the as-a-service segment. And within it, it's primarily from the Infrastructure as a Service, which we'll talk about in more detail. So Alex, why don't you walk us through the details.

Thanks, Steve. [ ISG ] generated $8.4 billion of ACV in the third quarter, which is essentially flat year-over-year. Year-to-date, ITO, which includes network, EUC, ADM, data center and multi-tower deals, was down 3.5%, which marks the first year-to-date decline for this segment since 2019. As you can see here, the Americas was down over 9% year-to-date, primarily due to weakness in ADM and data center builds. Europe was a different story, where data center was up 11.4% likely driven by data and AI sovereignty concerns, but like the Americas saw weakness in ADM. ADM overall was down double digits through the first 9 months. There were 15% fewer ADM awards compared to 2025. Globally, infrastructure continues to be resilient and being up 1.4% year-to-date. Now let's dive a little deeper into some of the reasons behind the weakness in ADM. Last month, we released a study on enterprise adoption of AI. One result that we wanted to highlight here which is AI adoption activities generated the most value for enterprises over the last 12 months. The AI value comes from process reengineering and data analysis and workflow automation, not the stand-alone use of individual AI tools by individual people. And this poses a problem for the application sector, specifically because application development work has to be downstream of that process redesign. So there may be way less capacity constraint on software development with the improvements from AI, but the difficult work is still a blocking step before the development can continue. Our bookings data on infrastructure spending is up slightly. While AI may be the driver of infrastructure utilization, the demand we see to build data centers has a secondary effect driving up individual costs for hardware. So RAM, hard drives and GPU shortages are all driving up infrastructure costs for clients. And this is contrasting with their need to contend with AI demand and also to modernize their legacy hardware and make the required changes to address emerging cybersecurity threats, also largely from AI. This creates opportunities for service providers where automation and contract consolidations are driving savings of around 40% on average over their deal terms. Now let's take a look at our second service line, engineering services. Engineering ACV and contract activity were both flat year-over-year and year-to-date. But the big difference in this segment of the market right now is on a regional basis, EMEA continues to be very strong. ACV was up 21% year-to-date, supported by strong growth in the number of engineering awards. The Americas, however, is going in the opposite direction. ACV was down 17% and the award counts were down 7% year-to-date. Given the alignment of results here between EMEA engineering and the infrastructure results we saw, we expect that a lot of the physical engineering work is driven by the build-out of European data centers. In software engineering, there's a slowdown that is likely driven by the same forces weighing on ADM where process redesign is blocking progress. Let's look at some of our recent study results on engineering to dig deeper into why enterprises are struggling to modernize their engineering processes. As we said, the engineering services volumes are flat. And in our recent study here, we see that engineering modernization has many of the same limitations as ADM integration complexity and legacy technology are the largest impediments to adopting AI and making progress. Once again, before AI drives value, this time in engineering organizations need to rationalize their legacy footprint, and design their future state. This work may be aided by AI, but is not replaced by it. And over time, though, the net reduction of tech debt to enable AI will have long-term benefits. And in our view, we'll create more opportunities in the future as projects will not have this legacy tech debt factor that weighs on the price and time to deliver these projects. Now let's turn it over to Namratha to talk to us about BPO.

Thank you, Alex. This is the fourth consecutive quarter of double-digit year-on-year growth for BPO. It was up 12% year-on-year and nearly 35% year-to-date, and that's the fastest year-to-date growth since 2021. However, the sequential declines in the last 2 quarters in BPO is largely attributed to the sharp decline in HR after several strong quarters and also decline in multi-process BPO and F&A contributed to quarter-on-quarter decline in ACV. On the year-to-date results, industry-specific BPO continues to be very strong. It's up nearly 27%. Given the growth in GCCs and emerging focus on building or expanding data centers, facilities management also contributed to 77% growth year-to-date. We continue to see a decline in customer experience, and it's down nearly 20%. However, we are seeing strength in multi-process BPO. That's where more than 1 BPO offering is bundled with another. For example, ISTC is more back-office BPO being bundled, especially in areas like F&A procurement and supply chain. And that said, the BPO industry is witnessing an interesting shift as the buyers are increasingly moving away from labor arbitrage models looking for a holistic business process transformation, especially with AI, making inroads relatively faster in this space. Also, the scope of these business transformation deals are evolving and are becoming much broader and have started to include elements of IT that are integrated into them. Speaking of transformation and AI and BPO, as per our most recently released BPO study, 56% of the respondents said that their BPO contracts has significant AI-focused automation infused into their contracts. Processes and workload redesign have been happening since before RPA days. And most of these processes are now mature enough to adopt. Like I mentioned, enterprises are no longer looking for labor arbitrage. But instead, they are looking for partners who can enable business process transformation by combining people, process and technology. And technology focused focuses around analytics, cloud, automation, AI, coupled with deep domain expertise and industry vertical expertise. At ISG, we are seeing that shift across most of the BPO deals we advice. Changing nature of these deals is also paving way for a lot of IT providers to enter the space with technological strength, while BPO providers continue to reinvent themselves by investing and expanding their tech capabilities to address these evolving requirements. Now this is obviously changing the provider competitive landscape in the BPO space. With that, let's turn to our regional update. Kathy, over to you.

Thanks, Samantha. As Steve mentioned earlier, this quarter and honestly, the entire year has really been a story about strength in EMEA, combined with weakness in the Americas. The Americas was down nearly 6% in the 3 -- in the third quarter with declines in 3 of the past 4 quarters for the region. That said, the third quarter did see a slight bump, which might indicate recovery. CPG and retail and telecom were strong in the Americas in the third quarter, with the 2 biggest industry verticals, BFSI and manufacturing were both down double digits. The number of awards in the Americas was down 11% and that's the fourth consecutive quarter that award counts have trended negative in the Americas. As you can see here, year-to-date, the Americas is down 5.5%. However, Europe is a different story. ACV in the third quarter was up nearly 12%. That's 3 of the last 4 quarters, but the 2 biggest industry verticals, BFSI and manufacturing were both down double digits. The number of awards in the Americas was down 11%, and that's the fourth consecutive quarter that award counts have trended negative in the Americas. As you can see here, year-to-date, the Americas is down 5.5%. Europe however is a different story. ACV in the third quarter was up nearly 12%, and that's 3 of the last 4 quarters that have seen double-digit growth. in ACV. Manufacturing is very strong in Europe in the third quarter, up 74% year-on-year, which is the best result for this sector since 2020. U.K. and Ireland were strong again in the third quarter with over $1 billion of ACV awarded and DACH was also very strong, which has helped the some region turn positive year-to-date. While France was up 10% in the quarter, the smaller markets like Nordics, Southern Europe and Benelux were down at least 25% year-on-year. As you can also see, year-to-date, Europe is up 13% compared with 2025, and that's the best year-to-date growth rate since 2021. And finally, in Asia Pacific, ACV was up 16% year-on-year on a second quarter consecutive $1 billion quarter. The smaller markets of Southeast Asia and Korea supported this growth, while larger markets like ANZ and India were flat to down. Year-to-date, the Asia Pacific region is up 12% versus 2025. Now let's move on to our industry update. Look, this quarter, we're looking at BFSI, manufacturing and CPG retail. These 3 segments represent over 50% of the ACV in the market. If we start with BFSI, as you can see, ACV is down 20% year-on-year, and that's the lowest quarterly ACV since 2020. It's also the third quarter in the last 4 that has seen a year-on-year decline. As we discussed, the declines in the Americas. It accounted for the entire -- it accounted for the entire decline pullback for BFSI on a global basis. On a year-to-date basis, the BFSI ACV is just down over 5%. Industry insight shows banks remain under pressure from rising funding costs, deposit competition and uneven loan demand. And many institutions are focused on cost efficiency and operating leverage, while AI-enabled middle and back office transformation to reduce structural costs. However, this is not translating into any increased demand for sourcing is AI is often creating what we're calling a build insource before outsource mantra. If we turn now to CPG and retail, ACV in the quarter was up over 100%. Q3 was the best quarter ever for the combined segment. Award activity was very strong here with the number of awards in the quarter up over 60%. And on a year-to-date basis, CPG retail is up 56%. Now as a number Namratha pointed out earlier in her segment on BPO, we are seeing a market shift away from pure BPO with much of the demand tied to broader what we're calling enterprise business services, these engagements that frequently include legacy IT and technology modernization and the AI-driven solutions increasingly are combining business process technology, operations and legacy systems into an integrated transformation program. Think about examples like customer care and marketing operations where technology platforms and AI are reshaping where historically we had labor-based services. Now let's close out manufacturing, which represents 20% of the ACV in the market. And ACV in the quarter was up 13.5%, and it was the first $2 billion quarter since the second quarter of 2025. And However, contracting activity did slow slightly on the third quarter. And on a year-to-date basis, overall manufacturing is basically flat. Thinking about just manufacturing clients and what we see, they continue to really face macroeconomic and geopolitical supply chain risk. And these enterprise buyers require clear improvements in cost performance before signing new business. We also see manufacturing clients pursuing solution-specific initiatives and technical debt reduction, all while rebalancing a large set of providers. So in summary, across all of our industries, we are seeing a few key themes. We're seeing shaping up, but the headline is cost pressure, which is creating demand for transformation. And AI is really changing how buyers pursue it. Instead of broadly where arbitrage outsourcing, clients are consolidating their providers, bringing capabilities in-house, focusing on targeted transformations in combining their processes, technology and AI into larger service programs. This all supports why we see a change in the distribution of services while enterprise executives remain intensely focused on operating cost reduction. We're now going to shift over to our as-a-service segment and Stan, I'll turn it over to you for an update on software and Infrastructure as-a-service.

Thanks, Kathy. So as Steve mentioned in market level set, AI is reshaping the SaaS market, not killing it. And we've been writing about the fact that the SaaS apocalypse fears were overstated for a few quarters now, and I think you'll see that result in the results here today. So as you can see, SaaS ACV is up 21% year-over-year. That's a 560 basis point increase from last quarter, and it's the fastest growth this segment has seen since 2021. And on a year-to-date basis, SaaS ACV is up 12% compared to 2025. And as you can see on the right-hand side of the chart, just about every category we track here in SaaS is up year-to-date as well. A big reason for that is, as Steve mentioned earlier, software platforms are becoming 1 of the primary ways that enterprises are adopting AI. So for example, in IT service management, which you can see is up 7% year-to-date, AI is really moving this category from triage to resolution. So we're seeing platforms with agents that are investigating and remediating problems, not just reporting on them. CRM, which is up 15% year-to-date is a good example of how AI is aiming enterprise traction through new AI architectures. So with Salesforce, for example, you can now set up an MCP server that can access sales force workflows and data governance, that kind of stuff, but through other applications like Slack, Teams and Claude, for example, in analytics and BI, which is up almost 36% year-to-date. We're seeing lots of new entrants here. For example, Golden analytics is new kind of coming into this segment. And they, for example, and other smaller providers like them are using AI to compete with more established players. And finally, in HCM, which is up almost 10% year-to-date, agentic AI has really become kind of a core expectation for HR buyers with new Agentec capabilities from companies like Workday, Darwin Box and Paylocity, for example, and on the ground with our clients, we're seeing an increase in investment in the frontline workforce. So for example, areas like voice-enabled self-service, high-volume recruiting functionality and time tracking. So all of that said, while we very much believe that the SaaS disruption from AI has been overstated 1 area we are keeping an eye on is the fact that growth is slowing for the top 10 SaaS vendors. So you can see here on the far right hand of the side of the chart. Year-to-date, they are up 6%. But this time last year, the top 10 SaaS vendors were up 13%. Okay. Let's take a look at Infrastructure as a Service. Again, as Steve mentioned, the AI infrastructure boom is very much accelerating Infrastructure-as-a-Service ACV is up 115% year-over-year. That's the best ever quarterly result for IAS by a margin of almost $9 billion. On a year-to-date basis, IAS ACV is up over 85%, and the big 4 hyperscalers, so those are AWS, Microsoft, Google and Oracle were up 98% and this explosive growth is not just isolated to the Americas, all the regions we cover here are surging, as you can see. So given these really unprecedented results, we're going to use our AI index to dive into these results. So Steve, over to you to talk about what we really see driving this growth.

Yes. Thanks a lot, Stanton. And as a reminder to everybody, we started the AI index 2 quarters ago. really has 2 segments. The first segment kind of looks at the quantitative aspects. The second segment looks at the qualitative aspects of it. So if I look at the index this quarter, there are sort of 3 things that jump out. Infrastructure 1 isn't just growing. It's accelerating again. As we said at the beginning, and as Stanton just talked, we're really just seeing tremendous growth through that market. SaaS is really looking a lot healthier than it did a couple of quarters ago and some of the bigger problems on the software platforms are starting to show real momentum as we go forward. And managed services recovering a bit in the market but the underlying economics are still tough. So let's walk through each 1 of this. I think the story has changed. On AI index, we're now showing up 254% since the AI inception point, but what really stands out is that rate of change. The index is up 60% year-over-year and 24% quarter-over-quarter and our broader ISACV data is growing more than 100% as Stanton just talked through. So the hyperscaler growth is really accelerating. And you look underneath the hyperscaler results, I think there's 3 things driving it. The first is obviously different tier model companies. The scale of compute being committed by companies like OpenAI and Anthropic is extraordinary. And we're now talking about multiyear, multi-gigawatt commitments, custom silicon chips and very large capacity reservations. There's certainly some circular financing going on in that market. The cloud providers are investing in model companies. model companies are committing significant portions of that capital back to the cloud infrastructure and those commitments are showing up in the backlog and future demand. But I think there's more to the numbers. I think the second driver is really enterprise adoption. So when we pull back the numbers, Microsoft is a great example. Nearly 90% of its cloud revenue is coming from customers outside the frontier model companies. And this quarter, all of the sequential growth in commercial backlog came from non-frontier companies. Google Cloud grew 82%. And importantly, Google says the majority of this $514 billion backlog is typical GCP contracts from a broad mix of customers. Existing customers are also consuming more than 50% above their contracted commitments and AWS grew almost 37%, its fifth consecutive quarter of acceleration and its fastest growth in 18 quarters. I think the third driver is ultimately is the most important. I think we all understand that AI is pulling traditional cloud consumption with it. And when companies move AI workloads into production, they don't just consume GPUs. They consume CPUs, storage, databases, network security, all of the other services around those workloads. And AWS talked explicitly about this relationships as customers increase AI spending, AWS is seeing corresponding increases in core cloud consumption. We're also seeing enterprises accelerate cloud migrations because they increasingly want their applications, their data close in to their AI workloads. So AI is effectively creating a second catalyst for the cloud migration cycle. So I'd be careful about calling this simply an AI infrastructure boom. We're actually seeing AI demand enterprise cloud migration and core cloud consumption beginning to compound on 1 another. That's why the infrastructure growth curve has steepened rather than flatten. Now there's some risk in the system. The scale of investments in some of the circular financing around the Frontier model providers means we should not assume every dollar of backlog represents sustainable end user demand. But equally, the enterprise data tells this is not simply financial engineering. So from my standpoint, the key takeaway is the infrastructure cycle has not peaked. The more interesting question now is how much of the current acceleration is from your model capability build-out and how much represents the beginning of a much broader enterprise AI consumption cycle. On the SaaS side, we've seen a nice recovery. The SaaS sector is up 78% from its inflection point. Reminder that was December 2022. SaaS is also getting more interesting with the SaaS AI index up around 17% year-over-year and up 7% quarter-over-quarter, and the forward indicators really remain healthy. More importantly, we're starting to see some of the larger platforms put up very strong quarters. For example, ServiceNow continues to show strong subscription growth. Salesforce is showing better momentum in AIG and data. Snowflake continues to post strong product growth. Adobe is seeing growth in AI-related ARR and that the more aggressive and Palantir continues to show what happens when AI moves into real production workloads. So a few quarters ago, the question was Sasol whether AI was going to disintermediate large parts of the -- and I think today, it's which software platforms are going to use AI to strengthen their position. The companies that own the workflow, the data and the enterprise context still have a major advantage. So think of this as a system of records. So I would say SaaS is moving from resilience into recovery. And in parts of the market, we may already be seeing re-acceleration. Finally, on the managed services side, it's just a bit more complicated. The overall index is down 4.3% since the inception date. The stock performance in our basket improved materially quarter-over-quarter. So investor sentiment has clearly recovered from where it was. Revenue growth, though, remains modest. Profitability is still under pressure revenue per employee is improving, which tells us some of that productivity is moving in the right direction. And that's really where the tension is. If the revenue per employee is improving, but the top line growth and margins are still relatively weak, then part of the productivity benefit is probably being competed away, and we're seeing that in the pricing. Clients expect lower prices, providers are taking labor out. At the same time, providers are spending heavily on AI platforms, tools and up-skilling. So you get a productivity improvement without necessarily getting the same lift in revenue and margin. I think that's probably one of the more important things to watch in services. I think as we see it, AI can absolutely improve productivity. I think the question for the service providers is whether it's keep enough of the value to improve the economics of the business. So now let's take a look at the confidence analysis. So this quarter, we took a different look at AI sentiment. Rather than focusing only on the providers and the market commentary, we look specifically at enterprise sentiment and how the discussion around AI is changing over time. doesn't even need to be said, AI is a permanent boardroom topic. [ Medicines ] are still running at more than 6x the level we saw before the launch of ChatGPT and that matters because it tells that this is no longer temporary innovation cycle. AI has become embedded in strategic planning, investment decisions and operating model discussions. The more interesting story is what happened to confidence. Confidence declined through 2025 and into the first quarter of 2026, even as the volume of AI discussions continue to rise. Enterprises were not losing interest in AI, they were becoming more skeptical about the near-term economics. The conversation really shifted from excitement around what AI could do to a much harder question around what was actually working? How much was it costing and where the measurable value is showing up. What we see now is that confidence has begun to recover. The rebound is important because it suggests enterprises are starting to move beyond the first wave of experimentation and into more grounded phase of adoption. In 2025, again, the conversation was dominated by execution and commercialization with a relatively small share focus on risk and governance. In 2026, the share of discussion around cost, risk and governance has increased materially while execution remains the largest category. I would characterize the current market is moving from AI enthusiasm to AI accountability. That does not mean the opportunity is getting smaller. It means the bar for investment is getting higher. And that's exactly what we should expect as we move from experimentation into the core operating model. Now I do have a gift for everybody today. And so before I hand it off to Namratha, I want us to take us through the leaderboard. We did publish the annual State of Enterprise A adopter report last week. This is our annual report. Great job by the team really identifying it. So I'll give you just a minute scan on the QR code, you'll automatically get a free downloaded coffee. This year, the focus was really on value, value enterprises are, maybe even are not getting as they're realizing that from AI. And as I mentioned in the market level set the ROI reckoning has begun for enterprises, and this report goes into that in detail. So hopefully, we're able to snap that and enjoy the report. So Namratha, over to you now for the leaderboard.

Thank you, Steve. As a reminder, providers are listed in alphabetical order, and positioning is based on annual contract value signed over the past 12 months. The company is new to the list are denoted with an asterisk and a reminder that the regional leaderboards can be accessed on the ISG website. In the largest group, we saw very few changes to the leader board. With the top company is maintaining their strong positions. This quarter, we'll highlight Accenture, TCS, HCL Tech and some of their high-profile contract signings. During the quarter, British American Tobacco agreed to transfer 3,500 roles to Accenture with its fit-to-win AI transformation program, targeting GBP 600 million in annual cost savings by 2028. The deal impacts operations across multiple countries globally. In a similar move, we saw HCL Tech acquired Garden India and sign a 7-year strategic agreement with Guardian Life Insurance Company of America, absorbing 2,000 employees into a new dedicated business unit. The transaction creates a HCL Tech strategic business unit focused exclusively on Guardian's insurance, retirement, wealth management and employee benefits operations. Similarly, in another deal, we also saw TCS to cover U.S. retailer Best Buys India GCC, which includes taking over 450 employees working across engineering, data and AI capabilities to redesign workflows, develop EIB solutions and transition into an AI capability center over time. In the Building 15, we saw several new providers joined the leaderboard. Principal among those was Bechtle and Indra Sistemas. Bechtle is a German-based IT provider that combines IT strategy, consulting and managed services with direct e-commerce sales of hardware and software solutions. Indra Sistemas is a Spanish technology and consulting firm, specializing in aerospace, air traffic management, digital transformation services. We also highlight Genpact who won a transaction at Mondelez International to build an enterprise-wide identic operating model across their source-to-pay processes, spanning procurement through accounts payable. In the breakthrough 15 group, we observed a slight turnover this quarter. Due to the leaderboard is customer engagement provider of first-source solutions as well as Danish IT services provider, Net company. Net companies signed a 4-year deal with the Norwegian transport group, why for product development and digital solutions taking over several tasks previously managed by another supplier. And finally, in the booming 15, we did not observe as much turnover as we normally are accustomed to. We did see Birla Soft rejoin the leader board. Based in India, Birla Soft is an IT services consulting company that provides cloud computing, AI, data analytics and digital transformation solutions. Stanton, over to you for an update on our as a service leader board.

Thanks, Samantha. So in the as-a-service segment, a quarter, we're going to start with Microsoft. So it signed an agreement with 3M to deploy Microsoft's AI capabilities in key areas of 3M's transformation road map, including customer service, finance, sales and marketing. In the building 15, we're highlighting Workday and their deal with BMO and -- the bank scaled Workday self-service agent from a 500-employee pilot to all 55,000 employees. And in the breakthrough 15 business software provider Visma, won a deal at U.K. Bank NatWest and also renewed their deal with Brazilian Banking Group Bradesco. And in Booming 15, insurance software firm Guidewire announced that Belgian Insurance Group Atea had gone live with their software. The implementation was done by Deloitte. [indiscernible] IT subsidiary NRB and Guidewire Professional Services. And finally, Westpine Insurance Company announced the successful deployment of its claims operations on Guidewire Cloud deployed ClaimCenter to establish an AI-driven foundation for its regional business. Congratulations to all the providers featured on the ISG leaderboard this quarter. And Steve, I'll turn it back over to you one more time to close this out with a forecast.

Great. Well, thanks a lot Stanton and Namratha. And again, congratulations to everybody is on the leaderboard. That is not an easy accomplishment. So as we wrap up, let me leave you with a few observations from the quarters. On Managed Services, we had another record quarter for ACV, but the underlying growth remains fairly modest. The market is still healthy, but it's uneven. We're seeing stronger performance in EMEA and Asia, while the Americas remain softer. And at the same time, AI-driven productivity is starting to have a more visible effect on the economics of the services model. Providers are improving revenue per employee, but they're also dealing with price compression, lower labor intensity and continued investment in AI capabilities. So for that reason, sort of the overall market health, we're going to keep our 2026 managed services forecast at 2.1%, so essentially unchanged from the year. SaaS is a really different story. Infrastructure growth continues to accelerate its AI infrastructure demand and traditional cloud demand reinforce each other. We're just seeing extraordinary levels of hyperscaler investment and demand continues to run well ahead of what we would consider a normal cloud cycle. SaaS is also improving and the recovery is becoming more evidence across several of the larger software platforms. We've better revenue growth, stronger backlogs and increasing evidence that AI is helping strengthen rather than disrupt the software model because those 2 markets are now moving at very different speeds and because hyperscaler volatility has become so significant, we're really going to change how we present the forecast this quarter. Rather than relying on as-a-service number, we're going to break it out into its 2 components. So for SaaS, we're going to keep our 2026 forecast at 12.5%, unchanged from Q2, still really healthy growth. For IaaS, we're raising the forecast materially to 80%. That takes the overall as-a-service forecast to 60%, which is a significant increase from I'd say the one caution on the IaaS market, it's become really difficult to forecast the hyperscaler market. We're really confident to managed services and the SaaS forecast, but the scale of the hyperscaler investments, capacity commitments and circular financing across that ecosystem, it's just creating more volatility and more distortion in the traditional demand signals. So while we're moving as a service forecast up substantially, I'd say the forecast risk is still to the upside. We probably can't forecast high enough right now on what we're seeing with the cloud providers. So more broadly, the market remains really strong. The growth is increasingly concentrated as we spoke about infrastructure, still the clear growth engine. SaaS is improving. Managed services remains resilient. So I think the key takeaway for Q3 is that technology demand remains extremely healthy. AI is changing the shape of the market and the pace of growth across segments and the economics underneath that are all improving. So that brings us to the end of the formal call. We'll now open it up for Q&A. Please type your questions to the comments at the bottom of the screen. And Sumeet, can I invite you to start us off on the Q&A.

Yes. Hi, team. Thanks a lot for very vital presentation. My first question is around the demand for the managed services space. I mean we have seen throughout the year, the macro has been deteriorating the geopolitical issues as well as elevated level of crude prices and inefficient and the 2.1% managed services forecast that I would suggest has been [indiscernible] since the beginning of this year. I just want to understand how much of that is because of adverse macro? And how much of that is actually because of the AI-led inflation and compression to the overall have [indiscernible].

Steve, do you want me to make that one?

Yes. Yes, go ahead, why don't you kick us off?

Yes. I think, Sumeet, you've hit a really key point on the index this quarter. There's clearly some macro things on slowdown in certain sectors of work. coupled with this big investment that organizations are making in AI. So when I look broader across the managed services business, you get a bit of both. Financial services as an example, it's typically 25% to 30% of the market, and we've seen pullback there. And that pullback isn't just because of AI. It's because of some macro things where those deals haven't flown through and that clearly impacts it. We did some analysis, and we said, "Hey, if BFSI was just at its normal rate, not even where it is right now, the ACV would have been up 7%." So we would have seen really good growth across those others. But that's not how we look at the market. We look at it much more broadly. So when you have whether it's macroeconomic, geopolitical, different segments going on right now, you definitely get those headwinds in the managed services market. You couple that with the AIP, tremendous amount of investment from all of the major service providers on we call it harnesses but essentially their platforms to deliver AI services. With those investments, Obviously, there's going to be margin compression to do that. Labor is a little bit higher. So you got some subcontractor things coming into there. And the revenue just isn't growing because IT budgets aren't growing. So they're moving stuff away from whether it's managed services or other categories of IT into AI and cybersecurity, quite frankly. And as Alex can probably talk about hardware as well. So you're seeing this compression from multiple angles.

Yes, Sumeet, I think it's -- we kind of talked about this on the precall is just right now, we're in that phase where it's sort of this mix and you have the deflationary pressure that we talked about around managed services pricing. We're talking about a 40% reduction over the turn of the deal, which is typically going to be a 3-year deal with typically 3, 11 deals that we would call that. At the same time, you have very strong demand for cost optimization. Service providers are really, really good at that, but that optimization is increasingly looking like transformation in order to get that 40%. You're not going to get that through labor arbitrage. They've already gotten that through that onetime benefit. So that's 1 of the reasons we think contracts are getting longer. There's more transformation happening. But I think Alex said it really well when he was kind of talking about a lot of these prerequisites that has to happen before we start to see the real benefit of AI, especially in areas like apps and engineering, it's a huge amount of like process integration, operating model work that's going to need to take place in order to get this benefit around AI. So that would be kind of the volumes that we talked about going up. There's an unlimited supply of legacy tech that needs to be modernized. Ultimately, the productivity of the developer is only 1 part of that. So I just think we're in this phase and I think we're going to be here for a while. It's the pressure around pricing, but then the opportunity of greater volumes of what AI is going to be able to deliver?

Right, right. No, that's very helpful. So maybe my next question is around more of gaining the crystal ball next year. I know you guys give the outlook for next calendar year typically in January. But let's say, right now, we are in almost middle of October with midterm elections in November. So if one has to ask you the managed services outlook next year in terms of direction. Would it be higher than calendar '26 or would it be lower -- understanding or weighing the AI deflation as well as the macro factors, where would you set the direction higher than '26 or lower than '26 in 2027.

Steve, do you want to take that?