Cuộc họp công bố kết quả kinh doanh Q2 FY2027 của DXC Technology: Hạ dự báo khi giá trị đơn hàng suy yếu

Tóm tắt điểm chính

• DXC Technology báo cáo doanh thu quý 2 năm tài chính 2027 đạt 3,0 tỷ USD, giảm 4% so với cùng kỳ năm ngoái. Biên lợi nhuận EBIT điều chỉnh giảm 180 điểm cơ bản xuống 6,2%, trong khi EPS pha loãng phi GAAP giảm xuống 0,67 USD từ mức 0,84 USD.
• Tỷ lệ đơn hàng trên doanh thu (book-to-bill) trong quý đạt 0,8 và giá trị đơn hàng mới (bookings) giảm 10% so với cùng kỳ năm ngoái. Ban lãnh đạo nêu nguyên nhân là do các hợp đồng lớn bị trì hoãn và sự suy giảm chuyển đổi của các dịch vụ theo dự án, đặc biệt tại khu vực Châu Mỹ và UKI.
• DXC đã hạ triển vọng năm tài chính 2027 xuống mức doanh thu tự thân giảm 5%–5,5%, biên lợi nhuận EBIT điều chỉnh đạt 5,75%–6,25%, EPS pha loãng phi GAAP đạt 2,25–2,50 USD và dòng tiền tự do đạt khoảng 600 triệu USD.
• Thương mại hóa AI tiến triển tích cực: hơn 5.700 nhân viên đã đạt được các chứng chỉ AI kể từ tháng 6, mức độ sử dụng Oasis tăng gấp 6 lần so với quý trước, và DXC đã giành được hợp đồng dịch vụ hiện đại hóa (modernization-as-a-service) lớn đầu tiên cho mảng Growth X.
• Mảng kinh doanh bảo hiểm tăng 23%, chủ yếu nhờ sửa đổi hợp đồng một lần. Doanh thu SaaS tăng trưởng ở mức hai chữ số mạnh mẽ, trong khi danh mục dự án tiềm năng (pipeline) của Smart Apps mở rộng hơn 20% so với quý trước.
• CEO Raul Fernandez xác nhận rằng DXC đang trong quá trình thảo luận về khả năng bán mảng kinh doanh bảo hiểm, nhưng không đưa ra thêm chi tiết.

Dữ liệu tài chính chính

Hiệu quả hoạt động kinh doanh và vận hành

Doanh thu CES giảm 3,3% so với cùng kỳ năm ngoái. Dịch vụ ứng dụng doanh nghiệp (Enterprise Application Services) và Growth X tiếp tục tăng trưởng, nhưng các ứng dụng tùy chỉnh vẫn gặp áp lực. Tỷ lệ book-to-bill của CES đạt 0,79, với giá trị đơn hàng mới giảm 17%, chủ yếu do mảng Dịch vụ ứng dụng.

Doanh thu GIS giảm 10%, thấp hơn một chút so với kỳ vọng của ban lãnh đạo. Tỷ lệ book-to-bill đạt 0,8 và giá trị đơn hàng mới giảm 13%. Ban lãnh đạo lý giải sự yếu kém này là do các hợp đồng lớn bị trì hoãn và các dịch vụ theo dự án giảm sút. Nhiều giao dịch lớn hiện dự kiến sẽ hoàn tất trong nửa sau của năm tài chính 2027.

Doanh thu bảo hiểm tăng 23%, nhờ sửa đổi hợp đồng một lần. Nếu không tính khoản này, doanh thu cốt lõi ổn định so với quý trước do sự tăng trưởng của phần mềm định kỳ đã bù đắp cho việc thu hẹp hợp đồng dịch vụ quy trình kinh doanh. Mảng kinh doanh SaaS ghi nhận mức tăng trưởng mạnh mẽ ở mức hai chữ số, được hỗ trợ bởi Asure, Horizon Solutions và các ứng dụng Smart Apps dựa trên AI.

Số lượng khách hàng bảo hiểm doanh nghiệp áp dụng Asure đã vượt quá 275. DXC đã bán các ứng dụng Smart Apps đầu tiên trong quý và danh mục dự án tiềm năng liên quan đã tăng hơn 20% so với quý 1.

DXC cũng báo cáo sức hút thương mại gia tăng trên toàn bộ danh mục AI đại lý (agentic AI). Mức độ sử dụng Oasis đã tăng gấp 6 lần so với quý trước, nhờ có thêm các quy trình sản xuất thực tế và các năng lực đại lý rộng hơn. Công ty đang triển khai hai chương trình thử nghiệm với khách hàng mới, trong đó quản lý bản vá (patch management) nổi lên như một trường hợp sử dụng hàng đầu.

Growth X đã giành được hợp đồng dịch vụ hiện đại hóa lớn đầu tiên với một tổ chức ngân hàng toàn cầu. Ban lãnh đạo cho biết công việc ban đầu đang tiến triển vượt tiến độ và có thể dẫn đến 7 hoặc 8 dự án tiếp theo.

Dự báo của ban lãnh đạo

Ban lãnh đạo kỳ vọng doanh thu GIS sẽ cải thiện nhẹ so với quý trước trong quý 3 và cải thiện rõ rệt hơn trong quý 4. Triển vọng này giả định tỷ lệ chuyển đổi dự án tiềm năng vẫn duy trì ở mức thấp của quý 2 trên cả hai mảng CES và GIS.

DXC cho biết khoảng 2/3 mức giảm dự báo doanh thu phản ánh lượng đơn hàng mới dự kiến trong quý 2 đã không diễn ra. 1/3 còn lại phản ánh giả định về sự chuyển đổi yếu hơn trong nửa cuối năm. Mức giảm này được chia tương đối đều giữa CES và GIS.

Rủi ro và các mảng cần theo dõi

Rủi ro ngắn hạn chính là hiệu quả bán hàng yếu hơn. Các hợp đồng lớn thuộc khu vực công tại UKI vẫn đang trong quá trình thương lượng, nhưng thời điểm ký kết chưa chắc chắn. Tại Châu Mỹ, một số dự án nhỏ hơn bị trì hoãn trong khi các dự án khác đã rút khỏi danh mục dự án tiềm năng, và ban lãnh đạo thừa nhận tỷ lệ thắng thầu thấp hơn cùng sự cạnh tranh gay gắt hơn.

Khách hàng cũng đang đánh giá lại các dự án công nghệ truyền thống so với các giải pháp thay thế dựa trên AI đại lý. Ban lãnh đạo cho biết điều này đang kéo dài chu kỳ ra quyết định và trong một số trường hợp, khiến khách hàng cân nhắc các hợp đồng 2 năm thay vì cam kết 5 năm.

Giá bán được mô tả là tương đối ổn định đến giảm nhẹ. DXC không nhận thấy rủi ro gia tăng đối với lượng đơn hàng tồn đọng đã ký, đồng thời cho biết mức độ suy giảm backlog vẫn nhất quán với các mức trong lịch sử.

Doanh thu thấp hơn đang gia tăng áp lực lên chi phí và biên lợi nhuận. CFO Robert Del Bene cho biết DXC đã mở rộng mục tiêu cắt giảm chi phí cho năm tài chính 2027 và đang đánh giá các biện pháp mạnh mẽ hơn cho năm tài chính 2028, đồng thời nỗ lực duy trì đầu tư vào các sáng kiến tăng trưởng.

Điểm nổi bật trong phần Q&A với chuyên gia phân tích

• Chuyển đổi danh mục dự án tiềm năng: Ban lãnh đạo đã phân biệt giữa các hợp đồng lớn bị trì hoãn, đặc biệt là trong mảng dịch vụ khu vực công ở UKI, và việc chuyển đổi dự án nhỏ hơn yếu hơn ở Châu Mỹ. Triển vọng năm tài chính 2027 không giả định sự phục hồi của tỷ lệ chuyển đổi.
• Hiệu quả bán hàng: DXC thừa nhận rằng một số yếu kém trong chuyển đổi là do nguyên nhân nội bộ. Công ty đang đơn giản hóa việc tương tác với khách hàng tập trung vào một đầu mối chịu trách nhiệm duy nhất và giảm bớt các vướng mắc nội bộ trong quy trình tiếp cận thị trường.
• Chi tiêu cho AI và dịch vụ CNTT: Ban lãnh đạo kỳ vọng việc triển khai AI tại doanh nghiệp vẫn đủ phức tạp để cần đến các nhà cung cấp dịch vụ bên ngoài. Tuy nhiên, sự chuyển dịch này đang tạo ra một giai đoạn chuyển tiếp trong đó khách hàng đánh giá lại các dự án và quy trình làm việc hiện có.
• Chất lượng khối lượng công việc tồn đọng (backlog): DXC cho biết khối lượng công việc tồn đọng đã ký không có dấu hiệu suy giảm gia tăng, với các xu hướng hiện tại nhất quán với kinh nghiệm trong lịch sử.
• Cơ hội an ninh mạng: Ban lãnh đạo nhấn mạnh Frontier Code Assurance là một giải pháp ưu tiên được thiết kế để xác định, xác minh, xếp hạng và hỗ trợ khắc phục các lỗ hổng phần mềm bằng AI cùng các năng lực an ninh mạng của DXC.
• Danh mục bảo hiểm: Ban lãnh đạo cho biết mảng kinh doanh bảo hiểm có đối tượng khách hàng gắn kết cao, doanh thu định kỳ và các rào cản pháp lý, đồng thời từ chối cung cấp thêm thông tin chi tiết về các cuộc thảo luận chiến lược tiềm năng.

Toàn văn biên bản cuộc họp báo cáo kết quả kinh doanh

Toàn văn cuộc gọi công bố kết quả kinh doanh

Phần trình bày của ban lãnh đạo

Operator

Hello, everyone. Thank you for joining us, and welcome to the DXC Technology Second Quarter Fiscal 2027 Earnings Call. [Operator Instructions]

I will now hand the conference over to Tiffany Horvath, Senior Director, Investor Relations. Tiffany, please go ahead.

Tiffany Horvath

Thank you, operator. Good afternoon, everyone, and welcome to DXC Technologies Second Quarter Fiscal 2027 Earnings Conference Call. We hope you had an opportunity to review our earnings release, which is available in the IR section of DXC's website. Speaking on today's call are Raul Fernandez, our CEO; and Rob Del Bene, our Chief Financial Officer. Here's today's agenda. First, Raul will update you on our strategic initiatives. Rob will then review our quarterly financial performance as well as provide thoughts on our third quarter and full fiscal year 2027 guidance. Raul and Rob will then take your questions.

Please note, certain comments made during today's call are forward-looking and subject to risks and uncertainties that could cause actual results to differ materially. Details of these risks and uncertainties are in our annual report on Form 10-K and other SEC filings. We undertake no obligation to update any forward-looking statements. Unless otherwise noted, year-to-year or quarter-over-quarter revenue growth rates discussed on today's call refer to organic revenue growth on a non-GAAP basis, which excludes the impact of foreign exchange and inorganic activity. We will also be discussing certain other non-GAAP financial measures that we believe provide useful information to investors. Reconciliations to the most comparable GAAP measures are included in today's earnings release.

And with that, let me turn the call over to Raul.

Raul Fernandez

Thank you, Tiffany. I would like to take a moment to comment on the market speculation regarding a possible sale of our insurance business. While I don't want to get into specifics, I can confirm that we are currently in ongoing discussions.

As we have said previously, Management and the Board are committed to exploring all avenues to create shareholder value. We believe we have multiple paths to do so, including improving the core, scaling our fast-track Agentic and AI-based solutions and evaluating strategic actions across our portfolio. We do not intend to answer additional questions at this time.

On the quarter, our revenue, adjusted EBIT and non-GAAP diluted EPS were above our guidance. For the balance of the year, we are adjusting our full year guide to reflect sales performance in our core business that fell short of our expectations. At the same time, we are pleased with the traction in our Agentic solutions.

One of DXC's greatest competitive advantages is the combination of our technical talent, deep industry expertise and access to the world's leading AI platforms. Since June, more than 5,700 colleagues globally have earned AI certifications. We are building a global bench of engineers capable of moving from AI experimentation to production in complex enterprise environments. That is a capability few organizations can replicate at our scale. The next step is commercialization, deploying these engineers against customer opportunities and converting that capability into revenue.

We are also developing greater visibility into average AI deal size, close cycle timing and pace of follow-on work, giving us increased confidence in the revenue opportunity ahead. Creating a growth company out of DXC is an ongoing exercise that will be bolstered by our work in strengthening the CORTRAK and growing our Agentic Fast Track and AI-based solutions. We feel that the addition of Paul Taylor and his operational accountability and execution focus as our President gives us a path to achieving that goal.

A few of Paul's key focus areas are removing friction in how we operate globally and improving our sales execution. He will join Rob and I on our next call to expand on his initiatives. We have continued to invest in agentic and AI-based solutions to drive innovation and position DXC for future growth. The value of AI will go to the companies that control the workflows, data, rights and distribution.

We believe DXC is well positioned to capture the expanding total addressable market from AI because of our deep global technical bench, existing customer base and new Agentic solutions. We see evidence that what we are building is resonating with clients.

In Growth X, we secured our first large win with a leading global banking institution to modernize its mission-critical core banking system and accelerate its transition to the cloud. Delivered through our modernization as a Service offering, we are combining our AI-powered modernization platform, industrialized delivery factory and Anthropic's advanced quad model to discover, document and validate thousands of business rules embedded in decades of legacy systems.

By integrating Frontier AI capabilities with our deep banking expertise and proven modernization methodologies, we are transforming how complex, high-risk modernization programs are delivered at scale. The promise of modernization has been around for decades, but the ability to do it at scale quickly is here.

We are very well positioned to win in this business category. This win demonstrates our ability to turn advanced AI technologies into differentiated repeatable offerings that accelerate client transformation and create new growth opportunities for DXC.

This quarter, we made progress increasing the value Oasis delivers to our existing clients. Our teams are expanding the capabilities of our agents to solve more complex customer pain points and increasing the number of agents that can make controlled changes within customer environments.

We are also increasing the number of agentic workflows in production. For example, we recently launched DXC Workplace Services, an AI-native workplace offering enabled by Oasis that reimagines the digital workplace around employee outcomes.

With workplace services, agents and DXC experts help resolve employees' IT issues before they disrupt work. As a result of this capability expansion and increased workflows in production, utilization surged 6x from last quarter. Demand for Oasis continues to be strong with increased adoption from existing clients and interest from new customers with patch management emerging as a leading use case.

We are currently running 2 pilots with new customers with incremental opportunities in the pipeline. In our insurance business, we continue to build momentum in our SaaS-based Asure platform with customer adoption exceeding 275 enterprise insurance accounts. This expanding platform installed base allows more customers to use our AI-enabled smart apps.

We sold our first Smart Apps in the quarter, and our pipeline has grown more than 20% from last quarter. Additionally, we are leveraging Smart Apps internally in our BPS business to drive efficiencies and provide proof points for our customers.

As we shared last quarter, DXC is accelerating the development of our AI engineering talent through strategic partnerships and hands-on training programs. In August, I attended Anthropic's Partner Base camp in San Francisco, alongside DXC engineers and forward deployed engineers from across the industry. Working and learning alongside dozens of DXC teammates and FTEs in training from other companies gave me a close and upfront appreciation of the operating complexities of being an AI builder. It also gave me insight into what human capital skills are needed to be an A player in an AI world. It demands a combination of engineering excellence, deep industry knowledge and the ability to translate AI capabilities into measurable business outcomes.

More than 500 of our engineers participated in a multi-platform FDE boot camp in India, working with Anthropic, 11 Labs and AWS. Our approach is multi-platform, building expertise across Amazon, Microsoft Copilot, 7.AI, Anthropic and 11 Labs to deliver solutions aligned with our clients' needs. Additional base camps are planned for Sydney and Tokyo.

Most importantly, DXC engineers are standing out in the marketplace for our operational and industry-specific context that is critical to creating value for our clients. We are already turning these capabilities into differentiated solutions for our customers. We have recently been approved to productize ClaudMethos into DXC Frontier Code assurance, a managed service that combines frontier AI reasoning with DXC's cybersecurity expertise and governed delivery to identify, validate, prioritize and help remediate vulnerabilities in critical software. A scan that produces every alert at once creates a backlog, not security.

DXC goes beyond identifying vulnerabilities by validating each finding, ranking it by risk to the client and fixing what matters first. That is a tangible example of DXC's differentiated commercial offerings for our customers in the cyber space.

Last quarter, I talked about customers wanting the potential and promise of AI, but also needing to trust the architecture underneath it. Since last quarter, trust and safety considerations have become even more important for customers. Given our work operating the most critical systems, applications and infrastructure, oftentimes in heavily regulated industries, trust and safety remains a key priority as we embed AI in more of our solutions.

Our architectures are being developed around safe, responsible and resilient AI operations. DXC's focus is on agent assurance, agent engineering and harness engineering, which provide the controls, safeguards and operating mechanisms needed to manage AI systems. DXC has also partnered with LangChain to support agent and harness development for Oasis. From the beginning, Oasis used LangChain to develop agents, and we have continued improving the harness as agents move to the production environment. We are licensing LangChain to support the development, evaluation and observability of Oasis agents.

We have more work to do, particularly in improving sales execution and stabilizing revenue in our core business. That remains a priority, and we're not satisfied with our current performance. But when I look at where DXC was at the beginning of this year and where we are today, the progress is meaningful. We have moved from strategy to products, from products to deployments and increasingly from deployments to commercial adoption. We have a stronger leadership team, a growing portfolio of agentic offerings, thousands of AI-certified engineers and customer relationships that give us a unique opportunity to scale.

Our strategy hasn't changed. What has changed is the evidence that it can work. The priority now is execution, improving the core, scaling what works and creating sustainable long-term value for our shareholders. Through ElevenLabs, my script will be available in 6 languages following this call.

Let me turn it over to Rob to review the second quarter results.

Robert Del Bene

Thank you, Raul, and good afternoon, everyone. Today, I'll go over our second quarter results, provide guidance for the third quarter and update our full fiscal year 2027 outlook.

Now starting with the second quarter results. Total revenue was $3 billion, down 4% year-to-year, above the high end of our guidance range. Our adjusted EBIT margin was 6.2%, down 180 basis points year-to-year. Non-GAAP EPS was $0.67, down from $0.84 in the second quarter of last year, largely driven by lower adjusted EBIT.

In the second quarter, we had a onetime contract modification in our insurance business, which was not anticipated in our guidance. Excluding the onetime modification, we were within the guidance range on revenue and in line with our adjusted EBIT margin and EPS guide.

The book-to-bill ratio for the quarter was 0.8, and our trailing 12-month book-to-bill ratio remained at 1.0. However, we came into the quarter with a robust pipeline and the bookings performance of minus 10% year-to-year was below our expectation. The performance was driven by large deal timing and project-based services pipeline conversion that declined versus recent history. This modestly impacted our revenue in the second quarter and affected our expectations for the second half of the year. The decline in performance was most pronounced in the Americas and UKI.

And now turning to our segment results. CES revenues declined 3.3% year-to-year, in line with our guide and consistent with recent performance. Enterprise Application Services and Growth X continue to grow with pressure continuing in our custom applications business. We're building momentum in engineering services, supported by an expanding pipeline and year-to-year bookings growth. The CES book-to-bill ratio for the quarter was 0.79, reflecting a 17% decline in bookings year-to-year, primarily driven by Application Services.

As I referenced earlier, we had a strong pipeline coming into the quarter and our bookings performance was below our expectation, particularly in the Americas. For GIS, the book-to-bill ratio was 0.8 with a year-to-year bookings reduction of 13%. The decline was driven by the timing of several large deals that we now expect to close in the second half of the year and by weakness in our project-based services performance. The short-term project dynamic impacted our revenue, which was down 10% and slightly below our expectation.

Our insurance business grew 23% year-to-year, well above our expectation, driven by the contract modification previously mentioned. Excluding the modification, the underlying insurance revenue was consistent quarter-to-quarter with growth in recurring software revenue, offset by the wind down of a business process services contract that we wrap in the fourth quarter.

We continue to build momentum in our SaaS business, which grew at a strong double-digit rate in the quarter. As a reminder, our SaaS business includes our Asure platform, Horizon Solutions and AI-based Smart apps, where we're starting to gain traction. We sold our first Smart Apps this quarter, and the pipeline grew more than 20% from the prior quarter. These sales will begin to modestly contribute to our SaaS growth in the second half of the year.

Now turning to cash flow and the balance sheet. We generated $91 million of free cash flow during the quarter. And as expected, we made a cash deposit of $122 million with the IRS related to our tax litigation case from currency losses from 2009. Excluding this payment, free cash flow totaled $213 million. For the first half of the year, free cash flow is $405 million, including our TCF settlement from the first quarter and the cash deposit made with the IRS in the second quarter.

Consistent with our capital allocation plans, we retired $400 million of our U.S. dollar bonds that matured in September 2026. Over the past several years, we've made it a priority to reduce our total debt, strengthening our balance sheet. During the quarter, we also repurchased $60 million of shares. Our capital allocation plans for the full year remain the same with flexibility on timing of the repurchases.

Now let me provide you with our updated full fiscal year 2027 guidance. Before getting into the specifics, I want to give a little color on our guidance. As discussed in recent quarters, our guidance anticipated better second half revenue performance supported by an improving GIS opening backlog. These backlog dynamics remain unchanged. However, there are 2 factors impacting our second half that are leading us to reduce our full year guide. The first being the impact of the weaker-than-expected second quarter bookings; and second, our assumption that the lower levels of pipeline conversion we experienced in 2Q will continue through the balance of the year in both CES and GIS.

Specifically, we now expect our full year organic revenue to decline 5% to 5.5% year-to-year. In CES, we now expect revenue to decline at a mid-single-digit range. In GIS, we continue to expect a stronger second half profile with modest sequential improvements in the third quarter and a more pronounced improvement in the fourth quarter. We are now expecting a high single-digit revenue decline for the year.

In Insurance, due to the impact of the 2Q contract modification, we are increasing our guide from low single-digit to mid-single-digit revenue growth. We continue to expect an improvement in the fourth quarter revenue growth rate driven by the momentum in our AI and cloud SaaS offerings and by the positive impact of the previously mentioned contract runoff, which wraps in the fourth quarter.

As a result of the reduction in our revenue guidance, we now expect adjusted EBIT margin for the full year to be between 5.75% and 6.25%. Our non-GAAP diluted EPS outlook is now expected to be between $2.25 to $2.50. We now expect full fiscal year 2027 free cash flow of approximately $600 million, reflecting the reduction in our adjusted EBIT guidance.

The second half free cash flow is skewed toward the fourth quarter, driven by the timing of capital software payments being made in the third quarter. For the third quarter of fiscal 2027, we expect total organic revenue to decline approximately 6% year-to-year. At the segment level, we expect CES to decline mid-single digits. GIS is anticipated to decline at a high single-digit rate and insurance is expected to grow at a similar rate as the first quarter. We expect adjusted EBIT margin to be approximately 6% and non-GAAP diluted EPS to be approximately $0.55.

And with that, let me turn the call back over to Tiffany.

Tiffany Horvath

Thank you, Rob. We'd like to now open the call for your questions. Operator, can you please provide the instructions?

Operator

[Operator Instructions] Your first question comes from the line of Tien-Tsin Huang from JPMorgan.

Phần hỏi đáp

Tien-Tsin Huang

On my first question, just thinking about the lower pipeline conversion. Can you give us a little bit more detail there? Is that client in decision? It sounds like the deals were pushed out, not lost. Is that accurate? Just curious how you're seeing the next couple of quarters play out here?

Robert Del Bene

Yes. So Tien-Tsin, it's Rob here. So let me take it in a couple of pieces. First, a little more color. It was pronounced in the Americas and UKI. In terms of the deal dynamics, the deal -- the bigger deals being deferred was weighted toward UKI, public sector, and those deals are still in the pipeline being negotiated, and we have confidence that they'll get booked. It's just the timing is uncertain. So that was a delay dynamic as opposed to deals dropping out.

From a project-based services perspective, the smaller deals, that was more heavily weighted toward the Americas, impacted us mostly in financial services and the retail sectors, and they impacted both GIS and CES. And those are a combination of deal slippage -- deals being delayed, slipping and then some deals being dropping out of the pipeline. It's a combination. And the competition in the Americas has stiffened. We could see it reflected in our win rates. So it's most pronounced there.

Now we -- that being said, we enter into the third quarter with a really good pipeline also in the Americas. So we are generating -- and these are qualified opportunities. These are not raw pipe. They're qualified. So we continue to generate pipe, and we've got a lot of good content that Raul has mentioned in his remarks. So we're optimistic going forward, but we're not counting on it in our guide. We're counting on conversion rates continuing at the levels in the second quarter.

Raul Fernandez

Yes. Just let me add a little bit of color on kind of these agentic deals. The small deals are smaller, but they happen -- they can happen more quickly, meaning they start and end and deliver value, improve value very quickly. But they are smaller than traditional small deals in rolling out SaaS and other types of products or pilots or projects. They also, though, lead to a clear line of sight on unlocking a tremendous amount of value. So while they're upfront smaller, they move more quickly and deliver the proof points needed to then engage into much larger transactions that can be replicated.

And I think one thing that we'll be able to talk about over the period of the next few quarters is the modernization as a service win where we are executing ahead of time and the client is super happy, and there's 7 or 8 projects right behind it. And you can see that, a, the value is being delivered, it's being recognized and the follow-on work is being unlocked.

So I think there's dynamics in kind of the traditional older projects. There are dynamics in terms of the new AI projects, size, cycle time, different set of issues. There's a tremendous amount of IP issues, a tremendous amount of questions, obviously, around safety and security, around the right evals and harnesses and agents to make sure that everything that is being built is built in a safe way where information that our customers own are maintained and contained in a safe environment. So the complexity is increasing. The time to value in terms of delivering an AI project is decreasing, but the unlock behind that is substantial.

Operator

Your next question comes from the line of Keith Bachman from BMO Capital Markets.

Keith Bachman

I had a couple, if I could. Just to follow on from the previous question. Rob, could you speak to the why? If these projects are being pushed, is it your sense because it's a crowding out of other sources, other grabs for fun, if you will, or competition? I'm just curious as to the why because the messages that we're getting from the various leaders in IT services, frankly, is a little bit different in terms of the backdrop of the market dynamics. So I'm just, number one, I wanted to ask about the why. And two, you highlighted that they're more competitive. And so should we infer from that comment that pricing -- price competition is likewise increasing, which is perhaps impacting revenue growth and the bookings as well?

Robert Del Bene

Yes. So Keith, what we're seeing in the data, and again, this is pronounced in the Americas, meaning the U.S., is that our -- the pricing data we have is relatively stable, slightly down. The win rates dropped a bit in the quarter. And I think it's related to the dynamic that Raul was mentioning. I mean, traditional project-based services are being delayed in that they're undergoing additional evaluations, and priorities are being reset by our customers, and I think that's happening everywhere. So that does delay the decision cycles on traditional workloads.

Raul Fernandez

I've talked directly to customers that have been in the decision cycle with us and without us. And I hear consistently that key managers, Board members, et cetera, that review and approve go forward or not on some of these initiatives are asking more like, okay, should we do this upgrade with this traditional software player? Or can we look at an agentic way of doing it? And that -- those sets of questions are adding time to the sales cycle.

The other aspect to this is if they do move forward, they're saying, okay, instead of a 5-year deal, we should probably only do a 2-year deal because we don't know where the world is going to be in 2 years. So those are dynamics that are happening, and I can tell you because I speak with a lot of customers as they're making decisions on our work and on a lot of other companies' work as well.

Keith Bachman

Let me just ask a follow-up to that then. As you're speaking to a lot of customers, budgets are currently being formed for 2027. It's probably candidly still early on the cycle. And I think there's a risk or a concern rather with investors that of leading organizations are trying to allocate incremental funds to AI and frankly, security as well. How do you think that positions IT services? In other words, do you envision risk to crowding out? Or do you think there's participation as you look at budget demands associated with calendar year '27?

Raul Fernandez

So a couple of things that I think I've learned over the last few months. One, that while popular belief is that any of us and any of us can do something prompting and delivering at some level, the complexities to actually deliver something that's enterprise level is tremendous. I mean, sitting 2 days in base camp with not just our engineers, but engineers from competitors, I walked out of there going, first of all, it's going to take longer, the transformation. Second, it is super complex. So people are still going to need outside providers. So we and all of us will have plenty of work, but we're going through a rough patch. We're going through unpaved roads right now with not a lot of traffic, but that will smooth out.

And then finally, I think there is a bifurcation of reimagining workflows versus executing on your current business. And as I've talked to customers, I think there's a realization that people that are operating existing workflows, existing systems are doing a fine job at it, but they may not be the best people to reimagine an agentic approach. That's the opportunity we have, and that's the opportunity that, frankly, is nascent today that will emerge over time.

Operator

Your next question comes from the line of Bryan Bergin from TD Cowen.

Bryan Bergin

Just based on what you've seen here in discretionary demand and the pipe conversion commentaries, do you foresee any risk that signed work that is already in backlog is incremental at risk? And what I'm curious about is certain items you may have signed that may be of a similar nature of where clients may be shifting their priorities in their pipeline. Is there any of that dynamic occurring where there could be elevated backlog erosion risk?

Robert Del Bene

Bryan, I don't -- in the data, I don't see that. Our backlog erosion statistics for this year are very consistent with historical levels, and they're and they've been actually moderating at a very good level. So we just do not see that happening.

Raul Fernandez

Yes. And I think there was a comment earlier about AI spending priorities. I think one area that we are super well positioned because of our existing cyber business and remediation business and now with the ability to run our frontier code assurance, which essentially gives us the ability to take an application, any application, run it with [indiscernible], detect vulnerabilities, document those and help the client prioritize and patch those quickly.

That is an area because the offensive side, meaning the bad guys have the skills and capabilities, the defensive side has to get up to AI speed in order to fight AI intrusion at AI speed or intrusion at AI speed. So while there are areas that may hit a longer decision cycle, there are areas where it is going to be required to move very quickly. And having that product offering just released in the market last week and making substantial progress in the limited amount of time we've been in market, I'm very, very optimistic that we've got the right product set available and the right offerings available for the demand that's there today.

Bryan Bergin

Okay. Okay. Understood. And then just as it relates to the 2027 guidance. So Rob, just really trying to unpack the moving parts. So you have the insurance modification this quarter that helped. But I'm not sure if I heard that right, there was a separate offset potentially later. So first that. And then if you can kind of rank order the most notable drivers in the reduction elsewhere across CES and GIS. And I think it sounds like you're still assuming GIS ramps in 4Q. If I heard that right, what gives you that confidence?

Robert Del Bene

Yes. You did hear that right. Well, first, on the onetime benefit from the insurance transaction, there was no real offset to that. So that was a onetime second quarter benefit carries through to the full year with no downside offset.

In terms of the dynamics of our guide change, just going back a little bit to -- we had a strong opening backlog coming into the year. That opening backlog projected an improvement in our year-to-year performance in the second half of the year, pronounced in GIS. We entered the second quarter with a strong pipeline. We assumed the recent history, historical conversion rates of that pipeline. And what occurred was what I mentioned in my prepared remarks, we had delays in larger deals in GIS. Again, they're not lost, they're delays, but the revenue impact of those delays carry into the second half of the year.

And then we had the drop, particularly in the Americas and the conversion rate of the pipeline. That impact -- that also impacts the second half of the year. So if you look at the dynamics, kind of unpacking them a little bit, half of the takedown is GIS, half of the takedown is CES in that range. They're very close. And the drop in second quarter performance, meaning the bookings we had anticipated that didn't come through are about 2/3 of our revenue decline in our guide. And the last 1/3 is carrying forward those lower conversion rates into the third and fourth quarter. Now all of that had very little impact to 2Q. The impacts the go-forward third and fourth quarters.

Operator

[Operator Instructions] Your next question comes from the line of Jonathan Lee from Guggenheim.

Yu Lee

Look, pricing is stable yet when rates are falling, while peers describe what is effectively a steadier backdrop. Can you help us understand if this may be a DXC-specific share loss issue and who is winning those deals that you lose? And what are you changing in sales execution to potentially recover here? What gives you confidence that, that dynamic can potentially inflect?

Raul Fernandez

Yes. We're simplifying our interaction with the customer so that we've got a single point of accountability between the company, DXC and the customer with the offerings backing up that single point of accountability. That friction before, we had multiple touch points. It wasn't coordinated. We could -- we were moving at different speeds for different deals and different offerings. The clarity and simplicity that we're putting in place now, and Paul has been here 2 months now is quickly showing up in our ability to engage and put the whole portfolio.

So I think to the extent we've had delays or we definitely have had delays in deals. But to the extent we haven't been able to capture or convert some of those, some of those are self-made. Some of those are absolutely go-to-market imperfections that we have that we're now clarifying. So I feel confident that, that will help.

But if you just step back and you say, at the end of the day, what do you have to offer and what are people buying, we're good. We have the set of skills, the set of technology, the set of partners, the set of capabilities that are needed today across the world. We just have to match the supply signal with the demand signal at the right time and chase enough deals so that the conversion rates so that we're not stressing over conversion rates that slip a point or 2 in any direction. So I'm confident that the portfolio is solid. I'm confident that we're getting better in execution and go-to-market. I'm super confident that we have the ingredients to win, and I'm confident we're going to win.

Yu Lee

And just as a follow-up for Rob. With revenue now declining faster than planned, what cost actions protect margins as you go into next year and as you think about your 3-year outlook? And how do you ensure you're driving cost actions without meaningfully impacting the investments that you need to make to drive revenue growth?

Robert Del Bene

Yes. Yes. So Jonathan, we came into the year with a cost takeout plan -- cost reduction plan that was pretty consistent in terms of year levels. With the revenue dropping, we've increased that cost takeout objective. We've baked that into the -- into this guide. And going forward into fiscal 2028 for us, we are currently evaluating what steps we take under the assumption that we're conservative. So we're going to take the assumption that we have to go deeper. So we're currently evaluating the plans for '28. A little premature to comment on specifics at this time, but it's -- believe me, we're working hard on developing the plans for next year.

Raul Fernandez

And I know we all talking percentages here. But when you just step back a second and you go, okay, you've got a $12 billion backdrop. The difference between winning and losing against the $12 billion backdrop is an incremental $155 million. When I look at what we have to sell, when I look at the size of those deals, when I look at the new offerings that we have, I feel confident when you put it in that context, we can close that gap.

So while percentage-wise, it is what it is. On a total aggregate, can you add enough opportunities into the pipeline to close a gap that today, again, is a run rate gap of about $150 million. The answer is absolutely yes. We just have to execute better.

Operator

Your next question comes from the line of Jamie Friedman from Susquehanna.

James Friedman

I'll ask mine upfront. So, any update on Growth X because that seems important [Technical Difficulty]. And then about the Guidewire announced [Technical Difficulty].

Raul Fernandez

Jamie, we can't hear you, but let me take your Growth X. The modernization as a service is part of that offering. And that's -- as I mentioned in my remarks earlier, that is a great example of an initial win. Again, the promise of legacy code transformation and modernization has been around for a long time. The reality that it can happen at a high quality that can get documented that the deliverables meet the expectations. We feel super confident there.

And we had our Investor Day 140 days ago. We launched many of these initiatives 140 days ago. We've been in market. We've won some deals. We've now completed the initial phases, and we see behind the initial phases of follow-on work as a result of those. And again, those are all kind of agentic in nature. You cut out at the end of your question, so I think I answered your Growth X question, but please repeat the rest of it.

James Friedman

I'm sorry, guys. I'm in a public space. So Guidewire, which is an insurance software competitor to DXC acquired [ hyperxponential ] software earlier this week for 20x revenue. And I think we'd be tone deaf not to ask about this. I'm not asking about your own process. But we don't -- we're not software insurance experts, you are. So do they do something that you do or can do? And how should we be thinking about the scarcity of assets like that in this market?

Raul Fernandez

Yes. Look, I'm not going to get into comments about [ ISV ]. What I can tell you is that the profile of highly sticky customers a very high reoccurring rate of revenue, new products that are agentic in nature and SaaS-like in nature are all the ingredients for high valuations. And that's a fantastic valuation by any measure, right? But I think we have those dynamics. I think -- and I know we have those ingredients in ISP. And I know we have them in other parts of the company as well.

But -- so to answer your question, the attributes that make that interesting for that level of acquisition and that level of consideration for an acquisition are pretty clear. Sticky customers, moats, regulatory moats, highly regulated, a lot of reoccurring, those are -- all those attributes are in our insurance business. And frankly, they're in other parts of DXC as well.

Operator

Your next question comes from the line of James Faucette from Morgan Stanley.

James Faucette

I want to just ask on -- back on the data security point that you're making, particularly as potential customers and customers are evaluating kind of their solution sets, et cetera. Just wondering how -- like if you can give us examples of like how that's being addressed in the cases you've seen. It sounds like you feel comfortable with your offering. Like how do you match that? I think the data security and if we expand it wider to the data sovereignty question is an important one for a lot of potential customers. I'm just wondering how we should think about like what capabilities you have that are matching that?

Raul Fernandez

Yes. As I've looked around and talked to competitors and other players in the space coming at it from different angles, I do believe one of the key ingredients for long-term success when we look back in a period of time is the deep engineering bench that we've got globally and the cross reference of that bench with certain industry skills. I think that combination is key. We have done a really good job with Anthropic and others to certify our engineers 140 days into this relationship. We've got thousands of engineers certified, and we're in the top 10 of all companies in the world with regards to FTE certified for Anthropic.

So first, you have to have the people that have the skills. Then you have to find the demand for those people who have the skills, and you've got to match those in a timely way. So the first step is there. We know our customers have the demand. There are data security, there are sovereignty, there are what goes in a public model, what goes in a private model, how you use these models. There's a lot of complexity here that the complexity fuels the need to get additional heads in the room and help figure out how to move forward. So the demand backdrop for us and for others in this space is very strong if you've got the right combination of raw talent. And I think we do have that as a very strong foundation at DXC.

James Faucette

Got it. Got it. And then back on your execution point, it's interesting because it sounds like there's lots of reevaluation, but at the same time, you feel like when decisions aren't getting made is that maybe you're not quite in the right position, et cetera. Can you just expand on that? And what are the things that you think need to be done internally at DXC to kind of, as you put it, better meet the opportunity?

Raul Fernandez

I think it's clarifying and prioritizing the parts of the product suite that we know have traction, where we have proof points, making sure that those proof points are internally distributed and then ultimately, that sales and marketing and case study material delivered to new prospects. I also do think it's the right set of offerings at the right time. I think our cyber business, our cybersecurity and now the ability to offer the frontier code assurance, where we can take one of your applications, Jamie, we can scan 10 million lines of code really quickly. We can document the vulnerability. We can create a patching framework. We can use our products to go help your teams patch.

That is a very, very hot offering. We are one of the few companies in the world that currently are to do that. We are selling it. And I think that is one of the avenues where there will not be hesitation in purchasing because it's a you must-do situation because, unfortunately, the offensive side, the bad side, the intruders have access to this, and they will take full -- they'll just take full opportunity to exploit. So I think it's a combination of right product, right fit, right timing and then obviously, the mechanical execution of engaging with our clients quickly and getting a decision moved along as quickly as possible without having internal friction and then minimizing external friction on the buy side.

Operator

There are no further questions at this time, and we have reached the end of the Q&A session. I will now turn the call back to Tiffany Horvath for closing remarks.

Tiffany Horvath

Yes. Thank you all for joining us today. We look forward to speaking with you next quarter.

Operator

This concludes today's call. Thank you for attending. You may now disconnect.