Cuộc họp báo cáo kết quả kinh doanh quý 4 năm tài chính 2026 của eGain (EGAN): Tăng trưởng AI và triển vọng FY2027

Những điểm chính

• Doanh thu năm tài chính 2026 tăng 3% so với năm trước lên 91,1 triệu USD. Doanh thu từ khách hàng dùng AI tăng 20%, trong khi ARR từ khách hàng dùng AI tăng 13% lên 54 triệu USD và đạt 72% tổng ARR mảng SaaS.
• Doanh thu quý 4 đạt 22,2 triệu USD so với 23,2 triệu USD cùng kỳ năm trước, phản ánh doanh thu giảm từ các khách hàng sử dụng dịch vụ hội thoại và phân tích truyền thống. Doanh thu từ khách hàng dùng AI tăng 11% so với năm trước.
• EBITDA điều chỉnh năm tài chính 2026 tăng lên 13,6 triệu USD, tương đương biên lợi nhuận 15%, so với 8,6 triệu USD và 10% trong năm tài chính 2025. Dòng tiền từ hoạt động kinh doanh đạt mức kỷ lục 21,2 triệu USD.
• Cho năm tài chính 2027, ban lãnh đạo dự kiến doanh thu từ khách hàng dùng AI đạt 59,5 triệu USD đến 60,5 triệu USD, tăng khoảng 8% đến 10%, và tổng doanh thu đạt 84,5 triệu USD đến 86 triệu USD.
• Ban lãnh đạo dự kiến ARR từ khách hàng dùng AI năm tài chính 2027 sẽ tăng khoảng 20%, nhưng ARR từ khách hàng truyền thống sẽ giảm 60%. Các khoản đầu tư vào cơ hội AI dự kiến sẽ làm giảm biên EBITDA điều chỉnh xuống còn 1% đến 2%.
• eGain cho biết số lượng khách hàng mới thu hút được tăng 27% trong năm tài chính 2026, trong khi số lượng cơ hội tiềm năng trị giá ít nhất 500.000 USD tính theo ARR đã tăng gấp đôi so với năm trước.

Dữ liệu tài chính chính

Vào cuối năm tài chính, ARR từ khách hàng dùng AI chiếm 72% tổng ARR mảng SaaS, tăng từ mức 63% tại thời điểm giữa năm tài chính 2026. Tổng ARR mảng SaaS giảm 1% so với năm trước do các khách hàng truyền thống không sử dụng AI thu hẹp hợp đồng.

Tỷ lệ duy trì doanh thu thuần dựa trên giá trị USD tính trong 12 tháng gần nhất đạt 104% đối với khách hàng dùng AI, so với 120% của một năm trước đó. Tỷ lệ duy trì thuần tính trên tất cả khách hàng đã giảm xuống 93% từ mức 105%. Tổng nghĩa vụ thực hiện còn lại giảm 5% xuống 87 triệu USD, trong khi RPO ngắn hạn giảm 2% xuống 62 triệu USD.

Trong năm tài chính 2026, eGain đã mua lại 1,6 triệu cổ phiếu với giá 11,5 triệu USD, mức giá trung bình là 7,16 USD mỗi cổ phiếu. Công ty còn lại 9,7 triệu USD trong hạn mức ủy quyền mua lại cổ phiếu trị giá 60 triệu USD vào cuối năm.

Kết quả kinh doanh và hoạt động

Ban lãnh đạo đã tái định hình báo cáo của mình xoay quanh các khách hàng dùng AI, được định nghĩa là những khách hàng đang tích cực sử dụng một hoặc nhiều sản phẩm AI của eGain. ARR và doanh thu từ khách hàng dùng AI bao gồm tất cả các sản phẩm mà các khách hàng này mua, không chỉ riêng các sản phẩm AI.

Số lượng khách hàng mới thu hút được tăng 27% trong năm tài chính 2026. Số lượng cơ hội tiềm năng được định giá từ 500.000 USD trở lên tính theo ARR đã tăng gấp đôi, trong khi các cơ hội trong lĩnh vực ngân hàng, dịch vụ tài chính, bảo hiểm và chăm sóc sức khỏe tăng 40%.

Công ty báo cáo việc gia tăng sử dụng các chương trình thử nghiệm trả phí trong số các khách hàng tiềm năng thuộc danh sách Global 2000. Ban lãnh đạo xác định việc chuyển đổi các chương trình thử nghiệm này thành việc triển khai sản xuất quy mô lớn là ưu tiên hàng đầu cho năm tài chính 2027. Theo công ty, một chương trình thử nghiệm tập trung vào tuân thủ đã mang lại kết quả ban đầu với 95% tỷ lệ tự giải quyết và 80% mức độ hài lòng của người dùng.

Các sản phẩm mới ra mắt bao gồm eGain IVA, eGain Agentic Studio và bản phát hành chính thức của eGain Evaluator. Công ty cũng đã giới thiệu eGain AI Knowledge Suite dành cho ngành chăm sóc sức khỏe. Ban lãnh đạo cho biết nhu cầu đang mở rộng từ năng suất của trung tâm liên lạc sang tự phục vụ khách hàng và các trường hợp sử dụng hạ tầng AI rộng lớn hơn.

Gartner đã vinh danh eGain là Đơn vị dẫn đầu trong Báo cáo Magic Quadrant đầu tiên về Các hệ thống quản lý tri thức dịch vụ khách hàng, xếp hạng công ty cao nhất về khả năng thực thi và xa nhất về tầm nhìn toàn diện, theo ban lãnh đạo.

Dự báo của ban lãnh đạo

Ban lãnh đạo dự kiến tăng trưởng doanh thu từ khách hàng dùng AI trong năm tài chính 2027 đạt khoảng 8% đến 10%. Công ty cũng dự báo tăng trưởng ARR từ khách hàng dùng AI khoảng 20%, cùng với mức giảm 20% doanh thu từ khách hàng truyền thống và mức giảm 60% ARR từ khách hàng truyền thống.

Đối với năm tài chính 2030, công ty đặt mục tiêu ARR từ khách hàng dùng AI và tổng ARR mảng SaaS đạt 100 triệu USD đến 120 triệu USD. Công ty cũng mục tiêu doanh thu từ khách hàng dùng AI đạt 105 triệu USD đến 115 triệu USD và tổng doanh thu đạt 110 triệu USD đến 120 triệu USD. Ban lãnh đạo dự kiến đến thời điểm đó, khách hàng dùng AI sẽ chiếm khoảng 100% ARR mảng SaaS và khoảng 95% tổng doanh thu, với biên lợi nhuận gộp SaaS gần 80% và EBITDA điều chỉnh tiếp tục duy trì ở mức dương.

Rủi ro và các điểm cần theo dõi

• Doanh thu từ mảng hội thoại và phân tích truyền thống gây áp lực lên kết quả quý 4. Ban lãnh đạo dự kiến mảng kinh doanh không sử dụng AI về cơ bản sẽ giảm về 0 vào năm tài chính 2030 trên góc độ mô hình hóa, mặc dù một số khách hàng có thể chuyển đổi sang các sản phẩm AI.
• Năm tài chính 2027 kết hợp sự tăng trưởng AI với mức giảm dự kiến 20% doanh thu từ khách hàng truyền thống, dẫn đến dự báo tổng doanh thu thấp hơn doanh thu của năm tài chính 2026.
• Các khoản đầu tư tăng cường vào tiếp thị và AI dự kiến sẽ thu hẹp biên EBITDA điều chỉnh năm tài chính 2027 xuống còn 1% đến 2% và dẫn đến khoản lỗ ròng GAAP.
• Tỷ lệ duy trì doanh thu thuần từ khách hàng dùng AI giảm xuống 104% từ mức 120%. Ban lãnh đạo lưu ý rằng con số của năm trước được hưởng lợi từ hợp đồng mở rộng quy mô lớn với JPMorgan Chase.
• Ban lãnh đạo nhận thấy áp lực tiềm tàng lên giá mảng SaaS từ AI khoảng một đến hai điểm phần trăm trong vòng hai đến ba năm tới, mặc dù các giải pháp AI giá trị gia tăng mới có thể bù đắp một phần tác động này.
• Các chương trình thử nghiệm trả phí tạo ra tiềm năng mở rộng, nhưng việc chuyển đổi chúng thành triển khai sản xuất quy mô lớn vẫn là ưu tiên hàng đầu về mặt thực thi.

Điểm nhấn phần Hỏi & Đáp với chuyên gia phân tích

Ban lãnh đạo cho biết việc thu hút khách hàng mới sẽ là động lực chính trong mô hình tăng trưởng dài hạn của công ty. Chuyển đổi các khách hàng truyền thống sang các sản phẩm AI là một mục tiêu nhưng không dự kiến sẽ là yếu tố đóng góp tăng trưởng chính.

Về chi phí hạ tầng AI, Tổng Giám đốc (CEO) Ashutosh Roy cho biết các dữ liệu tri thức đầu vào chính xác hơn có thể giảm đáng kể chi phí token, đôi khi lên tới 10 lần. eGain cũng sử dụng các mô hình khác nhau cho các nhiệm vụ khác nhau để quản lý chi phí và hiệu suất.

Ông Roy nói thêm rằng sự khác biệt về chất lượng giữa các mô hình tiên phong và mô hình mã nguồn mở thường không quá 10% đến 15% trên các bài kiểm tra chuẩn liên quan mà eGain đánh giá, trong khi chênh lệch chi phí có thể vượt quá 10 lần. Do đó, ban lãnh đạo dự kiến việc điều hướng mô hình sẽ trở nên quan trọng hơn khi các hoạt động xử lý tri thức theo thời gian thực làm tăng mức sử dụng token.

Toàn văn biên bản cuộc họp công bố kết quả kinh doanh

Toàn văn cuộc gọi công bố kết quả kinh doanh

Phần trình bày của ban lãnh đạo

Operator

Good day, and welcome to the eGain Fiscal 2026 Fourth Quarter and Full Year Financial Results Call.

[Operator Instructions]

Please note this event is being recorded. I would now like to turn the conference over to Jim Byers, Investor Relations. Please go ahead.

Jim Byers

Thank you, operator, and good afternoon, everyone. Welcome to eGain's Fiscal 2026 Fourth Quarter and Full Year Financial Results Conference Call. On the call today are eGain's Chief Executive Officer, Ashu Roy; and Chief Financial Officer, Eric Smit.

Before we begin, I would like to remind everyone that during this conference call, management will make certain forward-looking statements, which convey management's expectations, beliefs, plans and objectives regarding future financial and operational performance. Forward-looking statements are generally preceded by words such as believe, plan, intend, expect, anticipate or similar expressions. Forward-looking statements are protected by safe harbor provisions contained in the Private Securities Litigation Reform Act of 1995. These forward-looking statements are subject to a wide range of risks and uncertainties that could cause actual results to differ in material respects.

Information on various factors that could affect eGain's results are detailed in the company's reports filed with the Securities and Exchange Commission. eGain is making these statements as of today, September 3, 2026, and assumes no obligation to publicly update or revise any of the forward-looking information in this conference call. In addition to GAAP results, we will also discuss certain non-GAAP financial measures such as non-GAAP operating income. The tables included with the earnings press release include reconciliation of the historical non-GAAP financial measures to the most directly comparable GAAP financial measures.

eGain's earnings press release can be found by clicking the Press Releases link on the Investor Relations page of eGain's website at egain.com. And along with the earnings release, we will post an updated investor presentation to the Investor Relations page. And lastly, a phone replay of this conference call will be available for 1 week.

And now with that said, I'd like to turn the call over to eGain's CEO, Ashu Roy.

Ashutosh Roy

Thank you, Jim. Good afternoon, everyone. Right at the end of fiscal 2026, the category we've been building toward for years got a name. In July this year, Gartner published its first ever Magic Quadrant for customer service knowledge management systems and named eGain a leader, positioned highest for ability to execute and furthest for completeness of vision. This inaugural Magic Quadrant matters more than just our position in it. This is the first time a top analyst firm has drawn a sharp boundary around this market and explicitly called out knowledge management for customer service as its own category of enterprise infrastructure.

Now they base it on the volume and kind of client inquiries they get in this area. And therefore, they have chosen to invest Magic Quadrant level resources and attention to it. It's a very important signal for the market and the category that's building around it. As we have said, there's good reason this buying category is emerging now. Generative AI has collapsed the old separation between instruction and data. What an AI agent or agentic workflow does in any live customer or employee assistance conversation is determined entirely by the policies, procedures and know-how it is fed.

When that knowledge is wrong, the AI is confidently wrong. When it's stale, the AI doesn't know it's out of date. So knowledge is no more documentation just for humans to optionally use. It is instruction for AI. Wrong knowledge equals wrong AI. Engineering that instruction layer, governing it, operating it continuously is what we call AI Knowledge ops, a term that Gartner reflected in their Magic Quadrant report as something unique and important that eGain brings to this solution. It is the discipline enterprises are now realizing they cannot skip if they want AI to reliably work in production, not just in pilot.

With this market trend and the analyst acknowledgment, let me walk through how fiscal 2026 came together. Before I do that, let me define a term that we will use moving forward, and that is AI customer. An AI customer is an eGain customer who utilizes one or more of our AI offerings. So with that said, let's look at full year fiscal 2026. Our total revenue grew 3% to $91.1 million. Our AI customer revenue grew 20% year-over-year. AI customer ARR grew 13%, and represented 72% of total SaaS ARR at year-end, up from 63% at the midpoint of fiscal 2026. This is an intentional shift in the shape of our customer base. A growing majority of our SaaS ARR now sits with customers who are using one or more of our AI capabilities.

Turning to new business. Our momentum continued to build. In the fourth quarter, we won several new logos. A couple of examples here. First, a leading European insurance company, they set out to automate their service operation with AI and recognized that they needed to put in place a governed knowledge foundation before they could deploy AI automation at scale. So they selected eGain to modernize their knowledge environment and establish that foundation.

Second, a global multi-energy operator serving millions of customers. They faced a familiar barrier to scaling service, fragmented knowledge leading to inconsistent service quality. They are deploying our knowledge platform and AI agent in one contact center. Based on the successful blueprint from that deployment, they will extend to the rest of their contact centers. They also plan to activate self-service channels and leverage the Knowledge Hub across the entire business.

In addition, we added several new paid pilots this quarter. Increasingly, we see buyers wanting to extensively validate our platform in their own environment before committing to a full rollout, and they're willing to pay for it. This is a shift from where we used to be where we were doing a lot of free quick trials and pilots as part of our innovation in 30 days, the 30-day and no-risk pilot that we have.

Converting these paid pilots into at-scale production rollouts is a focus for us this fiscal year. Give you a couple of examples again. One of the world's largest pharmaceutical companies. Their use case is that their experienced scientists and specialists retire or change roles in their R&D teams and the company risks losing a lot of deep tacit expertise. They're using our AI Knowledge Hub to capture that tacit knowledge on a continuous basis and turn it into valuable knowledge for their AI engine.

Second, a global leader in testing inspection and certification. They were facing a hard regulatory deadline, and they needed accurate instant guidance in a compliance-heavy environment. Early pilot results of our deployment indicate that the AI agents deliver 95% self-service resolution, and it's enjoying a strong 80% customer user satisfaction surveys.

Third, a global leader in gaming technology. They operate in a complex environment where every answer has to be guided and correct. Stepping back to the market, I want to share 2 trends that we see emerging in the last couple of quarters. First, businesses are treating knowledge as core AI infrastructure, and their tech and AI teams are actively building on top of this infrastructure, which drives demand for richer platform capabilities like real-time knowledge APIs and stringent service levels. So our growing developer-facing capabilities on our Composer platform are being well received.

Second trend we see is growing interest in customer self-service projects. Several new logos in the recent quarters have started out with self-service deployments, something we did not see a year ago when it was more common to start with contact center-based use cases. While contact center productivity is still of great interest, we sense that businesses are increasingly driving for ROI at scale on their AI investments.

Moving to business momentum in fiscal 2026. Our new logo wins increased 27% year-over-year. As I mentioned earlier, several of the new logos we acquired in fiscal '26 have paid pilots in Global 2000 accounts, and they have significant upside, something we intend to pursue this fiscal year. Our pipeline opportunities valued at $500,000 ARR or more, doubled in count year-over-year. And our core verticals, which are compliance heavy like banking, financial services, insurance and health care, we grew our opportunities in the pipeline by 40% year-over-year, exactly where a trusted knowledge foundation matters the most.

Turning to products. Our innovation continues to accelerate with focus. Everything we launched in last quarter, which is in Q4 during our London eGain Solve event in May, fueled the cycle of knowledge and AI. First, we're increasingly deploying AI in our platform to dramatically automate knowledge management. And the result of that generation and maintenance of trusted knowledge with low effort then drives better instruction to AI that is being used to reliably automate customer service and customer operations.

So a few of the noteworthy announcements of new capabilities we made in May. The first was the eGain IVA, which is an intelligent voice agent. What's unique about it is that it is using the same trusted knowledge platform as we use for all our digital self-service tools. So that consistency and quality is something that now we can offer as a complete omnichannel self-service offering.

Secondly, our eGain Agentic Studio, which is a zero-code application building environment we have launched so that business users can assemble these service use cases end-to-end, multistep complex processes with every step grounded in verified knowledge using assured tools and actions and invoking human oversight when needed. It's a complete platform for service automation using agentic capabilities.

The third, which we had announced in the past is the eGain Evaluator, which is our continuous evaluation tool for AI pipelines, but that is -- we made it generally available, and it's a capability that's getting a lot of interest from our large customers who are looking to drive continuous quality assurance of their agentic pipelines.

And finally, we announced a new vertical for health care, which is our eGain AI Knowledge Suite for health care. And this is a governed knowledge foundation purpose-built for health plan and health systems. We will build on this momentum at our upcoming Solve event in Chicago on October 13 and 14 this year. We'll lay out our view of the year ahead, the shift from knowledge management to knowledge automation and the value of agentic AI assembly on top of trusted knowledge. And of course, we'll announce new capabilities and hear from our customers and partners.

So in conclusion, our sustained bet on AI knowledge, the market and products in fiscal 2026 is showing results. And so we are doubling down, and we intend to lead this market. With that, I'll turn it over to Eric Smit, our CFO, to take you through the financial details. Eric?

Eric Smit

Thanks, Ashu, and thanks, everyone, for joining us today. Before I begin, I'd like to note that we are again using slides to support today's call. We believe this provides helpful context and makes it easier to follow our results and outlook. You can access the slides in the Investor Relations section of our website alongside the webcast.

As Ashu noted, fiscal 2026 demonstrated solid financial execution. Total revenue increased 3% to $91.1 million. AI customer revenue grew 20%. Adjusted EBITDA increased to $13.6 million and cash provided by operating activities reached a record $21.2 million.

I'll review our fourth quarter and full year results, explain the transition in more detail to our customer-based AI metrics and discuss our fiscal 2027 outlook and long-term financial framework.

Starting with the fourth quarter results and starting with revenue. Total revenue was $22.2 million, exceeding both our guidance and Street consensus compared with $23.2 million in the prior year quarter. The year-over-year decline in total revenue primarily reflected the lower revenue from our legacy conversation and analytics customers. AI customer revenue grew 11% year-over-year in the fourth quarter. Looking at gross margins, non-GAAP total gross margin for the quarter was 72% compared to 73% a year ago. Non-GAAP SaaS gross margins were 78% compared to 80% a year ago.

Turning to operating expenses. Non-GAAP operating costs were $14.1 million, up 6% year-over-year and 2% sequentially. Sales and marketing expenses were $5.5 million, up 21% sequentially, reflecting our planned investments in go-to-market initiatives, including the eGain Solve event that we held in London.

Looking at our bottom line, GAAP net income was $1.3 million or $0.05 per basic and diluted share compared with GAAP net income of $30.9 million or $1.13 per basic share and $1.11 per diluted share in the prior year quarter. The prior year results included an approximately $29 million tax benefit from the release of the majority of our valuation allowance. Non-GAAP net income was $2.1 million or $0.08 per share on a basic and diluted basis, exceeding our guidance and Street consensus. This compares with $2.4 million or $0.09 per share on a basic and diluted basis in the year ago quarter.

Adjusted EBITDA was $2.2 million, representing a 10% margin and exceeding our expectations compared to $4.5 million and a 19% margin a year ago. During the quarter, we repurchased 1.4 million shares for $10.1 million at an average price of $7.32 per share.

Turning to our full year results. Looking at our revenue, total revenue was $91.1 million, exceeding our guidance and up 3% year-over-year. Within total revenue, AI customer revenue grew 20% year-over-year. AI customer ARR grew 13% year-over-year and represented 72% of total SaaS ARR at year-end. Looking at gross margins and operating expenses. Non-GAAP total gross margin was 74%, up from 71% in fiscal 2025. Non-GAAP operating costs were $55.3 million compared to $56 million in the prior year.

Turning to the bottom line, balance sheet and cash flows. GAAP net income was $8.9 million or $0.33 per basic share and $0.32 per diluted share compared with $32.3 million or $1.15 per basic share and $1.13 per diluted share in fiscal 2025. As I mentioned, the prior year results included approximately $29 million tax benefit. Non-GAAP net income was $13 million or $0.48 per share on a basic basis and $0.47 per share on a diluted basis, up from non-GAAP net income of $5.7 million or $0.20 per share on a basic and diluted basis in the prior fiscal year.

Adjusted EBITDA increased to $13.6 million, representing a 15% margin, up from $8.6 million and a 10% margin in fiscal 2025. Cash flow from operations reached a record $21.2 million, representing a 23% operating cash flow margin, up from $5.3 million or a 6% operating cash flow margin in fiscal 2025. Cash and cash equivalents totaled $73.3 million at June 30, 2026, compared to $62.9 million at June 30, 2025. During fiscal 2026, we repurchased 1.6 million shares for $11.5 million at an average price of $7.16 per share. At year-end, we had $9.7 million remaining available under the $60 million buyback authorization.

Now turning to our AI customer metrics. As Ashu mentioned, instead of reporting by product hub, going forward, we're now reporting based on whether a customer is actively using one or more of our AI offerings. We call this AI customer ARR and AI customer revenue. And we believe it's a cleaner, more forward-looking way to show our AI adoption spreading across our installed base since many customers now use AI capabilities across multiple parts of our platform rather than within a single hub. This is the framework we'll use going forward.

The strategic rationale is straightforward. We have found that the customers' overall adoption of our AI capabilities, not the specific product SKU or hub they originally purchased is the strongest predictor of long-term retention expansion. To better measure and ultimately maximize that dynamic, we completed a full review of our customer base this year and segmented it into 2 groups: AI customers, meaning those actively engaged with our AI platform and all other customers. This is a meaningful shift in how we think about the business.

Our reporting focus is now on growing ARR per account, which we view as a primary measure of success with a specific mix of products or given customer consumes becomes secondary. We believe this customer base view better reflects how customers deploy our integrated platform, how we manage these relationships and the broader retention and expansion opportunity within our AI customer base. It is now our primary lens for measuring the health of our AI business.

AI customer ARR is defined as total SaaS ARR from customers who are actively utilizing one or more of our AI offerings. This amount includes all offerings associated with the customer and not solely the AI offerings. AI customer revenue is defined as the total revenue generated from customers who actively utilize one or more of our AI offerings, inclusive of their SaaS and professional services revenue. This amount also includes all offerings associated with the customer and not solely the AI offerings.

With that context, here are the metrics. AI customer ARR increased 13% year-over-year and represented 72% of total SaaS ARR at year-end. Total SaaS ARR declined 1% year-over-year, driven by the decline among our legacy non-AI customers.

Turning to our retention rates. Trailing 12-month dollar-based net retention for AI customers was 104% compared to 120% a year ago. As a reminder, we had closed a significant expansion deal with JPMC in Q4 of last fiscal year, which drove that increase in net retention. Net retention for all customers was 93% compared to 105% a year ago. Total remaining performance obligation or RPO of $87 million was down 5% year-over-year and short-term RPO of $62 million was down 2% year-over-year.

Now turning to our outlook. Starting with guidance for the first quarter of fiscal 2027. We expect AI customer revenue of between $13.7 million to $14 million and total revenue of between $20.9 million and $21.4 million.

Turning to the bottom line. For Q1, we expect GAAP net income of $500,000 to $1 million or $0.02 to $0.04 per share, which includes stock-based compensation expense of approximately $900,000. We expect non-GAAP net income of $1.4 million to $2 million or $0.05 to $0.08 per share and adjusted EBITDA of $1.4 million to $1.9 million or a margin of 7% to 9%. For the fiscal year ending June 30, 2027, we expect AI customer revenue of between $59.5 million to $60.5 million, representing growth approximately of 8% to 10%. Total revenue to be between $84.5 million and $86 million.

Our outlook reflects 2 different trends within the business. We expect continued growth from AI customers alongside an estimated 20% decline in revenue from our profitable legacy customers. We are using the cash generation from this non-core business to fund investments in the larger AI opportunity. We expect ARR from AI customers to grow approximately 20% in fiscal 2027, while ARR from legacy customers is expected to decline by 60%.

On the bottom line, we expect GAAP net loss of $2 million to $3 million or $0.08 to $0.11 per share. This includes stock-based comp expense of approximately $4 million, non-GAAP net income of $1 million to $2 million or $0.04 to $0.07 per share and adjusted EBITDA of $650,000 to $1.4 million or a margin of 1% to 2%. We expect weighted average shares outstanding of approximately 26.6 million for the first quarter and 26.8 million for the full fiscal 2027.

Today, we are also introducing a long-term financial model that lays out our targets through fiscal 2030. As we complete our transition to a higher-growth AI-led business, we see fiscal '27 through fiscal '29 as a transition period with total revenue growing both increasingly converging with AI customer revenue growth and fiscal 2030 is a year that convergence is largely complete.

Now turning to our long-term financial model. For fiscal 2030 relative to fiscal 2026, we are targeting AI customer ARR of between $100 million to $120 million, up from $54 million in fiscal 2026, a 17% to 22% CAGR as AI ARR compounds towards scale. Total SaaS ARR of $100 million to $120 million, up from $75 million in fiscal 2026, reflecting substantially complete runoff of non-AI ARR and migration to AI.

For AI customer ARR, we expect that's going to represent approximately 100% of total SaaS ARR, up from 72% in fiscal 2026, effectively a pure-play AI ARR base with increasing contribution from our AI business. AI customer revenue of $105 million to $115 million, representing a 17% to 20% CAGR from the $55 million we generated in fiscal 2026 and a 20% plus growth year-over-year by fiscal 2030, making our underlying AI revenue growth increasingly visible in our total results. And total revenue of $110 million to $120 million, representing approximately 15% to 20% growth year-over-year by fiscal 2030, with total company growth now closely mirroring our AI growth.

AI customer revenue representing approximately 95% of total revenue, up from 60% in 2026, supporting a higher quality valuation framework and SaaS gross margins of approximately 80%, maintaining our attractive software margin profile and adjusted EBITDA margin that remains positive while we fund AI growth, a deliberate balance between growth investments and profitability discipline.

We believe our leadership in AI-powered knowledge management, expanding market opportunity and increased go-to-market investment position eGain to pursue durable growth while maintaining an attractive profitability profile.

So to summarize, in closing, AI customer revenue and ARR both grew at double-digit rates in fiscal 2026, and we completed the shift to a customer level reporting that we believe gives investors a clearer view of the business and strengthens our positioning following Gartner's naming of eGain a leader in the inaugural Magic Quadrant for Customer Service Knowledge Management Systems. We also delivered total revenue growth, strong profitability and record operating cash flow in fiscal 2026.

With our strong balance sheet and cash generation, including the cash we generated from our declining but profitable legacy offerings, we are all in on the AI knowledge opportunity, investing to build on that position and pursue sustainable long-term growth.

Lastly, as Ashu mentioned, we will be hosting an Investor Day and Analyst Day in conjunction with our upcoming eGain Solve customer event on October 13 in Chicago. Additional information and registration details are available on our website. This event is a great opportunity for prospective investors and analysts to meet with customers and learn more about our business. We hope you can join us.

With that, I would like to open the call for questions. Operator?

Operator

[Operator Instructions]

Our first question comes from Jeff Van Rhee with Craig-Hallum.

Phần hỏi đáp

Vijay Homan

This is Vijay on for Jeff. First one for me, just in the target model and kind of here in the prepared remarks, you talked a little bit about running off the non-AI ARR. Is there a time line for that in mind kind of similar to what you had with the messaging business? And then just how does the profitability of those businesses compare to the rest of the business?

Eric Smit

Thanks for that. Yes. So for clarification, if you -- as we sort of described in the model, the expectation is the non-AI business should be substantially -- the goal, obviously, is to convert some of that into the AI business. But from a modeling standpoint, we'd expect that to be to 0 as we get to the 2030 time frame.

Vijay Homan

Got it. And then you talked a little bit on previous earnings calls about some of the potential impacts of AI more generally on the business, maybe pricing pressure on SaaS products. Are you seeing that show up in the business at all? Or is that still kind of expected later down the line?

Ashutosh Roy

This is Ashu here. So I would say that we are seeing some pressure of that, but we are also seeing our ability to create new product offerings, which layer on kind of additional revenue from these value-added AI capabilities. So all in all, the effect has not been as significant as I would have feared. Yet, I mean, we are prepared for it. We do think that there may be -- my sense is 1 or 2 points pressure over the next 2 to 3 years is how I see it. But Eric, do you have anything more to add?

Eric Smit

Exactly. Yes, I think that's sort of aligned at this stage. I think given the construction layer that this is building, it's sort of creating opportunities that are different from what we would have seen historically as well, which I think will obviously impact sort of the way the pricing we approach this.

Vijay Homan

Yes. Got it. And then just for the target model, obviously, I appreciate having that out there. As you look at the growth profile, is there any way you can segment that as far as if you expect 15% or 20% growth, how much of that will be maybe price or new customer adds or adding seats to existing customers or reducing churn? Just what do you think the biggest kind of drivers there will be?

Eric Smit

So I think most of the driver will come from new logo acquisition. I think when we look at the opportunity in front of us, especially with now the backdrop that we're seeing with the Gartner MQ, I think this investment to drive the brand awareness and scale up the customer base will be the primary driver. Obviously, we will work hard to move customers that are in the legacy bucket, but that will not be the primary driver for this growth.

Operator

[Operator Instructions]

Our next question comes from Erik Suppiger with B. Riley.

Unknown Analyst

This is [ Ethan White ] calling on for Erik. Just one question from me. As companies adopt an ecosystem of AI models rather than just using one of the frontier models, does that dynamic create more demand for a knowledge management solution? Can you maybe speak to that dynamic a little more?

Ashutosh Roy

Yes, I'll take that, Eric. Yes, you're right. What we are seeing now is in the last month or so, I'm sure you've seen as well, a lot of talk about people running into sort of token runaway costs and also just cost of AI as the adoption has been pushed hard in enterprises. And what we see with our approach to it is just by being sharper in what you are feeding into these AI tools, you can keep the costs down significantly, sometimes by a factor of 10.

So it's a big advantage by being more precise in how you instruct and guide rather than throwing the kitchen sink of content and context into these models. So that's one thing we see as a very interesting advantage that we bring to the party.

The second one is that we even internally inside the platform tend to be smart about using, if you will, horses for courses, the right models for the right need. And we see that as another way of managing the AI token cost for our clients.

Unknown Analyst

And maybe just one little follow-up. Does that dynamic matter at all in kind of the big frontier models versus open source? Or is that relevant?

Ashutosh Roy

It does matter to some extent, the quality advantage, as you know, in terms of benchmarks and stuff is probably not more than 10% to 15% for most of the relevant benchmarks that we are looking at. And the cost difference can be more than a factor of 10. So yes, it does matter. And what we see is as businesses are doing more and more real-time continuous operation to ensure that their knowledge and know-how is always up to date, and that's going to drive up token usage, and that will then require smarter routing to the relevant capable models.

Operator

[Operator Instructions]

At this time, there are no further questions. I would like to turn the conference back over to eGain management for any closing remarks.

Eric Smit

Thanks, operator, and thanks, everyone, for joining the call today. And again, I encourage all of you out there to look at joining us at the event in Chicago, again, details on the website. Thank you.

Operator

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.