eGain (EGAN) 2027财年第一季度业绩电话会：AI ARR增长11%，重申业绩指引

核心要点

• 2027财年第一季度营收为2190万美元，同比下降7%，主要因传统业务营收下滑约28%，抵消了AI客户带来的增长。
• AI客户营收同比增长7%至1440万美元，占总营收的66%。AI客户ARR同比增长11%至5450万美元，占SaaS总ARR的74%。
• 盈利能力同比下滑。Non-GAAP净利润为240万美元，合每股收益0.09美元；调整后EBITDA从500万美元降至270万美元。
• eGain重申了其2027财年业绩指引，预计总营收为8450万美元至8600万美元，AI客户营收预计增长约8%至10%。
• 管理层继续以截至2027年6月30日AI客户ARR同比增长约20%为目标，同时加大销售和营销投入。
• 截至季度末，公司拥有7130万美元现金及现金等价物，无负债，股票回购授权额度尚余约600万美元。

核心财务数据

AI客户基于美元的净留存率为104%，环比持平，但低于上年同期的117%。受传统客户持续流失影响，全客户净留存率为92%，低于上年同期的102%。

剩余履约义务（RPO）为7100万美元，同比下降18%；短期RPO下降11%至5200万美元。管理层将下滑归因于大型多年期合同续约的时间节点以及传统业务的过渡转型。

业务与运营表现

Gartner在其首份《客户服务知识管理系统魔力象限》报告中将eGain评为领导者。管理层表示，这一认可正在北美和欧洲带来新的业务洽谈，且本季度来自线下活动的潜在客户线索同比几乎翻倍。

eGain将其AI知识中心（AI Knowledge Hub）定位为支撑其更广泛产品组合的核心平台。公司还推出了具备五项能力的AI KnowledgeOps框架：WYCA、kForge、知识图谱（knowledge graphs）、Genie和Smart Evaluator。这些工具旨在识别知识断层、准备AI就绪内容、改进导航、自动化平台任务并测试回答质量。

管理层展示的客户案例包括COUNTRY Financial，该公司报告称在首月内回答准确率超过96%，现场客服升级处理量减少25%。Healthfirst报告称在30天内错误转接的联络减少了72%，而菲利普莫里斯国际（Philip Morris International）则依靠单一受治理的知识库支持64个市场及多种语言的运营。

新签业务包括Technology Credit Union，以及一家为多达1,000名车间操作员部署eGain AI智能体的包装消费品公司。管理层还提及在第一财季之后的10月初签约的一家大型投资管理公司。

本季度，eGain以5.64美元的平均价格斥资360万美元回购了约64.4万股股票。自2026财年初以来，该公司已累计斥资1520万美元回购了约230万股股票。

管理层业绩指引

管理层重申其目标，即截至2027年6月30日AI客户ARR同比增长约20%。该预期假设全年传统业务营收下滑约30%。

对于2030财年，eGain继续将AI客户ARR目标定为1亿美元至1.2亿美元，总营收目标为1.1亿美元至1.2亿美元，AI客户营收目标为1.05亿美元至1.15亿美元。管理层预计届时AI客户将占总营收的约95%。

风险与关注点

• 预计传统业务营收的下滑趋势将从第二财季开始更加显著，其中包括计划逐步关停的传统消息业务。
• 在托管成本基本固定的背景下，传统SaaS营收的下滑导致Non-GAAP SaaS毛利率从81%降至78%。
• 销售与营销支出预计将在本财年余下的三个季度中增加。管理层提醒称，第一财季的盈利能力并不能代表今年余下季度的预期盈利表现。
• AI客户净留存率虽维持在104%，但较上年同期的117%有所下滑。
• 剩余履约义务（RPO）可能因大型多年期合同续约的时间节点而出现较大幅度波动。
• 管理层表示，Gartner的认可尚未显著改变传统客户向AI赋能平台迁移的速度。

分析师问答亮点

管理层指出，AI知识中心（AI Knowledge Hub）是eGain产品战略的基础。在此核心平台之上，公司通过AI KnowledgeOps框架进一步加载相关工具和预构建的AI智能体。

计划中的销售与营销投入初期将重点用于提升知名度并获取更多潜在销售线索。eGain还打算扩大销售开拓团队能力，不过管理层认为目前在销售后期推进和成单促成方面已有充足资源。

管理层表示，传统消息业务的关停已于第一财季末完成，这意味着相关营收下滑将在第二财季显现。

公司将AI客户ARR视为先行指标。管理层预计已签约合同将随着时间推移转化为确认收入，这也解释了为何当前11%的ARR增速高于AI客户营收7%的增速。

财报电话会议完整实录

---

完整财报电话会议逐字稿

管理层陈述

Operator

Thank you for standing by, and welcome to eGain's Fiscal 2027 First Quarter Financial Results Conference Call. [Operator Instructions] I would now like to turn the conference over to Jim Byers, PondelWilkinson, Investor Relations. Please go ahead.

Jim Byers

Thank you, operator, and good afternoon, everyone. Welcome to eGain's Fiscal 2027 First Quarter 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, November 5, 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 eGain's 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 News and Press within 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, and good afternoon, everyone. I'll keep my comments on the quarter brief. Eric will take you through the numbers, but I do want to spend some time on our market because we believe the market and the category we've been building toward for a while is now taking shape. So our first quarter came in as we planned with revenue and profitability ahead of our guidance. Total revenue was $21.9 million, above the high end of our guidance. As expected, that was down 7% year-over-year, reflecting the planned runoff of our legacy business.

AI customer revenue was $14.4 million, up 7% year-over-year and represented 66% of total revenue. AI customer ARR was $54.5 million, up 11% year-over-year and grew to 75% (sic) [ 74% ] of total SaaS ARR from 72% at the end of fiscal 2026. Now I want to turn over to the category that I was talking about. In July, Gartner published its first ever Magic Quadrant for Customer Service Knowledge Management Systems and named eGain a leader. We were positioned highest in ability to execute and furthest in completeness of vision.

In the companion Critical Capabilities report, we received the highest product score for the compliance-driven service center use case. I want to underline why this matters beyond our position in it. Gartner does not initiate a Magic Quadrant on a whim. They do it when enough enterprises, their clients, are asking about a problem that becomes its own buying category. This is the category eGain has been building for a while. It now has a name, a boundary and a recognized leader.

But let's be candid, for many people, knowledge management, quote unquote, still conjures document libraries, intranet and maybe SharePoint on steroids. That is not what this category is about anymore, and it is not what eGain does. So let me explain what we do because it is key to the opportunity. Let me start by summarizing a little bit of what we discussed at our Solve event in Chicago last month, which is our annual customer event in the U.S. I talked about what I call the expert paradox. Simply put, AI works well for experts and struggles otherwise.

Let me explain. As you and we all know, AI is successfully used now in business for coding and personal productivity. In both cases, the user controls what goes into the AI system and the same user can judge the AI output quality. A developer, for example, can tell whether the AI-generated code works using tools and their expertise. Customer service use cases are at the opposite end of that spectrum in terms of user expertise.

So the person being given the AI answer, whether that is a service associate, a sales rep or the end customer is in no position to tell whether the answer is correct or compliant. And in the off chance that they do, they cannot control the input to AI anyway. This paradox largely explains the yawning gap in AI adoption for customer service adoption -- sorry, automation today. So how do we make AI work at scale for such use cases? We do 3 things together. We ensure that AI is fed trusted knowledge. We make sure it has effective guardrails, and we continuously check for output quality.

The policies, procedures and know-how a business feeds its AI systems are, in effect, the instructions that AI follows. Wrong knowledge means wrong AI results. That is what eGain provides, trusted knowledge and guidance for enterprise AI. Whether a business runs our AI agents, Salesforce's Agentforce, Microsoft Copilot or whatever they built themselves, we make sure every answer and every action performed by AI is grounded in knowledge the business can stand behind, kept current as products, policies and regulations change and continuously checked for quality.

At Solve, IDC presented new research sponsored by eGain. Based on a survey of 520 U.S. organizations with 1,000 or more employees, 3 findings stood out. 95% of organizations surveyed are already evaluating, piloting or running AI in their knowledge processes. Trying AI is universal, getting a return is not. The leaders report roughly twice the improvement of laggards on every business impact IDC measured. This is leaders and laggards as measured by their knowledge maturity. And the returns are measured in terms of AI ROI.

What separates them is the foundation. Half of the leaders run a fully operational unified knowledge platform versus 5% of laggards and 94% of leaders have automated more than half their knowledge processes versus 36% of laggards. IDC's conclusion is the one we have been making for some time. AI success does not hinge on the model. It hinges on the knowledge foundation beneath it.

The most powerful part of Solve for us was hearing our customers describe their AI ROI success in their own words. I'll mention 3. COUNTRY Financial, a Fortune 1000 insurer, has given more than 2,500 field agents trusted answers using the eGain platform instead of a list of documents to consult when they're sitting across the client or on the phone with them. Answer accuracy, they have seen is above 96%. Calls into their home office for escalation from their field agents has fallen by 25% in the first month and adoption has reached 70% with no mandates or incentives.

Their 2 lessons for the audience was, first, content beats model. And second, AI is not a project. It's a product you keep operating. Healthfirst, the second one I want to bring out, a New York-based health insurer, talked about publishing 1 governed update to 6,500 users in their business across 50 personalized portals. They feed that knowledge into call routing now, which cuts misrouted contacts by 76% (sic) [ 72% ], and that was the result they saw within 30 days. Knowledge is not only answering questions, it is driving decisions.

And the last one I want to bring out is Philip Morris International. They are now running 64 markets across multiple languages on 1 governed knowledge base on eGain. Their presenter put it best. Knowledge has left the customer care group, he said, it is now used across IT, marketing and field operations. This is a pattern we see again and again. We land in customer service where the bar for accuracy and compliance is high. Once the foundation is working, it becomes the knowledge layer for the rest of the enterprise.

The common thread in all these stories is that trusted AI is not a one-time project. Products launch, rates change, regulations update and the knowledge behind the AI starts getting stale pretty much the day you lock it. So at Solve, we introduced AI KnowledgeOps, a new framework along with a set of capabilities for running knowledge as a continuous operation. It runs as a loop. You start by learning what customers are actually asking from your customer conversations, you find the answer gaps, you prepare and organize the knowledge so AI can use it precisely, you test the answers and then you feed back what you learn from users into the loop.

We announced 5 new capabilities as part of this AI KnowledgeOps to make that loop fast and low effort. The first is a tool called WYCA, which stands for What Your Customers Ask. It analyzes customer conversations to show which questions matter most and runs inside the customer's own firewall. This is crucial for our compliance-oriented customer base. The second is kForge, which turns content scattered across silos and SharePoint, Confluence, ServiceNow, Salesforce, what have you, into AI-ready knowledge with experts approving anything that is high risk.

The third was knowledge graphs. We added that to our knowledge platform. It gives AI agents a map of how our business works, so they can reach and navigate to the right answer at much lower token cost and much faster. The fourth was Genie, which lets our users on the eGain knowledge platform ask the platform and the tools in a conversational interface to write or update content, run reports or build workflows. And finally, the last one and probably the most crucial to close the loop is Smart Evaluator, a product that allows AI engineers to continuously test AI pipelines for answer quality. It flags the wrong answer, it pinpoints where it likely went wrong in the pipeline and suggests a fix.

Here's why this matters commercially for us. For a major financial services company, our team ran the AI KnowledgeOps process 1 time, the first time in about 10 person weeks. That company had been expecting that, that whole exercise would have taken 1 to 2 years. Knowledge management has always been a great goal, but the effort it took to get knowledge ready and maintain it was the biggest barrier in this market, and we believe we have removed most of it.

A question which we ask ourselves and is fair is what stops the large platform vendors from doing it themselves? Well, 3 things. First is the depth of capability. The hard part is not search. It is governance, workflow, permissions, versioning, auditability for regulated businesses. We have solved that for many years, and this is where Gartner scored us highest. The second is we have deeply applied AI to our own platform to remove manual effort to make knowledge projects that typically took years now take weeks. We have automated knowledge management so businesses can operate this loop with ease and speed.

And third, we are open. We are modular. Enterprises will use many AI tools from many vendors. We sit underneath all of them. So customers can change AI models and agentic systems to drive their objectives without lock-in.

Turning to customers in the quarter. We signed several new customers and expansions during the quarter. Let me walk you through some examples because they'll show you where the market is going. The first one is Technology Credit Union. This is a growing multibillion-dollar credit union in the valley here. They recognize that their AI plans for members and employees depend on the knowledge underneath, and they're building that foundation across contact center, digital banking, branches and back office. So all employees, not just contact center agents.

They chose eGain over vertical-specific vendors for our governance, agentic capabilities and our connectivity into their core systems, in this case, Jack Henry and Alkami. Credit unions are now becoming a vertical for us that we are pushing in a dedicated way into. Alongside this new logo, we have customers like Rogue Credit Union, who presented at Solve, Oregon Community Credit Union and SELCO, among others, who are using eGain well beyond member service. And we are seeing that interest build in our pipeline.

The next new logo I want to talk about is a privately held multibillion-dollar consumer packaged goods company and their use cases outside customer service. They're deploying our AI agent to up to 1,000 plant floor operators, and this agent is going to be a virtual expert to assist them. Interestingly, this opportunity started with a self-sign-up on our website. This is a motion we've been building for some time, and it is starting to add opportunities to our funnel.

Another exciting new logo that we closed, but this one in early October, so not in Q1, is one of the oldest and largest investment management companies in America. They are deploying eGain as their trusted knowledge foundation to power all AI projects and thousands of employees. Our existing customers also continue to expand with more use cases and more functions. A large U.S. mutual insurer expanded the use of eGain alongside its investment in Salesforce using eGain Composer, our platform capability. They're making their governed eGain knowledge available inside their Salesforce AI. That is the role I described earlier. If the customer chooses a different AI application, eGain still provides the same consistent trusted knowledge underneath.

Another one I want to bring out is an international bank added our real-time knowledge access capabilities. So the trusted knowledge they have built to start with for customer service now flows automatically into many more of the bank systems and user groups. The pattern is consistent. Customers first establish the knowledge foundation and then AI application and ROI follows.

Looking at demand, the Gartner recognition is showing up at the top of our funnel. While it's early, I just want to bring out a few things. The Gartner Magic Quadrant campaign for marketing has quickly become one of our most productive digital programs, bringing in new conversations across North America and Europe. Our event-sourced leads have nearly doubled year-over-year this quarter as we have scaled marketing. We're seeing more RFPs from large enterprises, many of them inbound, which signals growing awareness.

In addition, we now have focused go-to-market teams, albeit small ones, targeting banks and credit unions, health care and government. And all this is complemented by our self-sign-up push and a growing partner network. To conclude my section, a year ago, very few people were talking about knowledge as a foundational requirement for AI. Today, Gartner has defined the Customer Service Knowledge Management category and named eGain a leader in it. IDC has quantified the AI payoff of getting the knowledge foundation right. And our customers are showing what AI ROI can be in customer service in production at scale.

Now our numbers do not yet reflect the size of that opportunity we understand, but we believe we are in the early innings of a large and durable market. We have the products, the proof points and the balance sheet to lead it, and we are now focused on effectively scaling our go-to-market execution. With that, I'll turn it over to Eric.

Eric Smit

Thanks, Ashu, and thanks, everyone, for joining us today. As a reminder, we are using slides to support today's call. You can find them along with our updated presentation in the Investor Relations section of our website. As Ashu mentioned, we started the year well, delivering first quarter revenue and earnings above the high end of our guidance. I'll review our first quarter results, take you through our AI customer metrics, and touch upon our capital allocation and then discuss our outlook for the second quarter and full fiscal year.

Turning to our first quarter results. Starting with revenue. Total revenue was $21.9 million, down 7% year-over-year, but above the high end of our guidance and Street consensus. AI customer revenue grew 7% year-over-year to $14.4 million, above our guidance and represented 66% of total revenue. The year-over-year decline in total revenue reflects lower revenue from our legacy business, which declined approximately 28%, consistent with the transition we outlined last quarter.

Turning to gross margins. Non-GAAP total gross margin was 74% compared with 76% a year ago. Non-GAAP SaaS gross margin was 78% compared to 81% a year ago. The decline reflects the lower legacy SaaS revenue against a largely fixed hosting cost base.

Turning to operating expenses. Non-GAAP operating expenses were $13.6 million, up 4% year-over-year and down 4% sequentially. Sales and marketing expense were $4.6 million, up 17% year-over-year and down 16% sequentially, primarily reflecting the timing of program spending. Our largest event of the year, Solve26 in Chicago, that took place in October. So that spending falls in our second quarter. We expect sales and marketing spending to step up over the rest of the fiscal year as we expand the go-to-market programs Ashu described with a focus on converting growing market awareness into pipeline, new customers and recurring revenue.

Turning to the bottom line. GAAP net income was $1.6 million or $0.06 per share compared with $2.6 million (sic) [ $2.8 million ] or $0.10 per share a year ago. Non-GAAP net income was $2.4 million or $0.09 per share, exceeding our guidance and Street consensus compared to $4.7 million or $0.17 per share a year ago. Adjusted EBITDA was $2.7 million, a 12% margin compared with $5.0 million or a 21% margin a year ago.

Turning to cash flow and the balance sheet. Cash provided by operating activities was $1.8 million compared with $10.4 million a year ago. The difference is primarily due to the timing of an annual payment from a large federal customer. Last year, we received that payment in our first quarter. This year, it arrived early in the fourth quarter of fiscal 2026, contributing to the strong cash flow in that quarter. We ended the quarter with $71.3 million in cash and cash equivalents and no debt.

Moving to our capital allocation. Our capital allocation priorities are unchanged. First, invest in AI growth across product development and go-to-market to convert our category leadership into AI customer ARR. Second, maintain a strong debt-free balance sheet with significant liquidity relative to our scale. And third, return capital opportunistically. During the quarter, we repurchased approximately 644,000 shares for $3.6 million at an average price of $5.64 per share. Since the start of fiscal 2026, we have returned $15.2 million to shareholders through the repurchase of approximately 2.3 million shares, and we have approximately $6.0 million remaining under the authorization.

Now turning to our AI customer metrics. As we introduced last year, we report our business based on whether a customer is on our current AI-enabled platform rather than by product. AI capabilities, as Ashu described, are embedded across our AI-enabled platform, including identifying knowledge gaps, structuring content, authoring, verifying answers and powering AI. We, therefore, measure AI adoption at the customer level rather than by the individual product. Our focus is on growing the number of AI customers and expanding those relationships over time.

AI customer ARR includes all SaaS ARR from customers on our current AI-enabled platform and AI customer revenue includes SaaS and professional service revenue generated from those customers. With that context, AI customer ARR grew 11% year-over-year to $54.5 million as of September 30 and represented 74% of total SaaS ARR, up from 65% a year ago. Total SaaS ARR was $73.4 million, down 4% year-over-year, reflecting the continued declines among our legacy customers.

Turning to retention. Trailing 12-month dollar-based net retention for AI customers was 104%, consistent with the fourth quarter compared with 117% a year ago. As we noted last quarter, the year ago period included the significant expansion with the U.S. megabank that we signed in the fourth quarter of fiscal 2025. Net retention for all customers was 92% compared with 102% a year ago, reflecting the expected runoff of our legacy customers. Total remaining performance obligations were $71.0 million, down 18% year-over-year and short-term RPO was $52.0 million, down 11%. The decline reflects the timing of large multiyear renewals and the continued runoff of our legacy business.

Because RPO can swing significantly up or down with the timing of large multiyear renewals, we focus on AI customer ARR and net retention as a primary measure for our customer base. As AI customers become a larger share of our business, we expect their growth to increasingly drive our consolidated results.

Now turning to our outlook. Before the numbers, let me give you some context on how we are thinking about the rest of the year. First, investments. Part of the first quarter profitability outperformance reflected the timing of sales and marketing spending. We expect that spending to increase beginning in the second quarter, including the Solve26 event. As we expand the go-to-market programs, our focus is on converting increased market awareness into qualified pipeline, new customers and recurring revenue.

Second, the legacy transition. We expect the runoff of legacy revenue to become more pronounced beginning the second quarter, including the planned wind down of our legacy messaging business. Our full year outlook continues to assume an approximately 30% decline in our legacy business revenue. We remain comfortable with our full year outlook. Our balance sheet and operating cash generation gives us flexibility to fund our AI growth initiatives. Our full year guidance reflects a meaningful increase in sales and marketing investments over the remaining 3 quarters. As a result, first quarter profitability is not representative of the quarterly earnings levels we expect for the balance of the year.

So for the second quarter of fiscal 2027, we expect total revenue of $21.0 million to $21.5 million, AI customer revenue of $14.5 million to $14.8 million, GAAP net loss of $500,000 to net income of $200,000 or a loss of $0.02 to an income of $0.01 per share, including approximately $700,000 of stock-based compensation expense. Non-GAAP net income of $200,000 to $900,000 or $0.01 to $0.03 per share, adjusted EBITDA of breakeven to $500,000 or a margin of 0% to 2%.

For the full year ending June 30, 2027, we are reiterating our guidance. We expect total revenue of $84.5 million to $86.0 million, AI customer revenue of $59.5 million to $60.5 million, representing growth of approximately 8% to 10%. We continue to target approximately 20% year-over-year growth in AI customer ARR as of June 30, 2027. A GAAP net loss of $2.0 million to $3.0 million or $0.08 to $0.11 per share, including approximately $4.0 million of stock-based compensation expense, non-GAAP net income of $1.0 million to $2.0 million or $0.04 to $0.07 per share. Adjusted EBITDA of $650,000 to $1.4 million or a margin of 1% to 2%. We expect weighted average shares outstanding of approximately 25.8 million for the second quarter and 26.0 million for the full fiscal year.

Finally, a brief reminder of the long-term financial model we introduced last quarter. We see fiscal 2027 through fiscal 2029 as a transition period, with total revenue growth converging towards AI customer growth as the legacy business runs off. By fiscal 2030, we expect that transition to be largely complete. Our fiscal 2030 targets are AI customer ARR of $100 million to $120 million, up from approximately $54 million in fiscal 2026, a 17% to 20% compounded annual growth rate. Total SaaS ARR of $100 million to $120 million with AI customers representing substantially all of it, up from 72% in fiscal 2026.

AI customer revenue of $105 million to $115 million, up from approximately $55 million in fiscal 2026. Total revenue of $110 million to $120 million with AI customer revenue approximately 95% of the total, up from 60% in fiscal 2026. SaaS gross margins of approximately 80% with positive adjusted EBITDA margins as we invest in growth.

The result we are building toward is a larger, faster-growing AI knowledge infrastructure company with a greater share of recurring AI customer revenue and significant expansion opportunity within our installed base. We have shown we can run this business profitably at our current scale. We are choosing to trade some near-term profitability for growth because we believe the size of this opportunity warrants it.

So to summarize, we delivered first quarter results ahead of our guidance, and we are reiterating our full year outlook. AI needs the right instructions, and eGain is the instruction layer for enterprise AI, recognized as a leader in the first-ever Gartner Magic Quadrant for our market. AI customers now account for 74% of our SaaS ARR on a path to substantially all of it by fiscal 2030. And with $71.3 million in cash and no debt, we have the resources to fund the plan and turn category leadership into durable long-term growth. With that, I'd like to open the call for questions. Operator?

Operator

[Operator Instructions]

And your first question comes from the line of Jeff Van Rhee with Craig-Hallum.

分析师问答

Daniel Hibshman

This is Daniel on for Jeff. Maybe, Ashu, if you would want to touch a little bit on the AI product suite, obviously, has been expanding a lot. There's a whole suite of things that have been rolled out and every Solve event, there's a lot more that gets announced. Maybe whether it's AI KnowledgeOps or AI agents, too, there's a lot of products that have been announced. Maybe you could talk to what you expect to be the biggest needle movers in the AI portfolio and maybe what products or product developments you would want investors to be focused on?

Ashutosh Roy

Good question. So the way we think of it is the foundation first, and that's the AI Knowledge Hub. That is where it all starts. And then all these tools and products that go around the platform then get added on top. But what we have done with the AI KnowledgeOps as a framework is now to guide more and more people, what we have heard from customers, including at Solve was you have so many things to offer. We would love for you to tell us exactly how it all works together and how to go about it from where we are today to where we want to be.

So it all starts with AI Knowledge Hub and then gets -- so if you think of it as layers, it's sort of AI Knowledge Hub as the core platform. And then on top of it, you have tools and you have these prebuilt AI agents, right? So in the target market we are going after, which is the enterprise, I think AI Knowledge Hub with a platform is where it all starts.

Daniel Hibshman

Okay. That's helpful. And then maybe just on the S&M investments you're planning to make. Just walk through a little bit more of the plan there, what you're thinking to change in terms of is that about certain leaders? Is that about more quota-carrying reps we're targeting a particular geography? Just how you're thinking about what those plans entail?

Ashutosh Roy

Right now, we -- as I mentioned, so we've scaled up our front-end marketing investments to drive more awareness and more footfall, if you will, into our digital and physical presence, right? So that's where we are most focused on in terms of increasing our spend on sales and marketing.

And then the second bit behind that is building out the capacity to engage with those early leads and turn them into opportunities. So building out the SDR team, if you will, to make sure that we have got the right level of expertise and front-end engagement for those who we are able to bring in as leads. So I think those are the top 2 things for us right now. The back end of the funnel, which is more around selling and closing, we think that we have enough capacity for that. So it's really about filling the top of the funnel.

Daniel Hibshman

And then just 1 modeling question, Eric. In terms of the messaging business exit, I know the first half of that was achieved in 2Q '26. So it took half impact to Q2 and then another half of the impact came in Q3. And then the remaining half of the messaging business exit, I know that was -- that happened here in Q1. Just in terms of thinking whether we got a full quarter of impact or not, was that exited at the beginning of the quarter, near the end? How much of a quarter of impact did we get from that messaging exit?

Eric Smit

So that was the full quarter. It finished up at the end of the quarter. So that's why we mentioned that we expect to see that decline in Q2.

Operator

And your next question comes from the line of Eric Suppiger with B. Riley Securities.

Ethan Widell

This is Ethan Widell on for Eric. To start off with, so AI customer revenue up 7% versus 11% for ARR. Can you maybe help bridge the gap there between those 2 numbers and maybe what we should expect going forward?

Eric Smit

Yes. So I think, obviously, like you would typically see with the ARR is always the leading indicator as you grow business, then as those contracts get signed, that would take time for that to convert into the revenue going forward. So I think for us, as we -- as you look to the guidance, I think we're targeting a continued improvement in the ARR growth rate, targeting that by the end of the year, we'd be up to 20% whereas currently, our guidance for the revenue is sort of roughly half of that, and that's more just a lag between the ARR booking and then the revenue that gets recognized.

Ethan Widell

Okay. That makes sense. And then so when we look at the kind of legacy ARR decline, of those customers rolling off, maybe can you speak to what share of that is customers churning out versus migrating to like the AI platform? And has the Gartner recognition changed kind of what that looks like?

Ashutosh Roy

It's a good question. We are working on it. But right now, it's not like we have seen a significant shift there.

Ethan Widell

Understood. Well, it seems like a very powerful validation of your technology. So congrats on that, and that's all for me.

Operator

[Operator Instructions]

And I'm showing no further questions at this time. I would like to hand it back to the management for closing remarks.

Eric Smit

Great. Well, thanks, everyone, for taking the time today. Look forward to continue updating you after our Q2 results. Thanks. And I guess 1 final item, I'll be at the ROTH investor event later this month. So it would be great if anybody in the New York area that's participating. Great to meet up with you. Thank you.

Operator

Thank you. And ladies and gentlemen, this concludes today's conference call. Thank you all for joining. You may now disconnect.