Prepared remarks
Greetings, and welcome to the Varonis Systems, Inc. Second Quarter 2026 Earnings Conference Call. As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Tim Perz, Investor Relations. Thank you. You may begin.
Thank you, operator. Good afternoon. Thank you for joining us today to review Varonis second quarter 2026 financial results. With me on the call today are Yakov Faitelson, Chief Executive Officer; and Guy Melamed, Chief Financial Officer and Chief Operating Officer of Varonis. After preliminary remarks, we will open the call to a question-and-answer session. During this call, we may make statements related to our business that would be considered forward-looking statements under federal securities laws, including projections of future operating results for our third quarter and full year-ending December 31, 2026. Due to a number of factors, actual results may differ materially from those set forth in such statements. These factors are set forth in the earnings press release that we issued today under the section captioned Forward-Looking Statements, and these and other important risk factors are described more fully in our reports filed with the Securities and Exchange Commission. We encourage all investors to read our SEC filings. These statements reflect our views only as of today and should not be relied upon as representing our views as of any subsequent date. Varonis expressly disclaims any application or undertaking to release publicly any updates or revisions to any forward-looking statements made herein. Additionally, non-GAAP financial measures will be discussed on this conference call. A reconciliation for the most directly comparable GAAP financial measures is available in our second quarter 2026 earnings press release and our investor presentation, which can be found at varonis.com in the Investor Relations section. Lastly, please note that a webcast of today's call is available on our website in the Investor Relations section. With that, I'd like to turn the call over to our Chief Executive Officer, Yaki Faitelson. Yaki?
Thanks, Tim, and good afternoon, everyone. We appreciate you joining us to discuss our second quarter 2026 results. In Q2, SaaS ARR, excluding conversions, increased 25% year-over-year to $598.1 million and total SaaS ARR, including conversion was $726 million. The strong underlying trends we have seen in recent quarters continued in Q2, and we believe that our performance could have been even better if we had closed a few large deals that slipped due to the rumors in the news in the final week of the quarter. However, we have since closed some of those deals in July and Q3 is off to a strong start, which gives us the confidence to raise our full year guidance for SaaS ARR above the second quarter beat. We are also seeing momentum in our pipeline because of the secular tailwind from the need to secure AI and the data that's powered by it as well as benefits from our expanded offering. Let me spend a moment on the positive underlying trends that we saw in Q2. First, we saw SaaS ARR from new logos grow by more than 20% in the quarter, which is an important signal that organizations are recognizing Varonis' ability to secure their data and AI environments. Second, we saw growing momentum from our newer products, including Atlas, Interceptor and database activity monitoring, which reinforces the breadth of our platform and the expanding set of use cases we can address for our customers. Guy will review our results and guidance in more detail shortly. Before discussing our key customer wins from the quarter, I want to step back from our near-term results and highlight why we believe Varonis is best positioned to help companies safely adopt AI and prevent data breaches. Our opportunity starts with a simple reality. As companies look to adopt AI, the value and risk associated with their data increases. Customers want to move faster and unlock the productivity benefits of AI, but they cannot do that confidently if sensitive information is overexposed, permissions are too broad and activity is not monitored or managed. This is exactly the problem Varonis was built to solve. We see three areas where this shift is creating urgency: understanding and securing the data itself, controlling how AI systems and agents interact with data and defending against attackers that are using AI to move faster and operate at greater scale. First, customers need to know where their sensitive data lives, who can access it and whether that access is appropriate. AI makes this more important because it reduces the friction between employees, agents and the information those identities can reach. Data that was already exposed becomes easier to find, summarize and act on. Second, organizations need guardrails around the models, agents and pipelines that are being connected to enterprise data. It is not enough to know that an AI tool exists. Customers need to understand what data it can touch, what permissions it inherits, what actions it can take and whether its behavior is normal. In an agentic environment, activity that once required a person to invest time searching, downloading and moving files can happen in seconds. That speed raises the stakes for automated remediation and real-time detection. Third, customers are facing a threat environment where adversaries are also using AI. Attackers can generate more convincing social engineering campaigns, automate their attacks and scale techniques that used to require more time and specialized resources. The attack surface also expands as agents gain access to email, collaboration tools, applications and data stores. This reinforces the need for security controls that are continuous, automated and deeply connected to the data layer. These dynamics create a strong secular tailwind for Varonis. Customers are asking the same sets of questions with greater urgency: What sensitive data do we have, where is it exposed, who or what can access it and how do we reduce risk without slowing the business down? Our answer is to secure data from the inside out through automation. This is also why data security and AI security cannot be treated as separate problems. AI inventory is more valuable when it is tied to the data that AI systems can access. Runtime guardrails are stronger when they have context around identity, data exposure and accurate classification and labeling. Detection is more effective when it is informed by normal data access patterns. Customers want to move quickly, but they need confidence that the right controls are in place. Automation is the common thread. The traditional approach to security was built for a slow world when humans could review access, investigate alerts and manually clean up risk. That model does not work when data volumes are exploding, permissions are constantly changing and AI agents can act at machine speed. In our view, the only practical way to manage AI-driven risk is with AI-driven defense. With that, I would like to briefly discuss a couple of key customer wins from Q2 that show how these themes are translating into real demand. This quarter, a health care organization with more than 40,000 employees became a Varonis customer. This company was looking to deploy AI in order to deliver better patient care and needed to establish guardrails around AI usage, monitor how data was being leveraged across more than 100 AI projects and automatically remediate overexposed HIPAA and PII data. Our risk assessment uncovered significant exposure of sensitive data, including patient records and HIPAA data, and the customer ultimately purchased Varonis for Atlas Complete to safely adopt AI tools, including Copilot, cloud and internally built LLMs, as well as Varonis for AWS, Azure, Databricks, Snowflake, Microsoft 365 and NVIDIA. Another highlight from this quarter was a financial services company that expanded its Varonis deployment after performing a vendor consolidation review. The customer was using Varonis to secure on-premise file servers and wanted to consolidate multiple point products such as PAM and email security into our platform as part of a broader vendor rationalization initiative led by its CISO. They are also preparing to adopt AI and leverage automatic remediation to prevent attacks. We demonstrated how Varonis could secure data access across cloud, SaaS and AI environments while replacing several stand-alone tools. During the cloud evaluation, we automatically detected and prevented a significant security incident, which further validated the value of our platform. The customer ultimately expanded its coverage with Varonis for Atlas Complete, Interceptor, DAM as well as Varonis for IaaS, Salesforce and Microsoft 365. In summary, our Q2 results were highlighted by SaaS ARR, excluding conversion growth of 25%, SaaS ARR from new logos growing more than 20% and growing momentum from our newer products, including Atlas, Interceptor and database activity monitoring. Organizations are prioritizing data and AI security, and we are uniquely positioned to help customers secure sensitive data, govern AI and automate risk reduction. With that, let me turn the call over to Guy. Guy?
Thanks, Yaki. Good afternoon, everyone. Thank you for joining us today. Our second quarter performance represents strong new logo activity as our team continues to execute on the tailwinds in our business and help customers secure their data and AI, and we continue to see increasing adoption of our newer products. While we came in above the top end of our guidance range, we believe our Q2 upside was limited because of some large deals that slipped at the end of the quarter due to the rumors in the news. As Yaki said, we have since closed some of these deals, including a 7-figure deal, and the third quarter is off to a strong start. The Q2 outperformance, coupled with a strong start to Q3 and our healthy pipeline, gives us the confidence to raise our full year guidance above the beat for SaaS ARR growth, excluding conversions and also for free cash flow. In the second quarter, SaaS ARR, excluding conversions, increased 25% year-over-year to $598.1 million, and total SaaS ARR was $726 million. In Q2, we had $11.4 million of conversion ARR, and we finished the quarter with approximately $59.3 million of non-SaaS ARR remaining. Year-to-date, we generated $69.1 million of free cash flow, down from $82.7 million in the same period last year, which reflects the previously communicated headwind from the end-of-life announcement of our on-prem platform and also includes approximately $11.9 million of acquisition-related costs related to the accounting treatment of our acquisitions. Adjusting for the acquisition-related costs, year-to-date free cash flow would have been approximately $81 million. We are confident in our cash flow generation and are raising our full year free cash flow guidance by $5 million. Now I'd like to recap our Q2 results in more detail. In the second quarter, total revenues were $180 million, up 18% year-over-year. SaaS revenues were $171.7 million. Term license subscription revenues were $4.2 million and maintenance and services revenues were $4.1 million. Our SaaS renewal rate was over 90%. Moving down to the income statement. I'll be discussing non-GAAP results going forward. Gross profit for the second quarter was $139.9 million, representing a gross margin of 77.7% compared to 80.6% in the second quarter of 2025. Our gross margin continues to be healthy and in line with our long-term target set at our Investor Day. Operating expenses in the second quarter totaled $136.1 million. As a result, second quarter operating income was $3.7 million or an operating margin of 2.1%. This compares to an operating loss of $1.9 million or an operating margin of negative 1.2% in the same period last year. Second quarter ARR contribution margin was 13.3%, down from 16.5% last year. This is in line with our expectations and as a reminder, is impacted in 2026 due to the end of life for our self-hosted platform. During the quarter, we had financial income of approximately $7 million, driven primarily by interest income on our cash, deposits and investments in marketable securities. Net income for the second quarter of 2026 was $5.3 million or net income of $0.04 per diluted share compared to net income of $3.8 million or $0.03 per diluted share for the second quarter of 2025. This is based on 130.8 million and 135.2 million diluted shares outstanding for Q2 2026 and Q2 2025, respectively. As of June 30, 2026, we had $911.5 million in cash, cash equivalents, short-term deposits and marketable securities. For the six months ended June 30, 2026, we generated $80.1 million of cash from operations compared to $89.3 million generated in the same period last year, and CapEx was $9.4 million compared to $5.7 million in the same period last year. As a reminder, we will provide quarterly SaaS ARR, excluding conversion guidance for this year only. We are doing this because of the difficulty in modeling the year-over-year growth rates due to the impact of conversions in 2025 and 2026. We're also providing a bridge to quarterly total SaaS ARR in our investor deck, which again assumes zero conversions from our guidance perspective to the upcoming quarter. For the full year 2026, we will provide annual guidance for both SaaS ARR, excluding conversions and total SaaS ARR. For more information, please see our earnings deck in our Investor Relations website, which includes a more detailed breakdown of our financial guidance. For the third quarter of 2026, we expect SaaS ARR growth of 22% to 23%, excluding conversions, total revenues of $185 million to $188 million, representing growth of 14% to 16%, non-GAAP operating income of $2.5 million to $3.5 million and non-GAAP net income per diluted share in the range of $0.02 to $0.03. This assumes 131.1 million diluted shares outstanding. For the full year 2026, we now expect total SaaS ARR of $819 million to $850 million, representing growth of 28% to 33%. This represents SaaS ARR growth of 20% to 21%, excluding conversions. Free cash flow of $105 million to $110 million, total revenues of $735 million to $739 million, representing growth of 18% to 19%; non-GAAP operating income of $11 million to $13 million; non-GAAP net income per diluted share in the range of $0.14 to $0.15. This assumes 131.5 million diluted shares outstanding. In summary, we continue to see many positive trends, including healthy new logo momentum, strong pipeline activity and a growing urgency from customers to secure AI systems and the data that powers them, which reinforces our confidence in the durability of our platform and the massive opportunity ahead. This confidence allows us to raise our full year outlook for SaaS ARR, excluding conversions growth, above the second quarter beat to 21% at the midpoint to $769 million to $775 million, which is a $5 million raise over last quarter. Now I'd like to turn it back over to Tim.
Before we proceed to Q&A, I would like to make one additional comment. You may have seen recent speculation in the media related to a potential transaction involving Varonis. As a matter of policy, we do not discuss rumors or speculation, and we will make no further comment on this. With that, we will be happy to take your questions on the quarter. Operator?
Questions and answers
The first question is from Saket Kalia from Barclays.
I'll keep it to one. Guy, it sounds like there were some deals that pushed out at the end of the second quarter. Maybe the question is, can you talk about what the cadence was like of business through the quarter until those headlines hit? And how much of an impact could that have had on net new ARR in Q2 if you had to size it?
Thanks for the question, Saket. There were a lot of positive trends that we saw in Q2, and we think we could have done even better if it wasn't for some of those large deals that were impacted by the rumors in the news. We've already closed some of these in July, including a 7-figure deal, and we expect to close more of them during the third quarter. There are three factors which gave us the confidence to raise our full year free cash flow and SaaS ARR guidance, excluding conversions. Number one, in Q2, we saw new logo ACV grow more than 20%. Two, we saw increasing adoption of Atlas, DAM and Interceptor, which gives us confidence in our upsell motion over time; and three, we continue to see a healthy pipeline for the back end of the year, driven by the need to secure AI and the data that powers it. These positive tailwinds that we see give us continued confidence in our ability to grow 20-plus percent over the long term.
The next question is from Matt Hedberg from RBC.
Maybe as a follow-up to that, obviously, Ethos has been in the news, and it's seemingly generating a lot of cyber pipeline out there. I'm wondering if you could talk to how perhaps Ethos was driving some accelerated customer conversations? And have you had any luck, do you think, converting any of that pipeline thus far?
Yes. I think that what it does is definitely exposing vulnerabilities. But what we see that is really starting to drive the business and build a lot of pipeline is where customers are in their maturity curve. A lot of them are investing heavily in AI. Some of it is experimental and now they are trying to get ROI. But the bottleneck is connecting these agents securely to data. What you saw happen with OpenAI and Hugging Face, not exactly the same but in different shape or form, is now happening a lot within organizations: a misconfigured agent that you don't know what it is doing can inflict a lot of damage and can be a breach. This is what organizations are understanding: they need to do something. So we see CISOs and also Chief AI Officers coming and spending a lot of time with us. They understand that AI without data is nothing, and you need to have the full AI lifecycle in order to take it to production in the right way, to make sure you're in compliance and then to take care of the pipeline. This is what we see that's starting to generate a lot of pipeline. We also see massive synergies between Atlas and the DSP, the data security platform. This is something that works very well. So the net-net is that AI connected to data security can be a massive liability, and we really can be the foundation to realize gains from this technology.
The next question is from Meta Marshall from Morgan Stanley.
Just wanted to get into kind of on the new ACV that you are seeing, are there any kind of characteristics of those new customers that they're either larger or adding more products kind of upfront? Any details on that would be helpful.
What we see essentially is everything we said: you need really three ingredients. You need coverage, you need automation to define, fix and alert. And you also need scale. And this is what we have. What's happening is that with AI, before when it was all human-driven, you needed to put a lot of effort in order to expose the data security problems that you have. An agent that is running almost all the time will try to access data, create abnormal behaviors and be nondeterministic. You can give an agent the same instruction twice and it will do something completely different. So we definitely see that customers are buying more upfront. They are buying more platforms and trying to connect them to more AI systems.
The next question is from Brian Essex from JPMorgan.
I was wondering if I could actually follow up on Saket's question. Yaki, could you dig into some of the — I guess, the nature of some of those slipped deals? How did you get customers comfortable that your relationship maybe would not be disrupted given the news that was out there? How complex were these deals? And what gives you the level of confidence that you can close these deals for the remainder of the year?
Like any other software company, a lot is happening in the last week of the quarter. So just the timing was bad. For us, it's business as usual. We're building high-value products and making sure that our customers are successful with them. At the end, when you need approvals and such, these events can generate additional conversation that can elongate some sales cycles. But we're talking to our customers and explaining to them we cannot comment on any rumors, but we are committed to them, committed to their success, keep innovating and make sure that everything is working. As Guy said, some of them we already closed, and we are in conversation with most of them. So far, it's going very well.
And when you look at the guidance characteristics, when you look at the raise by $5 million, about half of that raise is coming from that Q2 beat and our raise above that beat only includes some of the slip deals that happened at the end of the quarter. We're keeping the same responsible guidance philosophy from our perspective. We haven't changed that. But when we look at the beginning of July, we had a strong start, and we're able to close some of those deals already, and the expectation is that we'll close for the most part most of those deals throughout the quarter.
The next question is from Rob Owens from Piper Sandler.
Nice to see the new product momentum that you guys talked about with the three solutions, Atlas, Interceptor and the Database Access Management. My question is really around, is this doing anything relative to deal complexity, which could be impacting sales cycles, number one? And number two, are you finding it's a new buyer persona within organizations as the portfolio has expanded pretty rapidly and seems more kind of runtime defensive. Are you selling to a different buyer in organizations at this point?
Mostly the buyer is the CISO. It's also very interesting that the Chief AI Officer is emerging; we're also talking to them, but the CISO is bringing them in because what is happening today in organizations, as I said earlier in the adoption curve, is that it's very hard for them to safely connect AI to data. So it became a business enabler. Interceptor is very synergetic to our MDDR. It's also very interesting how the MDDR is transforming to AI detection and response. With databases, customers want to cover all the data and the database activity monitoring. Remember, these are new products for us, and we also have a steep maturity curve with them, and we are starting to replace incumbents like Imperva and IBM Guardium. So this also works for us. But it's usually just the same motion with some contribution from Chief AI Officers.
And when we analyze the sales cycles, we haven't seen any change in terms of a longer period. So we're basically tracking the same relatively consistent sales cycles even with us selling a larger platform.
The next question is from Roger Boyd from UBS.
Guy, I know you're not guiding to ARR with conversions and that assumption for $50 million to $75 million in conversions for the year was unchanged. But can you talk qualitatively about what you saw in the first half of the year, kind of where that stood relative to your expectations for the first half? And anything we should be mindful of on conversions in the back half of the year, particularly when we look at federal in Q3?
Absolutely. First, I want to make sure it's clear from an investor perspective that the most important metric to track in terms of the health of the business is SaaS ARR, excluding conversions. That's the reason we're guiding that on a quarterly basis. We provided a bear case and a bull case range between $50 million to $75 million at the beginning of the year, and we said throughout the year we would not update that number because, a, that's not the focus; and b, we've seen customers that were up for renewal in the first part of the year asking to extend the on-prem subscription until the end of the year with the expectation that the conversion will happen in the second part of the year. So guiding that on a quarterly basis would generate more confusion. When I track the actual percentages, in the first part of the year we were roughly converting about 50% of the expected number. If we continue at that pace, we will be at the low end of our $50 million expectation. But if we convert in the second part of the year two-thirds of what is expected to come up for renewal and conversion, we will be at the midpoint. Because many customers wanted to wait until the later part of the year to actually do that conversion, getting to that two-thirds is achievable in our view. But again, I want investors to understand that the right focus should be on SaaS ARR, excluding conversion because that is the metric that shows how we're performing now and will indicate how we continue to grow post transition.
The next question is from Jason Ader from William Blair.
Can you guys talk about the go-to-market strategy? I know there was a question earlier about who the buyer is. I assume that's not changing that much. But maybe the overall strategy, how it might be changing to capture demand related to AI readiness? Are you guys doing anything different than what you've done historically?
Primarily now with Atlas and obviously database activity monitoring and everything else that we have, it's about educating the customers on what we have and explaining to them how they can get immediate value and ongoing value. We follow the same playbook: evaluate, show success, build an operational plan and then win the business and make sure the customer is successful. AI security and data security essentially are the same problem and very synergetic to everything we are doing. Also, you're starting to have this new technological stack where controls are moving to the data and AI level, and these are two places where Varonis plays very well.
The next question is from Shaul Eyal from TD Cowen.
Guy or Yaki, from a geographic perspective, was performance balanced around the three major theaters? And also on seven-digit transactions, would you characterize performance this quarter as equal to what you have seen over the course of the past few quarters, specifically on seven-digit transactions?
I'll start with the second part. We've talked a lot about our desire and philosophy of going upmarket, and we have definitely seen that work well for us. We're able to get larger deals as we land and show them value, give them protection with the MDDR, and then go back and do healthy upsells that continue to generate customer lifetime value and increase ARR. So yes, we have seen that occur, and it's part of our focus on going upmarket. In terms of geographies, this quarter was definitely led by North America with healthy numbers coming from there; that was the primary driver for the quarter.
The next question is from Joshua Tilton from Wolfe Research.
Maybe a quick one for me. I know everybody has kind of asked about the deals already, but I have more of a hypothetical one. I'm listening to everything that Yaki is talking about in the beginning. You guys are using lines like AI security and data security are one and the same. It feels like demand for the Varonis platform will never be stronger than it is today given all the secular tailwinds around the AI story. So my question at a very high-level point is: why are we seeing more? Where could we see more? Why are we hearing more about momentum in newer products that don't feel super related to data and AI security, and we're not hearing more along the lines of inflection in demand for the core products? Or is it a line out the door for Varonis because of this tailwind? Does that make sense? It just feels like now more than ever, you guys should be the #1 ticket item. So is it like a capacity thing? Is it a timing thing? How do we think about maybe seeing all the good secular tailwinds we're talking about show up just a little bit more from a growth perspective, if that makes sense?
So first, remember that we have been selling Atlas only for 3.5 months. In general, the demand is primarily related to the adoption of AI in organizations. Customers spend a lot on AI and now are trying to realize gains and they're understanding the risks. They need to be more mature and connect many more data stores to realize value. We believe that over time we will see more demand and it will be the foundation for organizations to realize value from AI. The reality is we see that when we sell the AI suite and the data suite, they almost always go together.
The next question is from Mike Cikos from Needham & Company.
I had a question for Yaki on the new product launches. We're talking about database, Atlas or Interceptor. Can you help us think about to what degree these new products are benefiting the model today? Our checks have indicated, for example, that Atlas is still in the process of education or technology enablement with partners. So the question is really: to what degree are the new products benefiting the model today? Is that quantifiable? Or are we still largely building the pipeline and these positive catalysts are on the come?
When we track the momentum in the pipeline generated by the new products, we're definitely seeing an increase in that pipeline, and Atlas is coming up in conversations with customers very frequently. Keep in mind the acquisition happened in the middle of Q1 and the expectation is that its contribution will be more meaningful in the second part of the year. In Q2 performance, we were pleasantly surprised with the contribution coming from Atlas already, with the expectation that it can do even better and have a more meaningful contribution in H2.
When you look at the offering itself, it's a natural extension to the Varonis platform and the way it's coming up in conversations gives us confidence we'll see better contribution in the second half. Don't forget we sell everything through a POC, and the preliminary indicators of a new product combined with the platform are very strong. The pipeline is very strong, and we converted enough Atlas deals to have confidence that it will do very well.
The next question is from Rudy Kessinger from D.A. Davidson.
Question for you, Guy. If we look at the SaaS net ARR ex conversion guide, the implied Q4 number is a really big step-up over Q3. On a year-over-year basis, it requires about 40% year-over-year growth in SaaS net ARR ex conversion. Is that more a factor of Q3 being a more conservative bar given you're not expecting all of those slip deals to close in Q3 in that guidance number? Or what else is informing such a significant ramp in the Q4 net new ARR versus Q3?
First, if you look at Q4 of last year, a large portion of the dollar amount was related to conversions. We had roughly $63 million of converted dollar amounts in Q4 of last year, and that required reps to focus on paperwork, security checklists, procurement and legal. Now they don't have to focus on that. Keep in mind the way commissions are structured: reps don't get quota credit in 2026 on conversions. So they are focused on new business and existing SaaS customers—those sales go towards their quota and accelerators. By having them not focus on conversions and focus on what they do best, I don't see Q4 as a stretch. Our guidance philosophy has been conservative and I feel good with the setup for H2. If you take away that $63 million of conversions, the uplift you're seeing in Q4 is not as significant as it seems.
The next question is from Richard Poland from Wells Fargo.
I noticed you mentioned GovRamp a few weeks ago. On top of already having FedRAMP, given past commentary around the government business and what's assumed around the conversion there, as we enter the big federal Q3, is there anything we should keep in mind for the back half? Any change or different perspective on how much of that could convert?
When you look at the federal business, in previous years it wasn't an area we did particularly well in. In our guidance and focus for Q3, we basically didn't assume any upside coming from that vertical. If things go well we'd be pleasantly surprised, and that would be on top of what we guide. But the expectation is to focus on the enterprise business. We know a good portion of customers who will not convert are state and federal, and we already baked that into our metrics for free cash flow and guidance. Even the conversion assumptions take that into account. So going into Q3, we derisked that component.
The next question is from Jonathan Ruykhaver from Cantor Fitzgerald.
Given the convergence we're seeing between identity and data security strategies, I'd love to hear your thoughts on where you see the boundary between Varonis and those identity vendors. And then just in terms of how you're doing on identity solutions, any color on adoption would be helpful as well.
We integrate a lot with identity and are doing identity behavior analytics to understand normal behavior and posture. We are not managing passwords and provisioning in the same way as identity vendors. Once identity reaches access to the actual data resource, that's where we play heavily, and it works very well for us. It's extremely relevant to AI security: understanding the identity of the agent and what it is doing is part of the overall value proposition and everything we are doing with MDDR and our UBA solutions.
The next question is from Todd Weller from Stephens Inc.
Just wanted to follow up on the topic of sales productivity. Where do you think you are in realizing the productivity benefits? And then how do we think about that from a tailwind perspective in terms of what inning we are in there?
When we look at productivity levels of our sales force, we have seen those levels increase compared to 2025 and 2024, particularly among mature reps—those with over a year of tenure. The reasons for that productivity increase include having a larger platform to sell and going upmarket. The new platforms we've acquired—Atlas, Interceptor and DAM—represent significant markets to go after, and those haven't fully kicked in yet. Our expectation is that with the additional pipeline build and the ability to convert that pipeline, productivity levels can continue to increase in the near and medium term. We're also focusing hiring on more tenured account managers to further help productivity. We're making the right investments to take advantage of this market opportunity.
The next question is from Joseph Gallo from Jefferies.
Any update you could give on the Microsoft partnership? They've been making announcements related to AI security. What momentum are you seeing there and do you expect that to further ramp?
The Microsoft ecosystem works well for us because a lot of MDDR and AI security use cases intersect with Microsoft environments. That said, the overall Microsoft environment is becoming a smaller portion of what we protect because organizations have many data stores—Salesforce, ServiceNow, Snowflake, Databricks and more. It's going well, but Microsoft is just one part of a broader ecosystem.
The next question is from Fatima Boolani from Citi.
Guy, with about a year or four quarters worth of upsell data points on Interceptor and DAM, can you put a little more quantitative context on what type of realized upsell levels you're seeing on a base core product deal for some of these newer solutions? Recognize Atlas is still early. So maybe just focusing on DAM and Interceptor because they've been in the market for about a year.
Interceptor was closed late in Q3 last year, so we've had Q4 and the first part of this year to get it as part of the product and sell it in an integrated way. Interceptor is mostly being adopted by existing customers—going to the base and connecting it with MDDR, which is a strong value proposition. DAM is an opportunity for both new customers and existing customers; it's part of addressing a large market opportunity. The mechanics are encouraging and we've seen it help us in Q2, but it's not near its full potential yet. We expect more contribution in H2 and in the years ahead. Add Atlas on top of those offerings and you have a really strong platform.
The next question is from Erik Suppiger from B. Riley Securities.
Several vendors have introduced products that have some overlap with Atlas over the first half of this year. Can you discuss the competitive dynamics for that product? And if you have any type of win rate, can you share any of those metrics?
Atlas is the most comprehensive when you look at the whole lifecycle of AI in terms of posture, compliance, penetration testing and so forth. It integrates exceptionally well with data. The founder of Atlas understands data very well—he pioneered database activity monitoring and built that experience into the product. When we bring the platform together on the AI side—covering the whole lifecycle, taking it to production, monitoring, automation and the ability to elegantly support AI systems and connect pipelines to data—this is where we feel we are uncontested.
The next question is from Shrenik Kothari from Baird.
On gross margins, sequentially they're down about 20 basis points and about 290 basis points from last year, as operating margin seems to be improving with the SaaS mix growing. How much of the gross margin trends are reflective of the pricing environment versus acquisition integration? Is this dynamic arising from new logos being stronger recently or new products structurally carrying lower gross margins as the mix shifts toward broader AI runtime security?
None of the above. The gross margins are tracking according to our plan, and I'd even say they're better than what we initially planned when we introduced the SaaS offering. The investments we made initially brought gross margins down, but they are expected to tick up as we complete the transition, and compared to internal plans they're doing better. We're not seeing pricing pressure or the other factors you mentioned driving margin changes.
The next question is from Junaid Siddiqui from Truist Securities.
Historically, you've benefited from data security budgets, but increasingly you're talking about AI security and AI transformation initiatives. Are you seeing deals funded by entirely new AI budgets? Or are customers largely reallocating spend away from existing security vendors and point products?
It depends by product. For DAM, we're taking deals from Imperva and Guardium. Interceptor can replace some layers of email security focused on social engineering. In data, we are dominant, but where customers have DSPM or similar solutions, we can take budget there too. We also see organizations allocating budget from digital transformation and AI budgets; there is a security component for this new technology and companies are increasingly allocating spend from that bucket.
There are no further questions at this time. I would like to turn the floor back over to Tim Perz for closing comments.
Thanks for the interest in Varonis. We look forward to seeing everyone at conferences this quarter. Please reach out if you'd like to call back. Goodbye.
This concludes today's teleconference. You may disconnect your lines at this time. Thank you for your participation.