Welcome, everyone. That lovely music just to sort of soothe everyone into our session today, so I'm Howard Wilson. I'm the CFO of PagerDuty. Good to see some familiar faces here in the room today. I met a few of you when we were out on the road, and I've got to know some of you by voice, through the course of investor calls and analyst calls. So thank you very much for being here today.
We also have a number of folks who are joining us via webcast, so appreciate you being here to share in this investor-analyst discussion session. So this of course is a very exciting week for us as PagerDuty. This week we hold our Fourth Annual PagerDuty Summit event, which is happening in this hotel. We're getting together over 1,000 enthusiastic participants. Well, that's what I'm sure they will be who are here to really chat about the world that we live in of Digital Operations Management, real-time work. We have both practitioners, developers, engineers, customer support agents, as well as managers, leaders, thought leaders who are all getting together for what promises to be two very exciting days, as we explore this topic, this journey to real-time operations.
So certainly gonna be a good few days, and I know that some of the folks that are here in San Francisco will be joining us, so we look forward to having you here. I did want to just remind you, pause for a moment, to remind you of our safe harbor statement. Our VP of Investor Relations, Stacey, has assured me that I do not need to read this to you. I just wanted to make sure that you were reminded of this. So with me today, we have Stacey, who is our VP of Investor Relations. We also have Jennifer Tejada, who you will hear a lot from today, our Chief Executive Officer, Jonathan Rende, who is SVP of Product, and Tim Armandpour, who's SVP of Engineering.
So between this group, we'll try and create some variety, and you'll hear us all telling different parts of the PagerDuty story. Just to run through the agenda, Jennifer's gonna kick us off talking a little bit about the opportunity that we see out of this problem space together with the way that we go to market. We'll spend a little bit of time talking about our customers and sharing a few stories with you about our customer base. We'd like you also to get a glimpse into the product. So Jonathan Rende will take you through a demo of a portion of how a customer could use PagerDuty. We'll have a short break, and then Tim will talk through some of the advantages that we have from a technology perspective that sets us apart from our competition.
We'll also take you a little bit through the journey, the evolution of PagerDuty, a little bit about where we've come from, but more importantly, where we're going and what's important for us from a product perspective. And then we'll close out the discussion with the financial piece and a Q&A with myself and Jennifer. So I'd like to invite Jennifer Tejada to kick us off to chat about the opportunity.
Thanks, Howard. Hi, everybody. Jeez, we got the after-lunch. Wow, this is gonna be a challenge. Howard mentioned this is our fourth Summit, and our first Summit was like on day 12 or something for me with the company, and it was down the road on Market Street in like one of those purpose-built event venues, meaning like an office space that you really can't do anything else good in. So let's just have events.
It was a bunch of dudes in hoodies and me, and it's really amazing to see how things have changed. We had about, I think, 200-300 people at that event. It was free. It was about three-quarters of a day long, including happy hour, I think. Here we are, kind of three and a half years later, with over 1,000 people coming. I think we were over 1,200 people, actually, that'll be here this week.
And the one thing I'd like to change is to have the analyst discussion after my keynote, not before my keynote. So I hope you'll all save for tomorrow because there we have a number of, I think, interesting announcements and exciting news coming out. I wanted to spend a little time talking about the opportunity that we see as PagerDuty's leadership and, you know, why we're inspired and compelled to come to work every day. In the last three years, I've seen several hundred of our customers. I've spent a lot of time getting out on the road and talking to both practitioners but also leaders. For those of you who aren't familiar with PagerDuty, we often start in the developer community, championed by the CTO, and sort of virally work our way into IT, IT Ops, security, customer support, etc.
And just last week, I got a message from a Fortune 500 CIO that I had met a couple of years ago and had later invited to do a fireside with us at a regional Connect event. And she reached out to say, "Hey, I just took a new job at another Fortune 500 company, and I got my new leadership team in place, and I think I'm ready for PagerDuty. I need the experts on your team to come help me drive." And she said two things: digital transformation and DevOps adoption in the same sense. And that really kinda underpins how PagerDuty has grown over the years. We engage with teams.
We engage with leadership. We have a product that is trusted and used regularly by our users. And increasingly, we build strong relationships with more senior executives as they're trying to transform our organization. The role of the CIO has changed as well. The CIO used to be the program manager role in the executive team, and the place where you managed a lot of technology but also people change management. And increasingly, we see the CIO being a much more technical and technically capable person and a lot of CTOs who have actually developed and, you know, written code and carried the pager, sitting in that CIO role.
And in this particular case, this CIO was a developer, and actually, you know, experienced the product herself as a user. And this company that we're talking about now is the third company that she'll be bringing PagerDuty into. So, let's. Oops, I just turned off the slide thing. That's the button I was not supposed to push. Now, let's talk a little bit about digital transformation.
So every company is becoming a software company. I think that's not disputable. What I would say is they're all relying on technology to transform, and that technology is becoming increasingly complicated, and the proliferation of data, and systems that humans contend with on a daily basis is outside of their control and impossible for humans to manage. Consumer demands continue to change as well, and we know that a customer will give you anywhere from one to three seconds to deliver the perfect end experience, or they will abandon your app. They will find something else to do. And when I was on the road, I shared that I only need to stand outside the front door of PagerDuty's headquarters here in San Francisco to watch people toggling back and forth between food delivery apps and rideshare apps.
And I mean, the patience is very limited, and that perfect opportunity is increasingly more elusive in terms of delivering it. Cloud adoption is very, very powerful. Whether you're adopting cloud for the first time or you're migrating parts of your environment to the cloud, it can be super powerful and super efficient, but it's also very hard and comes with a lot of risk. And so developers and IT employees in their own right wanna make sure that as they move a single workload to the cloud, then they'll earn the right to move the second workload to the cloud by having it all go very well. And as a result, we've developed very strong partnerships with a number of the cloud providers in the market. DevOps is now mainstream and top of mind, not only within the technical community in the businesses we serve but in the C-suite.
We hear executives now talk about how important DevOps is, how important ensuring that you have the right tools and solutions and platforms for your developers are, as a way of keeping them with you, as a way of attracting the right people into your organization. Developers need to be armed with the right technology, the right information, not all the data but the most insightful information that they need to do their jobs well. We still see increasing growth in as a percentage of the headcount in our customers of technology-centric individuals like devs or IT workers. That continues to be a tailwind for PagerDuty as well. When I talk a little bit about the complexity that our users face, I wanted to use a bit of an example. Think of this as a large e-commerce provider, and you know, it's Black Friday is coming up.
You may do a little holiday shopping ahead of the season to try and get on top of things. If you're like me, you'll do most of that online. I mean, the closest I've been to a store in the last three to five years is here. And what's running behind the simple transaction of trying to buy a new electronic device for your child or new jewelry for your wife, etc. There are a number of applications that support the shopping cart environment, should support order management and logistics, support dynamic pricing, etc. And within those different applications, many, many things can go wrong at any given time. Underpinning all of that is infrastructure, is the networking environment, app monitoring, cloud computing, security databases, etc.
And underpinning all of that are a number of diverse, distributed, and cross-functional teams that are responsible for different pieces of this journey. And remember, I mentioned that that consumer, you know, you in this case, is gonna give somebody three to five seconds to make sure that that experience is perfect or move on. So PagerDuty has invested a lot of our time and effort in ensuring that the orchestration of unplanned work across these distributed teams, which you may, you know, if you're a developer working in an organization responsible for the shopping cart application, you may not even know of the person who's responsible for security or the person who's responsible for supply chain and logistics or the person who owns a pricing application. They're not in the same town. They're not in the same building, and figuring out their name is hard. So orchestrating the work across these different teams has become a very big challenge that isn't solved by monitoring.
It isn't solved by ticketing queues, etc. That's orchestration. And like I said, the stakes are getting higher, and the time that you have to be effective is getting shorter, and bad experiences go vir al. We did a study this year that I'll be talking about tomorrow in my keynote where we learned that the majority of companies find out about bad customer experiences from their customers via Twitter, via social media. They aren't calling customer support politely anymore and telling you what's going on. They're raging online about it.
So we find that every company is talking about digital transformation. It's a top three CEO initiative, and it's imperative that these organizations can deliver an always-on, real-time experience. In order to do so, they need visibility across these increasingly complex and far-reaching technology stacks. If you think about the shift that many organizations have made from managing monoliths to managing more distributed architecture, leveraging containers, all that Kubernetes and Docker, or even moving to serverless, that complexity's increasing. It's not getting easier, even if you're abstracting some of your legacy environment away from some of your customer-facing applications. And again, we need to orchestrate people and teams using intelligent automation and learning.
We're not talking about just taking a phone tree and sticking it into an automated form and running an escalation. We're talking about using insight and predictive analytics to get the right challenge to the right person in the right organization and exactly the right moment. And we know that slow is the new down. You may not hear about a lot of outages, which you'll just see as the spinning wheel of death, which really makes people crazy. So this need to orchestrate teams effectively in an automated way to learn every time you run an incident or issue and get better each time is becoming really important. And we think of digital transformation as a team sport and, frankly, real-time ops as an extreme sport because you're under pressure.
You're under a time crunch. And the challenge that our community is facing is they can't always see the field. They're not even, you know, when an incident happens, you don't actually have all the context of what's going on. You might have one piece of the puzzle. So you don't even know what sport you're playing. You can't see the field. You don't have field vision, and you don't know who the players are available to you on the sidelines. So PagerDuty brings all of that together.
And, I mean, that means bringing the complexity of people, systems, services, the technology environment, and the customer together in a meaningful way. And that flashy piece of art is just a taste of what you'll see tomorrow. So our mission is really to connect teams to the real-time opportunity that exists and elevate them to the outcomes that matter. When a team is working on unplanned work, when they're dealing with an incident under time pressure, when they know that incident is impacting their business or impacting their end consumer's experience, they're not doing the innovative work that really matters to the company. They're not doing the purposeful work that's meaningful to them. So PagerDuty sits at the center of machine data by consolidating and integrating to over 150, 300, sorry, over 350 of the most popular applications on the planet.
But we bring that together with over 10 years of human response data, where we understand how people behave under certain types of pressure, who's responsible for which service, what their specialty is, etc. So we're bringing together not just visibility, you know, what's happening, not just consolidating and correlating monitoring signals, but then orchestrating those signals, which we turn into insights to the right teams and using machine learning to help those teams triage and respond faster. And an example of that would be recognizing that an event as a number of events that are storming right now are actually likely to become a major incident, one like you've seen in the past. You can only do that if you have a significant amount of historical data about a particular organization, a particular set of services, but also can benchmark that against a much broader data set.
And then we're automating more and more of the process, like providing suggestions around potential triage actions to date, providing suggestions around the types of people that you need or a specific person you need in the call, and then providing analytics, helping a customer understand how their different teams are performing vis-à-vis one another. So if you have 50 teams on PagerDuty, why are 30 getting significant benefits and the other 20 are not getting as significant benefits? Is it because of the services that they manage? What is the health of those services? And how does the health of those services compare to other peers we could benchmark them against?
So really using analytics to now look back and help our teams understand how to become more productive, more efficient, and even predict burnout, predict involuntary attrition, customers that might or voluntary attrition, customers users that might decide to leave because they are burned out from managing too many incidents. We do this by sitting. Hi there. Good to see you. We do this by sitting at the center of today's modern digital ecosystem. And I think this is an important slide because I know all of you are not necessarily infrastructure specialists. You may not even be technology specialists. You may be generalists following lots of different sectors and categories. So I wanna spend a little bit on this.
The reason that we have been proactive about, one, thinking of integrations as product, and two, investing in our integrations heavily, both in building deep bidirectional integrations but also making sure we cover the platforms, the systems, the tools, the applications, the security environments that our customers care about, is to help, one, bring a neutral, independent view of what's actually happening in the market and, two, start to correlate and consolidate some of these things so you don't have teams working in silos. On this slide, we've got some of the most popular apps and services on the planet, the largest monitoring providers that you can think of, the social environment where customers will share their unhappiness with your product, network and security. Increasingly, we're seeing a lot of our customers use this in IT use cases.
And what I point out is that the majority of logos on this slide are also PagerDuty customers. There's not a lot of competitive overlap here. This is a very complementary framework. And I know that's hard because a lot of companies are using the same language to describe very different technologies used for applied to very different problems. So I just wanted to make that super clear. We've continued to invest in our APIs, and tomorrow you'll hear us talking more about our developer ecosystem and our developer platform, which is going to allow the developer community to also build more readily on the platform. So how this all works, just to kinda give you an idea, we use those integrations to capture data coming in from the modern technology ecosystem, in milliseconds and seconds.
We co-consolidate and correlate that data using machine learning and start to recognize what's an event storm or a set of singular events versus what is an incident where a number of events storm simultaneously to become a business-impacting issue. We then engage the right teams, the right set of skills, and hopefully that's a less-is-more problem, so instead of getting 100 people in a live conf call, which will often happen if you just run a bunch of escalations, we're getting the right one- or two-person team of folks in a call because they are the right skill set across cross-functional teams to triage and address something very quickly. We're then capturing their actions. We're capturing who those responders are, what they do, how long they take, which actions they take, etc., which allows us to analyze that incident response and learn from it.
It also allows us to automatically build the postmortem and the actions coming out of that process, which can be propagated in any of the most popular ticketing environments that you can think of. And just to give you a sense of how this is different than how maybe the APM providers might send an alert to a team, going back to that Black Friday example, what you see is that spinner. You know, you're trying to check out and just get something sent to your house today or tomorrow 'cause you don't like to wait, and something's not going wrong. In that moment, the way that will present itself to a number of teams across a company is as a set of separate, disparate, simultaneous events.
So, for instance, your APM provider, Dynatrace, New Relic, etc., will be providing events to the apps team, the development team, that something's going on in the app. Datadog might be providing an event to the IT team that an app is, that the network is slow. Zendesk is gonna send alerts to customer support folks, customer support agents. That was the word I was looking for, that there's an increase in inbound tickets and calls, and Amazon CloudWatch may be sending an alert to the sysadmin team to let them know that there's a problem with server capacity or virtual server capacity. All those teams without PagerDuty will go run five siloed separate responses, right, so multiply that times five or 10 or however many people are on those escalation lists.
And now you have a bunch of people chasing what they think are different tails across distributed teams, you know, regionally dispersed, when actually there should be a team of six to eight people working on one incident. And that capture, that consolidation, that correlation, getting the right team ramped and up and running, that's a seconds and minutes problem, not a days, weeks, hours problem. You then will run a response, and we know that our customers are getting better and better at running a response. We did a little analysis that I'm gonna share tomorrow that over 10 years, mean time to acknowledge has gone down from hours to less than a minute on average for our customers and platforms. So that's just getting somebody to say, "Yeah, I got this," right, which is a really important thing.
For those of you who don't know much about DevOps culture, it's all about ownership. So the best, most progressive, technology-centric companies, their developers have full 100% responsibility for their code in production, meaning instead of shipping some code on a Friday night at 6:00 P.M. and going to the bar and throwing it over the wall for an ops team to manage, as soon as that code goes into production, they're first on call, and they're the subject matter experts should anything go down. What that does is it intrinsically motivates developers to build better code and not to do stupid things like ship it at 6:00 P.M. on a Friday or 4:00 A.M. in the moring on a Saturday, right? It has them test and automate their testing more effectively.
So driving DevOps and getting to the right person at the right time instead of getting 100 people in a call is very well aligned to the empowerment and accountability associated with DevOps best practice and, in fact, Agile. So you're gonna hear a lot over the next two days, and I, you know, I encourage all of you to stay tomorrow for our main event. We've got some really incredible speakers. I'm very excited about. I'm just as excited about Jeff Lawson and Eric Yuan coming as I am about Andre coming. I'm probably most excited about Chris Bertish, who crossed the Atlantic by himself on a paddleboard, the pretty crazy dude, talking about, you know, what responding to unplanned things in real time and really perseverance and talk about accountability and ownership, staying alive in the Atlantic over, I think, it was 96, 97 days.
Pretty tough gig. But we're here to really talk to our customers about product and also ensure that they have the opportunity to connect with the brightest minds and hear about some of the best practices in the industry because so much about this week is about our customers sharing their stories. And I think what you'll find is this is a community that's very open and very transparent and excited to talk about what they're doing. So many of you have seen this before, but you know, we built our platform around our first product, which was On-Call Management. And what that's allowed us to do is collect data on signals, people, events, workflows, and metrics over the last 10 years that we've then been able to leverage m achine learning and build new products on top of.
Event Intelligence was our first foray into AIOps. It was also, it's also where our first offering that was underpinned by m achine learning kind of lives. Event Intelligence is really driving a lot of automation and a lot of. We have a lot of customers who refer to Event Intelligence like being a virtual, all-seeing, all-knowing member of the team because it actually captures a lot of this information, starts making suggestions, helps you get to better decisions faster. Modern Incident Response is about going from beyond On-Call to truly running Automated Incident Response in your organization, getting to not just acknowledgment but resolution faster. Analytics, which we announced last year at Summit, GA'd earlier this year, and it really helps you look back historically on what worked, what didn't, and why, what's the cost of some services and teams versus others.
And then Visibility helps you understand the impact that technology issues are having across the business and bring context for teams about the priorities of what they should or should not be working on in any given time. And, you know, just to give you an example, I had a customer who was going through a big SAP update and, you know, on a mission-critical timeline to get that rolled out but had a direct customer-impacting incident that was driving revenue loss of about $250,000 a minute to the organization. And without Visibility, their developers would have still been working on that SAP rollout instead of working on, you know, the issue that mattered the most.
So PagerDuty's value proposition, I don't think we've talked enough about this in the past, but we've started to really measure the value, and we're trying to get better at, and we can always get better, but we're trying to get better at helping our customers understand the value and the return on the investment that they're making, not just in improving their DevOps culture or improving On-Call or increasing the efficiency of their incident response, but the overall impact to the business. So, we commissioned a study this year, with a number of our enterprise customers, and what we found, and IDC did the study for us, what they shared with us was that they're seeing on average a 731% return on investment over three years and a four-month payback on the investment that they make.
And that's not taking into account either the protected revenue, which averages about $1.2 billion a year that they see as a benefit from using PagerDuty. So I wanna talk a little bit about a couple of use cases here. One is, you know, the job number one is to make sure that customers are happy, consumers are happy, and they're completing their experiences and using us to make sure that payment transactions get taken care of, or it's a security team dealing with inbound requests from law enforcement, or it's SightLife who uses us for corneal transplants to orchestrate the process of getting a cornea from a donor to a recipient. These new use cases are not separate products that we designed for. They're the application of a real-time operations workflow to real-time problems in a technically enabled environment.
When you think about the TAM, what I'd point to is we think this TAM gets bigger as you start to add event management and AIOps, as you start to leverage predictive analytics more effectively, as we leverage benchmarking and real-time visibility, but also as you see new use cases leveraged in a more ubiquitous way. One of our development efforts underway is really to focus on making the user experience easier and simpler, even as the back end of the product gets more complicated and enriched with many new feature sets. Am I doing okay on time, Howard?
Another minute on this, and then you can move on to the next one.
Okay. You know, other solutions in the market do not have the capabilities for real-time operations because they usually solve one piece of the puzzle.
I'd start by saying our number one competitor is either nothing or homegrown solutions. You know, many, many companies, the majority of companies out there have not even gotten their engineers on call. So there's a huge opportunity just in the first step. Most organizations are very manual. If you've ever heard of a network operations center, that's a bunch of folks in three shifts sitting usually in something that looks like a concrete bunker with a lot of old flat screens up in front of them, looking at speeds and feeds, trying to figure out where there's a problem, and then double, triple, quadruple handle those issues to manually escalate them and figure out who to call.
And figuring out who to call is a really hard challenge because Active Directory doesn't tell you who's responsible for the Java code in this app or who owns the secure firewall over here or who's available right now when it's 3:00 A.M. on a Sunday. So we're primarily greenfield when we go into an account. There are other alerting and notification tools out there that are narrow in functionality. They essentially take a dumb process and automate that dumb process. So putting people in a static phone tree and then automating that phone tree so that something just escalates, in order, it does not add a ton of value. They also lack data. So none of the other alerting and notification solutions in the market even had a data retention policy on their website until, until we started talking about it.
We have been collecting data in a very high integrity and secure way for the last 10 years. So we have a very rich data set that we've been able to build on that doesn't exist anywhere else in the market. And then I wanna talk about reliability and scalability because it's one thing to say you're reliable and scalable. It's another thing to prove it by having more than half of the Fortune 100 and over a third of the Fortune 500 use your product for many, many years as they've grown and scaled. And we're talking about some of the most progressive, technology disruptors that you use in your homes and at work and also some of the oldest, most highly valued brands in the world.
You will not hear people complaining about PagerDuty's resiliency because we are proven at scale, and we continue to scale ahead of our customers' needs. It's a hard thing to get right, and we put our money where our mouth is by having not only one SLA for availability but also an SLA for performance, so most of the SLAs you'll see in SaaS and in the industry are, "Is my website up?" Right? We are ensuring that there is actually data and events transaction across our platform, and then there are still some customers that wanna try and use ticketing for Incident Management. The problem is ticketing is an hours and days and weeks thing. Like you put something in a queue, and you wait till somebody else picks it up.
And I mean, how many of you have raised a ticket and waited for something delightful to happen? It doesn't happen simultaneously. It happens over time. That's a command-and-control process that wasn't built for a real-time world. So we don't really see ticketing solutions as competition, but we get asked about them a lot, so I thought I'd mention it here. You know, the other solutions out in the market are not focused on correlation and orchestration, and I put those two things together for a reason. There are some companies that do one but don't do the other. There are some companies that do monitoring but don't do the correlation, etc., and we do all of it.
And that's the reason why most of the players in the technology ecosystem also use us, whether it's Datadog or New Relic, or you know, I could just go on and on. Tomorrow, I'm gonna mention in my keynote a number of customers that have gone public, and it won't surprise you to note that the vast majority of the companies that have gone public in the last several years all use PagerDuty because they wanna make sure they're putting their best foot forward in terms of experience in the market. And we're very proud of that, and we're proud of the fact that they trust us with their reputation and their customer experience.
That's not to say that we don't think account engagement is very important, and, you know, that story I shared with you about the CEO that's moving, CIO that's moving from one Fortune 500 company to another is very common. We have a lot of developers that take us with them from one company to the next. We have developers that require PagerDuty as part of their work, good working conditions along with other important platforms that they think meaningfully help them get their work done. But I think it's just important to mention that we think this is a great combination of both a great product. Our growth is product-led, low friction. It's very easy to use and can be self-served, but also compelling account management. So when you get to tens of thousands of users, you're spending more than $1 million with us.
There's a person that understands your business that is gonna be responsible for your account. Besides me, I'm gonna try and understand your business, but I'm not all over every single one of them. So just to prove a point on how easy this actually is, I'm gonna show you that even a CFO can sign up for PagerDuty. All right. So do you wanna click? You wanna get you ready?
Yeah. So this is probably hard to believe, but I thought, well, okay. You know, I'm always telling people this is really easy. So is it really that easy? So here I go. This is Mina. I'm gonna try now. This is my trial. Needs my first name, needs my last name, needs my email address. So pretty straightforward at this point in time. No credit card required, right?
So some basic details coming in there, and I'm about to get started. What's interesting, I've got my welcome. The first thing I'm gonna do now is I'm going to begin my setup, and it asked me to say, "Well, what kind of service is this?" Well, I'm a CFO. I'm interested in payment processing. So the first service I've decided I want to monitor is really all about payment processing, and I wanna use, I use New Relic today. So I select New Relic. I put in a phone number. I use someone else's, not mine, just so that I could check. Create my first incident, which is now going to test that this is actually working. And here I am in PagerDuty. I've got an example incident to have a look at. I can see that, yes, this has actually worked. I'm able to then acknowledge it.
I'm able to then resolve it so that I can actually clear it out of my path. And I'm up and running now. I have New Relic involved, configured, but I also happen to use a couple of other services that I'd like to add. So let me put in, another integration, and I'll use AWS CloudWatch 'cause that's one of the integrations that I use. I put that in, I select it, and I add the integration, and it's now generating the unique keys that I'll need to just go and cut and paste into.
For a developer to bring more people into the platform, because the more people in the platform will let you get calls. It's simple math, but it's important math. And then what often happens is you'll see one development team champion PagerDuty for the rest of development, and it moves towards a standard within all of Dev, and then it moves across generally to ops, to IT Ops, to security and ChatOps, and on and on to customer support, etc., for a particular organization. Oftentimes we see a bottoms-up viral motion meet a top-down initiative, and that top-down initiative could be cloud migration, could be a team changing their behaviors coming off the back of an incident that wasn't handled well, etc. Back to our go-to-market, I just wanted to talk about some of the things that we do to ensure that our customers succeed, and this is very important to us.
So Alex Solomon, our Co-founder, leads our Community Forum. So he's responsible for everything from community evangelists to making sure that that community has a way to engage together, and that they can share best practices, etc. So we still are sponsoring a lot of DevOpsD ays. We are all over the place, talking about how to bring best practices to light, including open sourcing some of our own best practices. So we've open sourced our incident response documentation to, so to help people improve more readily. It's at response.pagerduty.com, and we've recently open sourced something else, like postmortems. We recently open sourced our postmortem practices as well. So driving a lot of thought leadership, but also doing it in a very transparent way, in service to the community.
Over the last couple of years, we've built an Expert Services team 'cause what we found was even though PagerDuty is very easy to implement, human change is not, and neither is DevOps ownership. Sometimes people need help with custom integrations. In a lot, a lot of our enterprise customers, we're frequently asked to integrate to something that's legacy or homegrown that we can't build a standard integration for. But important to recognize it's not a big portion of our revenue. 99% of our revenue's still coming from license, and then PagerDuty University. So what we found, and this is really response to demand from our customers, we found customers were constantly asking us things like, "How do you train an incident commander? How do you improve the way you triage? How do you set up your services most effectively? What's the best practice for postmortem?
What's the best practice for driving from postmortem actions into CI/CD? How do we do that?" So we launched PagerDuty University last year at Summit, and it's actually going on right now somewhere in this very modern-ish building. And then finally, award-winning support. So, one of the PagerDuty use cases that I engage with frequently is I get paged every time one of our top customers raises a ticket or reaches out to support. So I can see in the moment the transcript and the conversation that they're having with our customer support people. That's super important to me because I want to see the tone and the voice that our team uses when they're talking to customers. And equally, I want to understand the questions that we're asking, whether we're really solving the whole problem or just the first part of the problem.
And I wanna make sure that our customers are in those conversations, which are often chat, not phone anymore, in a delighted way. And when they don't, I can reach directly in through my mobile app, and I've demoed this for some people on the road, and I can email that individual while they're on the call with our support person, which freaks them out and is a very entertaining way for me to break up my day. But I end up having a one-to-one engagement with a customer in that moment. So this is really important to us. We're continuing to move toward things like premium support services, which was a request we got from a lot of enterprise customers, dedicated customer success managers, not instead of our account management, but in addition to our account management, to make sure the whole journey is well-supported and that our customers see their success as our success.
We have a very efficient go-to-market model that drives this reality and growth where we have a high-velocity inside sales team. So these are people that are in hubs in places like San Francisco, Toronto, Atlanta. They're primarily working via phone, chat, email, etc., and they're engaged with our customers in SMB and mid-market. Some of them will go see a customer from time to time, but the vast majority of our business can get done remotely with an inside salesperson. And oftentimes, what they're working on is finding the next new team, finding the next new opportunity, etc., because even customers served within this inside sales model can add users or add product on their own without the help of a team. Probably important to mention too that most of our salespeople are very technically capable.
They don't require an SE, although they are provided with solutions consultants. They are able to go toe-to-toe with a developer, with a user, with a CTO, with just about anybody else in the middle. And they're all certified to be able to walk through our product to leverage data with the customer to help them understand where to find value from the investments they've already made, as well as think about what could be next for them. Then our field-based sales organization is also high-velocity. They do quite a few deals over the quarter as opposed to a single deal here or there. They tend to find new opportunities, let them grow, come back to them to expand them over time, and they focus on the enterprise part of our business, which represents more than half of our revenue today and continues to grow in its mix.
So we're really happy with how our Enterprise business has grown. And we talked a little bit about this in our last earnings call, where the number of customers spending over $100,000 with us is up to 274 and continues to grow at a nice tick with, I think, a 51% growth rate, year- over- year this past quarter. So, you know, just to kinda give you a visual of how this works, this is an example of a large global software company that's been with us for several years. You know, they started out with a single developer team, added additional developer teams, grew over time to about 1,500 people, and then they started investing in the cloud, and that cloud expansion drove additional new users on the platform. So over a year or two, we went from kinda 800 to 4,000 users.
A new CIO came into the business with mantra to drive Agile and DevOps throughout the organization. W e're seeing this everywhere in large enterprise, and that top-down initiative led to an inflection point in growth where we saw ourselves go from a handful of teams in development to teams across development, IT, and then more recently, customer support. So this just gives you an example of how we grow in an organization. It doesn't happen in three POs. It happens in 20 POs over the course of a couple of years. It's a patient way to grow, but it's a very sticky way to grow as well because the customers internally are championing on your behalf. They're helping drive that viral adoption. And, you know, if it doesn't work, you hear about it.
So that scalability and resiliency becomes really important when you start looking at the growth opportunities inside of these customers. So we're gonna switch gears and talk a little bit more about our customers. Next year when I do this, I think we'll do Analyst Day after the keynote. So 'cause it just, there's a lot more to learn. So this is a little, I feel, a little upside down 'cause we can't tell you everything 'cause we haven't announced everything. So I'll just, you know, reiterate one more plug for coming tomorrow 'cause you'll also see and hear from a lot of our customers. And I just think it's a much more powerful way to learn and sort of feel what that relationship looks like and kinda understand the level of trust those customers have in us.
But, you know, just to talk about our customers for a minute, digital businesses run on PagerDuty. It doesn't matter what vertical they're coming from. It doesn't matter which organization within a vertical. It truly is a horizontal, ubiquitous application. And you could imagine we're growing in every vertical segment that's growing its revenue generated through mobile or digital assets, right? So it's pretty much everybody on the planet. And what I love about the frictionless land motion with PagerDuty is some really interesting customers find us and do really interesting things with our products and services that we didn't imagine. And as we learn more and more about that, we obviously build the integrations to support it, and we start to think about how do we bring those use cases to life.
But we've got over 12,000 customers, as I mentioned, more than a third of the Fortune 500, and we're in over 170 markets. So truly grow global in terms of use and application and growing in every segment from SMB to Enterprise, with, as you know, best in class Net Dollar Retention. So I'm going to invite the team up here to tell you some of these stories just so you get a break from the drone of my voice. First up is Jonathan. Jonathan is our Head of Product.
Hey, guys. Thank you, Jennifer. So first I wanted to talk about in the financial vertical. This is a consumer financial SaaS organization. They deliver tax, accounting, and personal finance software to 50 million users and small organizations around the world. They started using PagerDuty about five years ago, a little over five years ago, and a couple of their development teams took over and started using the product. They adopted it very quickly. What's interesting is they were also, what was mentioned before, trying to change the way they operate. They were taking a pure DevOps approach where the development teams not only build, not only deliver, but also have to support and manage that software once it's deployed. That's what we call Full-Service Ownership. Those teams over the next two or three years virally grew to other teams, the use of the product.
Very quickly around 2016, and just 2017, that grew to over 22 teams totaling over 1,400 developers across their organization. That's when they decided, since delivering an amazing customer experience was so important, that they needed to have one way, one system of record of how they dealt with incidents. And their central team, their central ops network operations center used a different product. So they decided at that point they wanted to standardize on PagerDuty. That was another 1,800 users in the centralized ops and network operations center. So now they have one single system of record when it comes to incidents and delivering a better customer experience. We've estimated, working with them, that this saves them just on people cost alone over $2 million a year annually.
Next up, Tim Armandpour, Head of Engineering, and the guy in charge when things don't go well inside of PagerDuty.
Thankfully, a lot of things do go well inside PagerDuty. Here we have a large fashion retail customer, a member of the Fortune 500 that sells tens of billions in apparel both online and through physical stores. In this world, high availability is incredibly important to this customer, not just on Black Friday, but also the other 365 days of the year. Every minute of degraded online performance typically results and translates into hundreds of thousands of lost revenue. The company's always been a technology innovator, and over two years ago, they brought in PagerDuty first to help with On-Call Management. They have since expanded their product suite to both include Modern Incident Response and, most recently, Event Intelligence.
Now, this company quickly adopted the DevOps mantra and the approach of, "You build it, you own it," which brought about improved productivity and responsiveness across the entire business. Users have doubled over the last six months to now totaling over 2,500 on the PagerDuty platform. The impact that PagerDuty has had on the company's operations has been significant. The data speaks for itself. 94% of incidents of the 6,000 human incidents that are happening monthly are being handled by whom they are assigned, which is a great signal over to PagerDuty telling us that we are actually helping to get the right person on the right problem at the right time. And again, showing that increased capability in responsiveness creates a ton of leverage for the company here.
The more challenging and complex remaining 6% of those manual incidents actually leverage our Modern Incident Response capability to dynamically orchestrate the right set of experts in real time, which then increases the number of complex incidents that can be resolved in under an hour. The time to mobilize the right team has gone from hours down to minutes, and the time to resolve is reduced from days to hours over this past year, allowing them to deliver a great customer experience and protect revenue. This company also just got started with our Event Intelligence product, and we've seen a reduction in incidents that needed human action by 44%. Given the sizable impact that we've had with regards to improving incident response and resolution times, we estimate that we've not only improved developer productivity but also helped save the customer over $1.8 million in a year.
And next up, we have Mr. Howard.
Yeah. So I have an example here of a company based in the U.K., a global leader in derivatives trading. And obviously, this is a fast-paced, highly regulated environment. Even you know, every second of downtime can have some very serious consequences. The regulators are watching them all the time in terms of both speed of transactions, accuracy of transactions, reliability, and transparency. And they started out with PagerDuty originally on our Business plan. And part of the motivation behind this was really about being able to orchestrate their teams more effectively. So they actually had good statistics in terms of availability and uptime, but they were doing it through humans by actually trying to make sure that they had enough bodies at any point in time trying to address the problem.
This is how they were trying to ensure that they could meet their performance obligations. But this is quite challenging because they were distributed across the U.K., Poland, and India. And what they found is that they were spending an immense amount of time just trying to schedule people to be on support and leading to gaps in terms of ownership. And once they put PagerDuty in place, PagerDuty was actually able to manage all of those aspects for them so that whoever needed to be responsible was clearly identified. There was a high degree of ownership around services. It was very clear who was going to be responsible if there was an issue.
And what they found that they moved from very manual processes where people actually had to physically be on site in one of their environments to an environment where they were able to mobilize teams a lot faster and people could work wherever they were. They used the PagerDuty mobile app extensively because people were now suddenly liberated and empowered to do their jobs elsewhere. They are a good example of a customer who then took on our Event Intelligence product, which you can see the growth in the ARR here was now driven by acquiring additional product in terms of Event Intelligence. The use of Event Intelligence is really around trying to ensure that they prevent unnecessary work and reduce incidents that would normally have gone to a human.
But because of the m achine learning that we have now, we're actually able to put that aside so that we're only getting people to work on the things that really matter. And from the study that we did, we estimate that the use of the platform has reduced their work by about 58%, and that results in annual savings of around $500,000. Stacey.
This on? Great. So I'll be talking about a mobile payments company today that does Point of Sale that I might or might not have worked at previously. So it's an example I know very well. So for this company, they're doing billions in GPV per year. Every time there is an outage or a slowdown, it's not even their customers that are gonna have an issue. It's their customer's customers. If they have to go to offline mode, they can no longer accept credit card payments. Way down the line, there's somebody who's not getting a transaction done. So making sure they're up and running is of utmost importance. So this company has actually been on PagerDuty since 2010, so quite some time given that they started in about 2009. So right at the beginning, and from experience, DevOps culture is paramount there.
So it makes a lot of sense. So recently, the company adopted Event Intelligence, as we've discussed before. It's a product we launched about in the summer of last year. So once again, it's taking m achine learning , correlating millions of signals, and they're getting millions of payment signals basically every moment. There are transactions that are going on across the world. So but we're only surfacing the incidents that need action, so providing that unique context and recommendations for resolution. The context allows the incident team to move more quickly to resolution because we combine machine data and, given some of the slides we saw previously about PagerDuty being part of the central nervous system, we combine machine data and human response data.
With each incident that gets resolved, the machine is getting smarter and smarter, and so the context and recommendations for the next incident are getting richer. We actually recently checked in with the company. Once again, as Jen talked about in our value proposition, getting our customers to understand the value that we bring to them. We recently checked in July. They'd been on Event Intelligence for only about a few months, even though the solution hadn't been in place for that long. They were already experiencing a 46% reduction in incidents. Equate that to thousands of hours of unplanned work, hundreds of thousands of dollars in operational cost, not to mention the revenue that they're protecting. Once again, getting back to this sort of thinking about it horizontally, so the broad applicability of our platform.
So not only are they're using our products in a traditional way, they've also been using our products for other teams that you wouldn't normally think of. So about 30 members of the Trust and Safety team now use PagerDuty to orchestrate responses to physical security issues across their global offices. And they're also using PagerDuty to triage and escalate issues for customers in five continents. And one thing that's really interesting is we had a little debate yesterday about, you know, there was another very similar company to this that actually has been using PagerDuty in different horizontal ways as well. So now I'm gonna turn this over to Jonathan, who's gonna show you an example in real time.
Perfect. Okay. I'm between you and break. Lucky me, so what I wanted to do is walk you through the PagerDuty platform. Now, we all know we live in an incredibly crazy, fast-paced world. Each of us, you know, myself, my family, I'm sure all of you rely more on applications and a lot of the business processes that fuel those applications more than ever before, and we do so both in our kind of private lives as well as in our professional lives. What I want to do over the next 10 to 12 minutes is walk you through a couple, a few capabilities on the PagerDuty platform.
And I'm gonna do that by showing you kind of a narrative of what a consumer who engages with two different organizations, two different products on the web, two websites, what it's like not only for her to experience and go through this process, but also, more importantly, double-click behind the scenes to those organizations and show you how they're using PagerDuty. What's also I wanna highlight in this is that in those organizations, it's not just the typical development and IT Ops that are using PagerDuty. It's other audiences. It's the executive team in one, and it's the customer service organization in the other. So meet Val. Val's a planner. Val has been looking forward to going on a vacation for a very long time and finally convinced one of her friends to go along with her.
So she has to, number one, book the cruise of her dreams and, number two, pay for it. So to do that, first, she goes to TravelAdventures.com. And what she doesn't know is behind the scenes at TravelAdventures.com, they're contemplating re-architecting their product, re-architecting their site, some of the major capabilities of what they deliver. Why? Because the travel industry is very, very competitive, and they know they need to provide a better user experience. After she does that, though, she needs to finance her trip. And for that, she's gonna turn to her financial institution, Grand Boulevard Financials. So Val diligently goes on site at TravelAdventures.com, and she starts to look at different cruises, and she realizes she's not exactly getting all the results that she wants, not a lot of options, and the search is a little bit slow.
So what I wanna show you now is behind the scenes what's going on and how a couple of executives, the head of development and the head of operations, are using PagerDuty in addition to the development teams to make some very strategic decisions about who in the organization they want to spearhead the rebuilding of this site. What I'm showing you right now are our automated postmortems. Jennifer referred to this earlier. This is all about how can we help teams, individual responder teams, not only better resolve more quickly and efficiently resolve incidents but then learn from it. We automatically create the report, the timelines, all the information, and the teams can sit in a meeting and learn from it so that they don't repeat the same problems again.
Increasingly, we're seeing executives sit in on these kinds of meetings because they really wanna understand how the teams are operating. Now, PagerDuty is a company of firsts. It's not about just providing a solution. It's about providing best practices. This was referenced earlier, and what I'm showing you right now are our best practices around our incident process, our incident management process. We open-source this and make it freely available to all organizations. We've had tens of thousands of downloads of this already, and customers are forking it and using it just like TravelAdventures.com to pioneer and set up the right practices and processes within their organization.
So what I wanna do now is actually show you how the Executive Team can look at part of our analytics, which is our operational business reviews, our quarterly reviews where they can double-click into each of the different teams, in this case, Team Hammerhead, and they can go in and find out how effective that team is. What they'll see is that if they look at information trending over time, the number of major incidents as well as the response cost is decreasing over the last three months. That's a great sign. It means that the organization, this team, is getting much more effective. Next, what I wanna show you is our Intelligent Dashboards. This is new to the analytics family of products, and this is where we're pulling in the power of our 10 years of data.
We're providing benchmarks not only for other teams within the organization but also the same vertical in the industry so we can see how Team Hammerhead is doing against other teams as well as against the industry average. And we see that there's an incredible volume of incidents that this team is dealing with. It makes us a little bit uneasy as the heads of engineering and operations, so we need to look at some other metrics in our dashboard. What I'm showing you here is the response effort in hours by the team. You can see in this case, the Team Hammerhead is the best in class for the entire organization. They're below the company average, and they're below the industry average. Now, that's making me feel good. It's making me feel great that they're the right team to spearhead this new redevelopment.
What I'm showing you right now are our Spotlights, a part of Intelligent Dashboards. This is us applying m achine learning to Analytics. And what are we doing? What you see in these three cards here, we have a recommendation engine behind the scenes that's looking at all the trending information, and it's generating a separate card. And each card is focused on a different team in the organization, giving them, telling them, number one, what's happening, and number two, giving them recommendations of what they can do to improve, to evolve, to get more effective at what they're doing. The last card is pretty interesting. It's about increased escalations. This is usually a sign that the alerts, the incidents are going to the wrong person. So they should probably take a look at their escalation policies and get better at it.
Again, the other takeaway is that Team Hammerhead, they aren't listed in any of these cards here. That's a great sign. So I think now they can feel confident both the Head of Development and the Head of Operation that this is the right team to start this project. So what I've just shown you really is how our Analytics with Intelligent Dashboards, which are making use of m achine learning , are offering recommendations looking at trending information over time so that an organization can make some pretty strategic directions and decisions. Okay. Back to Val. So Val's now booked her trip, but now she needs to finance that trip. So she goes on to Grand Boulevard Financials. And in Grand Boulevard Financials, she starts looking for different credit line options, different credit cards that she might wanna use.
What she doesn't know behind the scene is that not only is the development organization using PagerDuty, but another part of the organization is using PagerDuty as well, the customer service team. And one of the big problems in customer service organizations is just like other parts of a company in real-time work, they're often reactive. They have to respond when a customer calls up, whether it's your cable company or whether it's Grand Boulevard Financials with a problem. And only then do they start thinking about what is the problem, how do I need to respond to it.
What we're doing and what Grand Boulevard Financials is doing is using PagerDuty with all of their customer service agents so that they can proactively and immediately know when there's a problem on the engineering side, when customers are experiencing any kind of a search issue, and they can start preparing customer responses and orchestrate that among their teams and automate a lot of that process. So to begin with, as a part of our Modern Incident Response product, we have a new capability called Business Response. Now, there's two ways that an organization like this customer service team can use it. One, they can subscribe to the services that they care about so they can get real-time interaction. And two, the whole team can understand and look at a dashboard that we've provided as a core part of our product.
In this case, one of the customer service engineers just got an SMS message, and what it says is that one of the customers or several, a few customers potentially are seeing some slowdown in the search capabilities on the financial site and that, more importantly, engineering is on it, so this one agent is sitting there, and now he knows without having to walk across the building that engineering has identified an issue and they're on it. The rest of the customer support organization can look at the dashboard, our Business Response dashboard, our status dashboard, and see that same information. They can see that this P1 incident has been ongoing for a minute. It's potentially impacting customers, and we'll look to deploy a fix shortly. Now, this is a real-time view that the organization has.
They can see what the engineering teams are doing, again, without interrupting the teams that are working on that incident at that time, and that's really important. Now, what we can see is there's an update. Again, this is in real time, and no cases have been opened on the customer support desk. The engineering team is now telling the customer support team by automating this orchestrated response that the fix has been deployed, and it should significantly reduce any issues in the next 10 minutes with full recovery within the hour, so this P1 has been downgraded to a P2. What's also important, oh, some more information is showing here. We can see that only 2% of customers are being impacted by this. Still, no customer support cases have been opened yet.
Now, this is great information because they need to start putting together. Whoops, let me what just happened? Let's go back for a minute. I fat-fingered the keyboard. What just happened is a Slack message occurred. And that Slack message is absolutely critical because that's the first time that a customer has come in and opened up a customer support case, which is also integrated as a part of the PagerDuty platform. So now we're further ahead. Let's go back. Right here. Okay. So the Customer Service team has proactively given all this information so they can start putting together all of the communications that when they get cases opened up, they can push that information automatically out. And again, they know that it's just a small segment. There's that Slack message that came in. Slack is integrated as a core part of the PagerDuty platform.
And in this case, the Customer Service team is working not only in PagerDuty, but they're working in Slack. They can work where they are. They don't have to context switch between different products, and they can acknowledge that this case has been opened up, and they can also add actions. In this case, they could further automate the process for the customer support team, or they can add a note. So this one agent has gone in and said, "We have the communication template. If any more cases show up, use this template from this link and push it out to the rest of the organization." So in typically, what we've seen and we ran some studies last year, with an organization called IDG. What we saw was Business Responders, folks outside of the technical team.
On average, it takes 44 minutes for them to be mobilized and take action on a major incident. That's after the incident team is working on it, 44 minutes. In just a couple of minutes, we've communicated and educated the entire customer service agent team so they can go back to their dashboard, and now they can see that all their services, inclusive of the search, are in the green. What I've tried to show you today is a couple of things in this scenario. One, how customer service organizations and all of the agents are really able to break down the walls between development and themselves. They don't need to walk over and interrupt the team to understand what's going on.
Number two, with the information flowing in with Zendesk in this case, customer cases, and with PagerDuty with the Development team, they can get proactive in developing communications to push out to their customers. At the end of the day, our heroine, Val, has done a great job, both getting her vacation booked and financed, but more importantly, the teams behind the scenes, both of these organizations, have delivered a better customer experience and become much more proactive working with their customers. Thank you.
Thank you, Jonathan. So hopefully, folks, that's given you a little bit of an insight into how customers can use our product. That was just a peek into some of the elements of how PagerDuty can be used, a number of different use cases, a number of different aspects of the products we've released over the last year and a half, which is on display in that demo. So thank you very much, Jonathan. We're now going to have a break for 15 minutes. So if we can have folks back here at 3:25 P.M. to 3:30 P.M., we'll then continue with the session. Thank you. Righty. Well, welcome back, folks. Everybody's in the room. That's good. You're all used to being on time. That's amazing. So, for this next part of the afternoon, Tim Armandpour will lead off.
He's our SVP of Engineering, and he's gonna talk a little bit about how we think about engineering and technology and how this is an advantage for us. And then Jonathan will be back to talk a little bit about our product and from the perspective of the problems that we solve over time and the way in which we're thinking about it for the future. And then I'll be doing a discussion from a financial perspective and then Q&A. So over to you, Tim.
All right. Good afternoon, everyone. I am Tim Armandpour. I've been here about four and a half years and had the privilege and fun to see not only the company size grow by nearly 10X but also the business, so we've been hard at work in the world of engineering, since almost day one of the company starting by founders, so I've got a handful of areas that I'm gonna cover, but most important takeaways here are gonna be how not only our choices around technology, how we actually put technology out in the wild, but how we work, and how that positions us very well, not only for now but for future success. You know, we talked about how we have 12,000 customers globally.
When you're in the business of running an always-on platform, we don't necessarily have the luxury of predicting how the traffic volume is gonna come to us, right? We don't have seasonality similar to Black Friday or Cyber Monday. Quite the opposite. We actually have to be prepared for the unexpected at all times, not only for our largest enterprise customers but all the way down to, like, tomorrow's startup, but also there's a lot of shared infrastructure in the world now that many companies depend on. You read everybody's S-1, you're gonna see some kind of call-out to a cloud provider or a telco provider and whatnot. A lot of that shared infra is actually building up the next set of foundations for all these companies that also rely on PagerDuty coming into us.
One area that’s always fun to talk about for why this statement is very important is when there’s an interruption in service, let’s say, with a major cloud provider. When that happens, part of the world as we know it does get interrupted, right? Jennifer alluded to it earlier. But if you’re at home trying to stream content and it starts to lag a little bit, there’s a disruption in the household. You might be rushing home to order ahead for food to be delivered on time for dinner, and that doesn’t happen. It creates quite a bit of a ruckus, right? I know my household goes through that. Simultaneously around paying bills online, right? Frustration when you can’t actually move your money where it needs to go, right? Those are just kind of first set of examples, tip of the iceberg.
But that's why when we say that, you know, how we orient our teams around PagerDuty being at our best when all of our customers may not be is absolutely paramount. So we put our money where our mouth is. We invest heavily in the world of technology. Just recently, a couple Saturdays ago, we had a major cloud provider have an interruption of service, right? And those are always fun when they happen on Saturday afternoons, no matter where you are in the world, right? We have a ton of customers that were impacted. And when that happens, essentially, a bunch of data signals start coming to us where part of the world is starting to yell at us.
And, much of our role in that regard as a platform offering is to withstand that and do our job, which is to ingest those signals, analyze those signals, correlate, and make sure the right people get the right info at the right time. So, in that moment, a couple Saturdays ago, we were able to withstand a 300% surge in traffic. It's equivalent to a Category 4 hurricane hitting you out of nowhere, right? We didn't flinch. We didn't miss a beat. We continue to serve our customers as they expected. In our world, you know, we've cared very deeply about working very hard to earn the trust of our customers every single day. In addition to the key principles I'll touch upon in a moment, left-hand side, you see DevOps at the core. Jennifer talked about that a little bit.
DevOps is a cultural movement and a set of philosophies, tools, and processes that help an organization deliver at higher velocity, so ultimately, the name of the game is to deliver more and do it faster, and when changes are going in a live environment, right, and there's a new code going to live environments, right, typically that creates risk. Ultimately, part of the job is to figure out how do you actually work your way through both the notion of velocity and couple that with the expectation around increased stability, right? High rate of change typically results in higher risk profile to your ability to service your customers via your platform offering. Underpinning DevOps here at PagerDuty is the notion of Full-Service Ownership. Now, how that gets defined is developers being responsible for the code and services that they create.
So there's no handoff, separate operations teams. There aren't separate quality assurance teams. There isn't a separate support team just to be ready and waiting to inherit what you just built, right? We believe it's really important for the technologist at the edge of the customer experience, fully empowered. And what starts to happen over time is it leads to almost like a friction-free environment. So instead of typical toll roads of how to get your code out to environments working across a sequential set of processes, you now have an autobahn. Now, as you get on the autobahn, right, you can drive fast, but you better be responsible. And in the world of PagerDuty engineering, we create a very nice flywheel effect of this notion of I develop my code, I operate it, and then I go on call.
And once that starts to happen and you get better and better that more customers accustomed to it, you get this incredible amount of empowerment that takes place in the organization. And, you know, when we talk about owning your own destiny of outcomes, when that accountability becomes something that doesn't have to be manufactured and it's a new norm, you're going to see results that showcase in, for example, higher quality of code being developed, right? Last thing I want to happen when I'm a developer and I own it and I'm fully responsible for how it operates and it lives in the wild is be interrupted, especially when I'm sleeping, or it could be when I'm driving, or it could be when I'm hanging out with my kids.
So it's really important that here at PagerDuty, we harness that, and that becomes the absolute norm. Finger-pointing is not really a possibility, right? We are big believers in, depending on how you look at it, eat your own dog food or drink your own champagne, but PagerDuty is very critical to how we operate our own service offering. Once we find this leverage in the organization, that a culture of accountability becomes the just the way we start to see how we put this into the work so at the top of the circle, Cloud-Native. This company was born in the cloud, right? We started a little over 10 and a half years ago, and what that starts to manifest as, you know, our environments, our infrastructure is actually elastic.
There are other companies out there that will take their typically on-premises software, they'll package it up, ship it up to some new compute in the cloud, and call it cloud-ready or cloud-native. However, it's a world of different architecture when you have to architect for the cloud versus software you're gonna run in your own facilities or data centers. Now, in the cloud, your approach to architecture also has to be different. You're actually in the business of architecting for failures, right? There's a spider-web mesh of fun in the world of dependencies and abstractions you may not readily understand. So you have to be incredibly agile and nimble to get ahead and/or even better than what those dependencies are that you depend on, right?
And when you understand those abstractions in a highly distributed computing environment, you have to get very good at withstanding, just, you know, basically tolerance for disruption. At PagerDuty, we've been designed from the ground up for the utmost amount of resiliency, and honestly, we are tested every moment in the market every single day by the leading enterprises out there, you know, multi-region is something I'm gonna touch upon in the next slide. It's worth going a little bit deeper in there. However, the world of redundancy and resiliency is actually super important as well. Redundancy meaning there is no single point of failure, right? And so how that typically manifests is here at PagerDuty, at almost every single layer, that's from how we manage DNS, domain name look-up services, all the way down to our data stores.
There are very few single points of failure, at least that I know I'm aware of. But every single day, we're working harder and harder to make sure that we minimize that such that if there is a disruption somewhere, right, it's never just one thing. And that one thing most likely has a secondary and a tertiary way of continuing to operate. The world of resiliency here at PagerDuty has a lot to do with the assumption that the system will fail. Anybody who tells you otherwise is no doubt lying, right? The system and the environment's in a constant state of change, and we have to get better and better at withstanding failure. We know that, when we are typically in recovery mode, which honestly happens almost every single day in some part of our system, it's pretty rare that our customers actually notice.
You know, we use the word customer a lot. It is near and dear to our heart. We have a tremendous amount of passion and empathy to making sure that our customers are wildly successful. That speaks to the quality of service offering we have. We deploy quite often. This year, we're on track to deploy more than 20,000 times the live environments, right? That's a pretty high rate of change where we're also expected to continue to maintain that same high level of reliability, resiliency, and general availability. Last piece I wanna touch on that's important is how we approach operational excellence, especially in engineering and providing the service of the platform that we do is quite a bit of investment into chaos engineering.
If you were to look that up on Wikipedia, it's defined as the discipline of experimenting on a software system in production in order to continue to build confidence in the system's capability to withstand turbulence and disruption, right? Very much like I view it as like, you know, you've got, you're flying on an airplane. You're expected to go through clouds. You're expected to deal with that stuff. You designed for that from the ground up. That's where your starting point is 'cause you know something could go awry, but you gotta be well prepared, and here at PagerDuty, we leverage chaos engineering where every few minutes a day, there's something that's on purpose automatically creating some kind of disruption in our service offering, and that helps us build our muscle for continuing to get better and better and better at operating this platform at scale.
So this is ultimately why the largest enterprises choose PagerDuty. Nobody else in the market invests, executes, and delivers the way we do. We have multiple cloud providers in play, right? And these cloud providers have within these cloud providers, we have multiple regions, which also have multiple Availability Zones. So we have let's walk through what's a region. Region is a distinct geographic area. It could be Fresno, California, Montreal, Paris, London, Stockholm, etc. Within each region, you have what are called Availability Zones. These zones are actually a set of isolated physical facilities or their data centers where there's distinct power, cooling, network, and other infrastructure that's powering that particular facility. So here at PagerDuty, we have two cloud providers, four regions, nine Availability Zones in play, right? They're not working in isolation.
It's not, we are not in a mode where one Availability Zone is running, and if something happens, let's try and move over to the next one and keep trying until the system actually recovers well. It's quite, in fact, the opposite. When we say active- active, some of our workload is actually specifically designed to run in parallel irrespective of whether or not one Availability Zone is actually working or not. So in cloud speak, when there's a regional interruption, it's pretty rare you're gonna see PagerDuty post publicly and name the cloud provider or name a given provider, whether it's across other dependencies like DNS or telecommunications providers. We have multiple of those along with email providers that help ultimately deliver that last mile of the notification to the end users globally, right?
The reason why we won't publicly go on and post that that one's having issues is because we got others in play that automatically fail over. And we've created a set of proprietary heuristics that help us do that literally each and every week. There's a lot that goes unsaid behind the scenes, but that also speaks to our ultimate resiliency profile. Last piece I want to touch upon is the volume. So on our platform this year, we're gonna process approximately 19 billion events and API requests. That's also going to result in approximately a little over 400 million notifications dispatched globally, right? So this is again a platform, as Jennifer and Howard and Jonathan mentioned, has been operating at scale for some time. That is not new for us.
In fact, we do quite a good job of staying ahead of our business growth, again, across the leading enterprises. We talked about earlier about multiple SLAs, right? There's the availability one. Much of what you might read around in the industry, especially in the world of SaaS, as it's evolving, is people touting, you know, an incredible amount of uptime. Uptime typically means that the piece of hardware that's plugged in is actually running power, and the lights that are blinking continue to blink. That's actually quite easy in today's world. It's not that hard. What gets harder is standing behind your promise to your customers and saying the availability that you expect, the functionality you expect that you're paying us for, is gonna continue to work no matter what, right? That's one part of our SLA.
The second part has to do with performance SLA. That's we're going to dispatch the notification that you expect each and every time within five minutes globally, irrespective of that's going across SMS as text, voice, push notification via the two main providers there, and/or email. So collectively, across all those channels, the burden's on us to make sure it gets to where it needs to get to. The piece that I think sticks out the most here is this notion of no scheduled maintenance or downtime. I often say internally that, you know, scheduled maintenance is for cars and not for SaaS, right? Who are we to stop the clock of another person's business? I don't think we actually have any right to do that.
I actually find it quite unacceptable if I'm negotiating with vendors in my own right around them start stopping the clock. What am I supposed to do? Do I have to build more and more redundancy there? It's quite expensive and a pretty tall order. So we take that burden off our customer base as well. Our clocks never stop. We do believe and stay wholeheartedly true to the fact that the world is always on. It does need to stay that way. Which is why a lot of the literature you might see from us, or if you see some of my engineers out there on the talking circuit, we actually don't talk a lot about high availability. We believe that there's actually a need for continuous availability. That's how we approach much of our, much more of our platform offering.
And again, just to recap, this is exactly why the largest enterprises continue to choose PagerDuty time and time again. Nobody else in this market is investing and delivering in this manner on the technology side. And we continue to deeply care about our customer s uccess. Thank you. And coming back up is Mr. Jonathan.
Thank you, Captain. Okay. What I wanted to walk you through over the next 10 minutes is a little bit of the evolution of PagerDuty. This was touched on before. Again, we started over 10 years ago, and we started in the alerting and on-call area. And around 2012 is when we had our first kind of significant bump of customers, 20, 25, 25,000 users. And really, the problem we were solving there, and that's based on the backbone, the foundation of all of the resiliency and the scalability and the performance that Tim was just talking about, is really engaging the first individual in an organization and doing that at scale, right? And when you think, "What, what's the value proposition of that?" it's really about reducing that mean time to acknowledge. That's just the first person that gets notified.
Not a lot of intelligence there, but it is, you know, most of the time a manual process and can be slow. It can be improved and should be improved first. Around 2016, 2017, we really filled out our solution suite around Modern Incident Response, the full incident lifecycle. This is when organizations of all sizes, whether just a single team or multiple teams, where that has to be orchestrated a response to get all of the real-time work done is important. A great example of this that I'm gonna talk about tomorrow is Cox Automotive. So they're one of our customers. And they, the main benefit and the value of Modern Incident Response and doing it well with the PagerDuty platform is reducing not mean time to acknowledge, but mean time to resolve, right?
And they're seeing upwards of 56% improvement in the mean time to resolve issues using the PagerDuty platform. Now, in the last year, we started talking about Digital Operations. And this is really much broader than just Modern Incident Response. And this is why we're taking advantage of all the data that we manage within our platform and applying m achine learning to it because it's helping organizations transform digitally. One, another one of our customers I'm gonna talk about tomorrow is IG. IG is a company out of the U.K. They're using our Event Intelligence products. This goes beyond just incident management. They're not only seeing the benefits of, you know, reducing the time to resolve, but they're actually reducing the number of incidents, right? Because they're intelligently grouping alerts and incidents together. This has a savings to them of reducing that by 86%.
So these are just a couple of examples how we're helping organizations mature and seeing the returns that they're getting, which is really important in a world where complexity is really going up at an exponential rate. And the number of interruptions to teams is going up at that same rate. Last year, we conducted an extensive research project. We went out with a firm, and we looked at not only PagerDuty customers but non-PagerDuty customers because we realized in working with a lot of our market and with our partners and customers that there's a broad range of maturity, right? There are those who are just starting, and there are those that have been doing this, as we talked about in some of our customer examples earlier, that are doing a wonderful job of it.
So we created, working with that firm based on 600 in-depth interviews, our Operations Maturity Model around real-time work. And what we see is, you know, the beginning organizations will start out as being reactive. Then they continue to work through and automate, with on-call and alerting. They move then to being responsive, starting to deal a little bit about full Incident Management and Modern Incident Management. But then they need to get proactive. They need to use, what I'll talk about, and then more of the platform. They have to get ahead of all of these interruptions and start proactively with their constituents, with their customers, understanding and educating them, still for the most part within kind of an IT and development sense how they can get better. Lastly, we have preventative. So this is where most organizations we work with aspire to be.
This is where it's not just the development and IT Ops part of the organization, but it's also customer success, security, IT BizO ps, marketing organizations. In the physical world, in IoT, customers are starting to get preventative using a lot of our capabilities far beyond just the notion of a development and IT team, using our products for Incident Management. The other point I'd like to make is that as people mature, that value goes way up. The two examples I talked about with Cox Automotive and IG Group kind of underscore that point. And that's how we've designed the PagerDuty platform. We always refer to it as a real-time operations platform. And it all starts, if you think of that maturity model, how it complements the platform itself, with On-Call Management. And O n-Call Management, we think about it as made extremely simple and delivered at scale, right?
This is so that teams can, in a self-governing way, get in, create their own on-call rotations, and be very productive in automating what are typically a lot of manual processes. This takes advantage, again, of all of that foundational work that Tim was just talking about. We have to be on when other organizations aren't. Next, Modern Incident Response. Modern Incident Response really focuses on a couple of things. One, orchestrating communication. It's not just that first individual getting contacted, but it's all the other folks dynamically. At the same time, you may not even know what the team needs to be to work on a particular problem, and in that instant, kind of that moment of truth, we'll pull that team together automatically for you in the platform, and Modern Incident Response is kind of the backbone of that.
It's also about further communicating outside of just the core teams, so as I talked about in our demo earlier, it's not just about communicating with the technical responders and making sure they're on the same page. It's also about Business Responders and it's also about communicating and orchestrating response with stakeholders, people who just need to know. Next is Event Intelligence. Event Intelligence is about reducing the number of issues coming in, not just reducing the time it takes to work on one, but reducing the overall number, and that's where we apply our m achine learning in addition to all of the machine data that's coming in, looking at our human response data.
Human response data plus machine data, m achine learning on top, all of that allows us to intelligently group alerts, to find Similar Incidents, and allow organizations to actually train and educate newcomers into the response process, into the incident process, so that they have all the important context they need when those moments happen. It's a highly stressful event, and all of this information we provide to them about who previously worked on these incidents that were similar and what they did. That's incredibly important information, and we're the only organization that integrates that, only business and solution in the market that integrates that with the incident process today. Next is Visibility, and Visibility is all about showing the business impact in those moments during a major event.
In this case, what I'm showing you is on a mobile device, we're able to show all of the responders, even business owners above the response teams, what the actual impact is of an incident, correlating the outage time with how many tickets are being opened up in the support desk, how many people are unable to finish a shopping cart transaction, and tell, and communicate in real time what the financial impact to the organization is. And lastly, Analytics, and Analytics is all about continuous learning and improvement. I showed you in the demo how we're applying now m achine learning to that as well, with a recommendation engine so that teams can not only see how they're doing at a business, a service, and an individual team perspective, but more importantly, see how they can get better and make improvements over time.
And lastly, all of our use cases. This is something that we continue to invest in. We do that because real-time work isn't the exclusive domain of the development and the operations organization. As the world is always on, all parts of an organization need to be ready and effective in dealing with opportunities as well as major incidents. So where are we making investments across the platform? What I've tried to share with you today a little bit is how m achine learning is an overarching umbrella that we're applying. We have a fantastic team of data scientists and data engineers that are making use of all of our data. Why? So that we can get smarter, so we can start help our customers get smarter and help provide them recommendations to make better decisions and mature their practices.
At the foundation, though, is everything, again, that Tim spoke about: major investments in scalability, availability, performance, and security. We lead the industry in this area. But in the middle, there's really three major buckets of investment: visibility and context, so we can help organizations in those major moments be more effective, lower the time, and lower the number. Two, Automation and Orchestration, so that we can automate mundane activities across the entire, all of the constituents, all of the audiences. And then continuously learning. So some of how we're applying, again, what I showed you in Analytics, how we're applying m achine learning and a recommendations engine across the platform. This year was a busy year when it came to innovation for us. At the beginning of the year, we had our first release of Analytics, our what we called our Operational Scorecards.
These are curated reports that go to the teams, to management, and also to executives within a business so they can continuously get better. In Q2, we released the latest versions of our integrations with ticketing systems and with ChatOps products, and the big focus of this area was allowing teams in these critical moments to work in the environments that they need to and still interface and share all of that important information with PagerDuty. At just at the beginning of Q3, the end of Q2, as a part of our Modern Incident Response product, we released Business Response, which I showcased to you as a part of that customer service use case demo earlier, and here at PagerDuty Summit, we have a number of items and updates we're announcing as a part of our core products. In Event Intelligence, we'll be talking about Intelligent Triage.
That's dealing with the most complex problem within organizations, IT and where multiple teams will get spun up on incidents, and they need to have visibility and context to other related incidents outside of what they're working on so that they can know whether to engage or whether to back off. We often see in a lot of organizations that one team starts to work on something only to find out that in real time, another team is undoing what they're working on on a related or Similar Incident. Intelligent Triage eliminates that and provides that visibility across multiple teams. We're gonna be talking about more Service Directory, which is the one location where any team that's on call that is managing major incidents can go to find the latest status on any services and, more importantly, who's on those.
We find that we're putting that all together in one location and making that available to teams so they can be more effective. We talked about Intelligent Dashboards as a part of Analytics. We'll also be talking about our new partners and integrations we have with Zendesk and with Salesforce to further support the customer use case. And ultimately, we'll be talking about how we're enhancing our developer ecosystem. So we're making it that much easier for our whole community to go out, extend the PagerDuty platform, and also publish those extensions so everybody in our community can take advantage of that and use that as a part of their practices with PagerDuty. I guess it's over to Howard.
Thank you. Thank you very much, Jonathan.
You bet.
All righty. We're onto the final section of the afternoon. What I wanted to do was just run through a few of the financial highlights. These are as of the 31st of July, so end of Q2 for us. Our fiscal year, as a reminder, is January of each year. This is the numbers as of the end of Q2. Trailing 12 months revenue at $143 million, revenue growth at 48%. As Jen mentioned earlier, we are going after a total addressable market that we estimate to be around $25 billion. That's made up of 85 million users across development, IT, customer support, security. And of course, there's a larger TAM that Jen alluded to in her discussion. Our dollar-based net retention rate at 132%. Over 12,000 customers.
We now have 274 customers spending more than $100,000 a year with us. Our business is predictable because it's almost all subscription revenue, although we have a small percentage of services, around 1%, and gross margins above consistently around 86%, so I'm gonna just talk through a few things here. One is, you know, how is it that we grow, so there are a number of elements that are important to us from a growth perspective. We certainly see enterprise and mid-market as being an important area for us. We continue to grow in the enterprise. When we think about enterprise, we're thinking about companies that have revenues over $1 billion. Our portion of our revenue coming from enterprise continues to grow. We're above 50% in enterprise. We think that within mid-market, there's still a large underserved opportunity.
From an international perspective, we've certainly noticed the demand signal outside of the U.S. International revenues at the end of last quarter represented about 22% of our revenues. And we've seen growth rates in international of around 59%. And we see this as still a large opportunity for us as a company. In terms of new use cases, we went through a number of examples today that showed where PagerDuty is being used in a customer support environment or in a security environment or in a trust environment or compliance. And that to us is another significant opportunity for us for real-time work beyond those traditional places. And then our new products, we've seen good traction. One of our older, more established new products is Modern Incident Response, which has been around, as Jonathan mentioned. That's the longest of the newer products.
But certainly, Event Intelligence and Business Visibility and Analytics, we're seeing some good momentum there. And Jonathan alluded to the fact that we're doing more in terms of our ecosystem, allowing developers to engage around that. So these are the ways in which we grow and provide us with multiple engines for growth. In terms of the subscription revenue, as I mentioned, this is where we're sitting today. A small percentage of our revenue is related to services. Most of these services are either around education, PagerDuty University, or they're around best practices, advice that we provide to help companies deal with the change that they're going through. So we have a small team that's focused on that in helping customers be successful. When we look at our revenue, we have a predictable model with sustainable growth over time.
This has been, of course, fed by the fact that we have low churn in terms of customers that leave us. Our land and expand model means that we have a fairly predictable view on as customers land with us and their ability to continue to expand and grow within their existing teams. While we're chatting about the revenue, I do think it's worth mentioning we have about 20% of our revenue coming from customers who are on a month-to-month arrangement with us, and the rest are customers who are on term contracts. We've shared that information with folks before. Those term contracts period of eight quarters, we've continued to sustain an increase in the annual revenue per customer, annual average revenue per customer. This comes from a combination of things, them adding new users but also adding new products.
The expansion in the enterprise and mid-market is really seen when you look at those customers above $100,000. We purposely selected this group because we feel that it gives a strong indication of customers who are either enterprise customers, so companies with $1 billion in revenue, or mid-market customers, which sit for us between $50 million and $1 billion. Most of the accounts that you would see here are enterprise or the upper tiers of mid-market. We were pleased to see 32 new customers in Q2 who had matriculated into that group. Within our customer base, continue to see good growth of expansion within our customer base. Just as a point of reference for this particular slide, the FY 2020 cohort, of course, is not completed.
So some of the growth that you would see within FY 2020, that's been delivered into those other areas, would only represent a portion but fairly healthy layer cake view of how our customers expand with us. And then in terms of our gross margins, Tim spoke about some of this when he went through his part on our technology advantage. The things that contribute to this are the cloud-native architecture, a very smart architecture as well, architected specifically for the cloud, which means that it's a highly efficient architecture. We're also very good at managing those multitude of arrangements that we have with the redundant providers across every level of the stack. So we optimize that infrastructure.
When it comes to things that are typically resource-intensive, like customer support and customer success, we follow a very programmatic approach to those so that we don't try and solve every problem with humans but rather try and solve it through programs and technology. The fact that our solution is easy to deploy, there isn't a high service component, means that we're not carrying a services team with, you know, bad margins. We're able to ensure that we can sustain that and then our DevOps approach to production, which Tim spoke about at some length, which means that we have high levels of ownership around the code, which leads to an efficient operating model and then from a net retention perspective, this is how our dollar-based net retention rate subtracts over time.
Things that can play with these numbers, of course, is that when you have a customer that does a largish upfront purchase, if they're buying for a under a project initiative, then sometimes what that means is that as they deploy those users, the expansion opportunity is lower because they've already bought in some respects ahead. That's not typically what we see, but it is an element, a characteristic that sometimes plays into this. We also, though, typically see customers, though, landing, and then depending on the cycle that they're in, expanding with the users, as they need them when they need them. When we look at, you know, our view on investment and future growth, these are the numbers looking at, how we, we've tracked in terms of R&D, sales, and marketing, and G&A. The FY 2018 and FY 2019 numbers are there, and also then the last 12-month numbers.
So right now, from an R&D perspective, at about 26% of revenue, sales and marketing around 52%, and our G&A around 25%. This year, we've been intentional in terms of investments that we've made from a sales team perspective, hiring ahead of in terms of building our sales capacity in order to meet some of our longer-term needs, and also an increased spend in marketing, with events such as our Summit event this year, which is our flagship event, which happens in Q3. These events get bigger and bigger as we grow and we have more people attending, but also, you'll have noticed, hopefully, you've seen our advertising around town. We certainly, it's the first time we've done any above-the-line advertising, so a fair amount of out-of-home and other advertising campaigns to help promote our brand.
And then just with reference again, this is the same long-term non-GAAP target model that we shared with folks when we were on the road with the IPO. More effectively than anyone else can. Our 350 integrations provide our customers with access into an ecosystem that, again, is unparalleled. And our open API environment that we have means that customers are able to essentially work with us from any software-enabled system or device. The breadth of our functionality in terms of solving multiple problems and multiple issues for customers and the secure, resilient, and scalable environment that's been built for real-time, those are, for us, things that are powerful foundations for our success. And when we look at our growth engines, we've chatted a bit about enterprise and international.
You've seen a number of examples today of how, as a platform, those new use cases are gaining significant momentum. And of course, the opportunities we've created with our new products to drive some upsell of new products and our typical or traditional, growth method in terms of customers adding users. And those are the things that are, that have helped us be successful to date and constitute our view on long-term success. So at this point, I'd like to invite Jennifer to join me and Stacey to manage the microphone. We've gone a little bit off the original timing, but we certainly wanna make sure we have some time for questions. We'll use the microphone for questions because the folks on the line will not be able to hear otherwise, and we have a number of people who've joined us. So, Stacey, do you wanna take it, please?
Thank you, Jen and Howard for hosting us. This was really helpful in terms of refreshing the story and getting us up to date on all the things that you're working on. Well, I think one of the things that we get asked about most is just the broader platform story. With things like Modern Incident Response, Event Intelligence, Visibility been out for some time, still relatively early. What have you learned in terms of the uptake of those products? What sort of what does adoption sort of look like? And where could things like change in terms of the pricing strategy to drive some of that adoption? Then I have a follow-up.
I'll take a crack at that, so for those listening and those who are new, we only launched Event Intelligence last summer, so I just wanna point out it hasn't been around for a long time. Modern Incident Response is less than two years old, so I know we live in a what-have-you-done-for-me-lately sort of society, but I mean in SaaS, if I would tell you that if we built a startup on the back of those two products, we'd be very happy, and so would our investors, so I think we're making very good progress there. To answer your question around what we've learned, we've learned a couple of things. One is that the developer who often purchased our first solution may not necessarily be the decision-maker for instance, Event Intelligence because it derives a much broader ROI.
So those ROI numbers that I shared with you earlier, you know, one question you might ask is like, "Wow, that seems really great. Like, could you grow faster, etc.?" 'Cause everybody always asks, "Can we grow faster?" The way I think about that is we could do a much better job of articulating the value, like the quantifiable value that our products and services create, as opposed to just solving the technical problem that we started with 10 years ago. And that is a journey that we're on as a go-to-market organization. But we're now in a position to speak to phenomenal ROI numbers because of the leverage that we get from m achine learning and the data in the platform. And that data is one thing that makes us platform-esque. The reality makes us platform-esque.
The Analytics, being able to look back and drive and improve, you know, the way, not only the way you manage an incident and the benefit that you get from the efforts of employees or the health of your service, but the long-term value that you get from the investment as that investment grows, so it's very early in the race. We're not gonna share additional numbers from what we've already shared on new product adoption because it is still early, but we feel very strongly about them. I'm very excited about Event Intelligence, and you know, Jonathan glazed over this, but I hope you'll come tomorrow and see the demo on Service Directory because Service Directory, think of it as a registry. I mean, it is a dynamic registry that helps you understand which services are doing what, who owns them, and why.
These are dependencies that don't live in a CMDB, that don't live in a static Active Directory that you can only understand if you have both the responder and the people data and the service data. And this is something that CTOs asked us for over and over again. And the last thing I'd mention is the, the change in the way companies actually think about the strategy of managing their customer experiences. So it used to be that no one would really talk about incidents and outages. You know, that was the domain of the NOC and the people in the basement, etc. And now it's CEO conversation, right? I can't tell you how many CEOs come to me and say, "You're the gold standard. We wouldn't use anything else because my butt's on the line. My investors noticed.
My customers noticed." I've had CEOs who have called me from their war room and said, "We just had a major outage. We've gotta figure out how to get better. I need you to send a team in to help us understand what to do." So the last thing I'll mention is part of the platform is also our expertise. And the expertise that we've built, you heard Tim talk a little bit about the way we manage our own services. There are a lot of people that look to us and say, "Can you just send someone in from your tech team to talk to us about how you do what you do?" Because it is so differentiated.
So, you know, I think, I guess that I think when you have a really strong following on your first product, it's hard for people to see beyond that, particularly if your name is PagerDuty, which, you know, if I had a dollar for every time an adult told me I should change the brand name, including when I told myself that when I first took this job, you know, I probably would not still be working, but there is an enormous opportunity, I think, for us to continue to build a very strong business on the back of our first product, followed by the platform and, you know, the add-ons that we've brought to the table, and a lot of the customers that are here this week are here to talk about their experience with our new products, and they're really excited.
Yeah, and I think what I would say, you know, we're trying to make sure that we showcase now to people, in particular, those new products. So today, you would have heard a number of the examples we gave were about Modern Incident Response at scale. This isn't kind of like five people and their dog using it. It's at scale. Event Intelligence, you know, we're again at the point in organizations where it's delivering $500,000 worth of benefits. So these are real examples. So we're happy with the traction that we're seeing, particularly with Modern Incident Response and Event Intelligence. They're a bit ahead of the others. So we're seeing some really good signals in terms of how that's tracking.
One of the shifts I'm looking for in the industry, if you think about the way perpetual software was deployed back in the day, you know, you started with a small team and a beta, and then you like inch by inch crawled along until you could kind of stand up and walk, etc. These would be really long deployment cycles. With SaaS, you shouldn't have to like crawl, crawl, walk, run. But a lot of customers, particularly IT organizations, still think about deploying new technology that way. So dev really leads the way in showing how you deploy for everybody. Everybody gets benefit instead of a group of power users, etc. So that democratization of new feature sets, something enterprise customers are still learning. You won't see that in a startup per se.
I mean, we have a lot of startups where every PagerDuty, every employee is on PagerDuty.
If I could do one quick follow-up. In your presentation, Jennifer, you said an interesting word about patience.
Yeah.
A patient growth. And I wonder if you could help us understand how that maps to your go-to-market model. Are we gonna be emphasizing?
Yeah.
How we've grown historically with a world-class self-service?
I'll just give you a really quick example. Like, one of the things that developers don't like is being told that they're going to need to standardize on something. It's much better for us as a developer-led product if the developers adopt us first and it becomes the de facto standard over time versus we go and sell one deal to a CIO who forces it down upon everybody. I mean, the quickest way to get a developer to dislike you is, you know, force their hand. So, and it's also very efficient.
Like if you have an enterprise rep who only does six deals a year, six $1 million deals a year versus doing 30 deals a quarter or 25 deals a quarter, it's a much more efficient way to grow because while that rep is doing smaller transactional deals inside the business, it's growing, and so that's what I mean by being patient. I don't need to sweep up all the business at once in a $1 million, you know, or $2 million deal. I'm happy for it to grow over time. It also means that you get stickier because users are adopting it as they need it as opposed to buying shelfware that then you take two years to digest.
So if you think about like our high Net Promoter Scores and our high customer sat rates, I think a lot of that is because we're very focused on making sure that we're not deploying license ahead of requirements or ahead of our customer's ability to use them.
Thanks, Matt Hedberg from RBC. Thanks for having us here today, guys. I thought this was great. Howard, one for you and then a quick follow-up. Last quarter, your net expansion dipped down a little bit. I think on the call, you noted there were two competitors that churned off. You know, that net expansion can move around a little bit. How should we think about maybe on a go-forward basis, two customers like that? Should that have like a four-quarter sort of depression on that, or could that in fact pop up a little bit?
Yeah. So I, you know, I think net retention for it has always fluctuated, right? And it fluctuates for a few reasons. One is that the expansion of customers isn't always going to be 100% predictable. And so there's often a cadence of deals that will determine what that looks like. The other side of it is, you know, when you look at how customers' buying behavior, the one thing we have started to see that a characteristic that's starting to emerge is we're doing some larger deals. And when you do those larger deals, what it does is that then four quarters out, it means that it's, you know, you don't have the same growth opportunity because you basically kind of covered that customer for that period of time. So we're seeing some of that. It's not universal, but it's one of those characteristics.
You know, 130 for us, we think is good, right? 132, we think is a good number for us, but we watch closely any changes in those characteristics, but I think it is something that you need to look at over a period of time, given the fact that it has moved around.
And maybe just somewhat related, Jen, for you, there was a lot of talk today about, you know, how the platform is different competitively. And one of the other questions that we always get on, well, gee, competitors are cutting price. They're cutting price. But just maybe a little bit more thought on what impact are you seeing from that? What are customers saying? Clearly, there's a technology gap that you guys have, but maybe a little bit more on how you guys see the pricing reductions from peers.
Sure. I mean, we've always seen the same competitors compete on price, and we have a lot of respect for Atlassian and Splunk and others that are in the market. You can see that our gross margins are holding up nicely, so even as we see competitors trying to drive more intensity, particularly around that price point, we're focusing more and more on value and helping our customers understand the benefits of not just automating your on-call schedule, but actually driving Modern Incident Response and Event Intelligence, so you know, we also, I would say that you have to look at competitors on a segment-by-segment basis because there's far less competition in enterprise than there is in SMB, and mid-market is a bit of a mixed bag. What I'd also just reinforce is it's not a zero-sum game.
We continue to see the vast majority of our customers be greenfield as opposed to we're duking it out with somebody who is in there.
All right. Thanks, Rishi Jaluria, D.A. Davidson. Really appreciate all the detail today. Two questions. First, you know, all the kind of non-DevOps use cases that you talked about today, which were really helpful talking about, you know, customer service. Is that this stuff you're seeing from customers, and does it maybe in your mind make sense to have some of these horizontal specific applications like customer services or financial services, you know, down the line built on your behalf? Or is that more part of the developer ecosystem and app ecosystem story that we might expect to hear more about this week during the Summit? Then just a follow-up on Jennifer. You talked about the AIOps, you know. We've heard New Relic and we've heard Dynatrace talk about AIOps as part of their platform.
So maybe help us understand how that all fits together. Thanks.
Sure, so for your first question, I just wanna make sure I got it in.
About the new use cases.
New use cases.
Whether it's the.
Got it.
Yeah.
So new use cases were originated entirely from the customer base, right? This is like, think of that as true market pull, right? They're using PagerDuty in the developer community, and someone realizes like, "Hey, I got the same problem. It's like real-time. I'm monitoring stuff. I need to figure out how to get my team orchestrated around it," etc., and that's how security happened, right, and the next thing you know, you have tons of SecO ps or DevS ecO ps teams that are using PagerDuty. Datadog, for instance, has long been using PagerDuty for their security teams and have talked on our behalf about it many times. They're very passionate about it, so those use cases are coming from the customer, and we're following that demand signal, and the first thing we'll do is invest in integrations to support those use cases.
So earlier this year, we announced PagerDuty for SecOps, and we have 30 partners in that ecosystem that we integrate to, including Palo Alto Networks and Threat Stack, etc. And a couple of them are speaking on a DevSecOps panel this week. Same thing with customer support. So you're gonna hear us make announcements about integrations and features to support our support agents or customer success agents that will be dealing with customers at the time they're trying to understand what's going on with an issue that's impacting that experience and how we bring all of that together. But then what you'll see a customer support team do is, "Okay. Now I learned how to use PagerDuty for incident response. Could I apply the same technology for to reduce the time it takes to resolve cases?
Could I, could I do it to reduce caseload?" And, you know, the answer is hopefully yes. Some of the weirder ones that like, I mean, using PagerDuty to manage collections, using PagerDuty for order management at the end of closing a software quarter. I mean, we are even doing that.
We do that.
Like these, but we heard about that from another customer.
Yeah.
So it's more about, one, making sure we have the integrations in place, two, making sure that there's a platform for developers to build on top of the platform for use cases that we can't foresee. And then we'll continue to pay a lot of attention to how our customers are leveraging it. I mean, frankly, job one for me is we've gotta get the word out so that people understand these opportunities and value propositions exist. And, you know, that's why I'm excited to have Julie here with us because marketing historically has not been an area we've invested in. It's all been about frictionless product-led growth, sales-led growth so building a stronger marketing capability is an important next step for PagerDuty.
Yeah. And maybe one comment I would add to that, Rishi, just, you know, we don't report on percentages by use case. And part of that is because when someone takes a subscription from us, we don't tell them how they use us, right? It's a horizontal application. They can choose to use us in dev, IT, wherever they like. But what we have started doing is looking at the integrations that they use and also the job roles. And that's starting to help us get a picture. So, for example, if we look at our customer base today, about 15% of our customers are using us in what appears to be a security use case as well as other use cases.
So there's a momentum that's built around that was, you know, was not necessarily as intentional from our side until the last sort of year or so when we were very specific about building the integrations that would be required to support that environment. We're seeing similar trends in terms of customer support, growing number of customers there who are starting to use us in customer support use cases. And that just reaffirms our view in the importance of this being a platform. And then we look at AI.
So you had a question about AIOps.
Yeah.
So, you know, AIO ps used to be called Algorithmic IT Ops, and they conveniently changed it to Artificial Intelligence IT Operations, about a year ago. And I think it's still an emerging category. When we think of AIO ps, what we're talking about is leveraging AI and m achine learning to speed, consolidate, automate the process of dev and IT operations, period, right? And we've been doing that historically with Modern Incident Response through Similar Incidents. The add of Event Intelligence allowed us to bring in far more machine learning capabilities into that space that sees us playing in the event management space as well as some of the predictive analytics space. So the idea, for instance, of recognizing an event storm before it becomes an incident, getting out in front of it, I mean, that's a great example of AI Ops.
Where we're very different is most of the monitoring providers and the Moogsoft and BigPanda of the world is we're doing it with services and people and the dependencies between those two things at the center. So in any other AIOps solution I'm aware of, you can't actually execute action directly from the insight. Whereas in PagerDuty, you do and, in fact, you should look for us to think more and more about how we automate that action, how you go from the insight to automation as opposed to the insight to telling a person or orchestrating work on behalf of a person. So we see them as tightly integrated that you can't really have Event Intelligence and event management without the people action that closes the loop and actually gets the job done.
And that is, in our view, core, a core strategic advantage and core of value prop.
We probably have time for one more question.
Thanks. You made a comment about how your competitors either have correlation or they have orchestration, but they don't have both. Can you flesh that out a little bit? And then specifically related to Opsgenie and VictorOps?
So neither Opsgenie nor VictorOps have correlation engines that I'm aware of. They haven't even had data retention up until the last several months. So, you know, the way I think about that is you have alerting and notification tools that don't correlate and don't have any kind of event management capability. You have event management tools that don't have any orchestration or alerting capability, or at least broad-based alerting capability. And then you have monitoring that might do monitoring in some event or AIOps, but no orchestration. So you've got a whole, and by the way, the philosophy with which we come at this is very different. So we take a whole worldview, right? New Relic started in the app and is trying to move down. Datadog started in infrastructure and trying to move up. We've always looked at the whole picture.
So we're architected to look at a much higher volume and variety of signals than anybody else's, right, from the get-go. And then we come at it from the perspective of resiliency. So all those people I just described, pretty much, with the exception of a couple of competitors, are using PagerDuty for their own reliability practice, right? So they're not necessarily architected for the same reliability that we have. We're not going through any kind of transition from on-prem to the cloud. We were cloud native. We're architected for speed and scalability. We continue to try and stay ahead of the market by leveraging microservices, serverless, etc. It is not easy. Developers don't like to have to manage multiple cloud environments, but we make ours in service to our customers. And Tim probably doesn't like it either.
But even customers like, you know, HashiCorp, Mitchell is speaking tomorrow, in General Session. You know, they're big PagerDuty users, and they're the kind of customer we love because they push us, we push them. I was in seeing Slack CTOs the other day, and we're, you know, talking about all the things that we could be doing differently to support each other and including how we might think creatively about the way you could integrate Slack more tightly into PagerDuty and vice versa because we're also big Slack users at PagerDuty. We know a lot of our customers are.
Okay.
Can I get one more?
Okay. One more. If there's one more question.
There you go, Jonathan.
Sure.
Am I gonna regret this?
No, no.
Okay. Good.
You talk a lot about learning from data and combining the human intelligence in with the actions that are or the signals coming in on the platform. What percentage of the activity on the platform or the actions that people are doing are somehow informed by that versus just the brute force of like, "I did X, Y, and Z and I to solve a problem"? Does that make sense? How much are you actually seeing the leverage and the lift in the business today because of learning from that data?
It's growing progressively and it's growing rapidly. So, even our customers that use our simplest solution are getting some benefit.
Mm-hmm.
In terms of that insight because they're getting some ability to suppress noise that they don't like, some ability to reduce to improve the signal through the noise.
Mm-hmm.
Just in our core product, and then they're getting benefits like through automation and postmortem, etc. The benefits obviously increase dramatically if you're using Modern Incident Response and Event Intelligence.
Mm-hmm.
And likewise, I think the benefits from Analytics to speed your learning also improve. But it kind of depends on what their current product adoption is. But I mean, even in our basic product, you will see some of that benefit.
Okay.
The other thing I would say is the more users you have the platform, the more integrations you have on the platform, the more benefit you get from insight.
And just one follow-up. Are you doing taking data across customers, you know, 'cause they're all using the same cloud services, you know, etc., and integrating that in, or does my insights are they confined to that which has only been learned in my environment?
Totally. Great question. So just to be super clear, we're very careful about data and data privacy and the integrity of the way we leverage data and making sure that data is encrypted, etc. So I just wanna lay that one out there. We don't collect a ton of PII data. We are GDPR compliant. We've killed ourselves to make sure that we do that reasonably well. But to answer your question specifically around how much leverage do customers get from the total data platform as opposed to just their data, a lot, particularly if they use Analytics or Operations Health, and Similar Incidents or because you start to get a benefit from pattern recognition. And in Analytics in particular, you can benchmark. So you actually start to benchmark your capability of your teams versus teams within your industry. You can choose a peer group and do that.
So they get benefit both ways. What you're alluding to, though, is also a future opportunity. And I won't speak about future so I can continue to work here. But what I will tell you is I think there's a lot of opportunity going forward to get more to increase the leverage that our customers get from the total data and total data environment.
Okay. Thank you.
Okay.
All right.
Anything else we need to do?
No, I think, yeah. Well, thank you all for joining us today. It was great to have you here. We were pleased with the numbers. We initially didn't think we were gonna have so many people, but good to see the level of interest. So I appreciated you all being here today. There will be cocktails served out in the library. Sounds like we're playing a game of Clue in the library. So watch out for Colonel Mustard with a candlestick. That could be a problem. But again, thank you very much. Hope it was helpful.
Yeah. I'm gonna just add one more thing too. Last plug. Tomorrow, there are a lot of interesting things happening on the main stage tomorrow, including Jeff Lawson will be joining me, as will Eric Yuan, as will Andre Iguodala. You'll get to see a product demo keynote and a much more detailed demonstration, which I think could be really helpful in just sort of helping to separate for you guys the signal from the noise because one of the things I've learned about this industry. I mean, remember I started my career selling toilet paper and diapers at Procter & Gamble? There's a lot of people using the same words for very different technology, and a lot of people using big buzzy words that but can't actually execute on them without customer success agents and services and systems integrators, etc.
We don't talk about something unless it's in the hands of customers today. So everything you see tomorrow is in GA or has been in design and in a long beta with customers and has already got hundreds of customers on it. And I think that will help you sort of clarify. And then the last thing I'd just like to say is thank you because all of you have asked us a lot of great questions over the course of the last several months. We're new at this. You know, we're just loving this public company thing. You know, the earnings calls. The question, I mean, everybody is trying to compete against us apparently.
But it really does help us to, you know, think about how to continue to do a better job of articulating the value that we're creating, the business that we're building. So keep it coming. Thanks.
Thank you.
Will that be streamed tomorrow?
Somebody in here knows more about that than me. Streaming? Julie?
Is the keynote gonna be streamed live tomorrow for Jen?
If you want it to be streamed live tomorrow, you'll have information on our website as far as the ops as well as on our customer website.
Okay.
But it's way more fun if you just.
More fun if you hear.
People live. There you go. You just hold up the all right. Thanks, guys.
Thank you very much.