SiliconANGLE theCUBESiliconANGLE theCUBE
  • info
  • Transcript
Co-Founder Alex Solomon shares the direction PagerDuty is taking in the next year.
Clip Duration 01:04 / September 24, 2019
Alex Solomon, PagerDuty | PagerDuty Summit 2019
Video Duration: 12:49
search

Announcer: From San Francisco, it's theCUBE, covering PagerDuty Summit 2019. Brought to you by PagerDuty. Hey, welcome back everybody. Jeff Frick here with theCUBE. We're at PagerDuty Summit. It's the fourth year of the show, CUBE's been here for three years, it's amazing to watch it grow. I think it's finally outgrown the Westin St. Francis here in lovely downtown San Francisco. We're really excited to be joined by our next guest. He's Alex Solomon, the co-founder, co-founder and CTO of PagerDuty, been at this over 10 years, Alex, first off, congratulations and what a fantastic event. Yeah, thank you very much, and thank you for having me on your show. So things have changed a lot since we had you on a year ago. This little thing called an IPO, so, I'm just curious, you know, we have a lot of entrepreneurs that watch the show, as a founder, and kind of going through this whole journey, what was that like, what are some of the things you'd like to share from that whole experience? Yeah, it was, it was incredible. The word I'd like to use is surreal. Like, just kind of going through it, not believing that it's real, in a way. And joining by my lovely wife, who came along for this festivities. And just being able to celebrate that moment. I know it is just a moment in time, and it's not, it's not the end of the journey, certainly, but it is a big milestone for us. And being able to celebrate, we invited a lot of our customers, our early customers who have been with us for years, to join us in that celebration, our investors who have believed in us from back in, you know, 2010. Jeff: Right, right. When we're just getting going. And, we just had a great time. I love, overnight success, 10 years in the making, one of my favorite expressions. And it was actually interesting when Jen pulled up some of the statistics around kind of what the Internet was, what the volume of traffic was, what the complexity and the systems are, and it's really changed a lot since you guys began this journey 10 years ago. Oh, it has. I mean back then, like, the most popular monitoring tool's Nagios. And New Relic was around, but just barely, and now it's like DataDog has kind of taken over the world, and the world has changed, we're talking about not just microservices, but containers, and serverless, and the cloud, basically. That's the, kind of, recurring theme that's changed over the last 10 years. But you guys made some early bets, you made bets on cloud, you made bets on dev ops, you made bets on automation. Those were pretty good, those turned out to be pretty good places to put your chips. Oh yeah. (laughs) Right place, right time, and, you know, some, some, experiential stuff, and some, just some, raw luck. Right. All right, well let's get into it, tell me some of the product announcements that are happening today. What are some of the things you're excited to finally get to showcase to the world? Yeah, so one of the big ones is related to our Event Intelligence release. We launched the product last year, a few months before Summit, and this year, we're making a big upgrade. We're announcing the big upgrade to the product, where we have related incidents. So, if you're debugging a problem, and you have an incident that you're looking at, the question you're going to ask is, is it just my service, or is there a bigger widespread problem happening at the same time? So we'll show you that very quickly. We'll show you are there other teams impacted by the same issue? And we'll, we actually leverage machine learning to draw those relationships between ongoing incidents. Right. I want to unpack a little bit, kind of, how you play with all these other tools. We're, you know, we were just at SuperLogic a week or so ago, they're going to be on later. They're a partner. And people, I think it's confusing. There's, like, all these different types of tools, and you guys partner with them all. I mean, the integration list that you guys have built. I wrote it down, it's ServiceNow, it's Splunk, it's those ZenDesk, it goes on and on and on. So, explain to folks how does the PagerDuty piece work within all these other systems? Sure, so, I would say we're really strong in terms of integrating with monitoring tools. So any sort of tool that's monitoring something and will emit an alert, when something goes down, or an event, when something's changed, we integrate it, and we have a very wide set of coverage with all of those tools. Think, your, like, DataDog, AppDynamics, New Relic, even old school Nagios, of course. Jeff: Right, right. And then we've also built a suite of integrations around all the ticketing systems out there, so, ServiceNow, Jira, Jira ServiceDesk, Remedy as well. We also now built the suite of integrations around the customer support side of things. So, that would be ZenDesk and SalesForce. That's interesting, Jen had a good example in the keynote and kind of, in this multi-system world, you know, where's the system of record? Cause it used to be everybody wanted to be the system of record, they wanted to be the single pane of glass. But it turns out, that's not really the answer, there's different places for different solutions to add value within the journey, within those other applications. Yeah, absolutely, I think the single pane of glass vision is something that a lot of companies have been chasing, but it's really hard to do, because, like for example, New Relic, they started an APM and they got really good at that, and that's kind of their specialty. DataDog's really good at metrics, and they're all trying to converge and do everything, and become the one monitoring solution to rule them all. But they're still the strongest in one area. Like Splunk for logs, and New Relic and AppDynamics for APM, and DataDog for metrics. And I don't know where the world's going to take us, like, are they, is there going to be one single monitoring tool, or are you going to use four or five different tools? My best guess is you're, we're going to live in a world where you still going to use multiple tools, they're each going to do something really well, but it's about the integrations. About really bringing all that data together. Right. >> That's, from early days, we've called PageDuty the Switzerland of monitoring, cause, we're friends with everyone, we're partners with everyone, and we sit on top, work with all of these different tools. I thought her example she gave in the keynote was pretty, it's kind of illustrative to me. She was talking about, you know, say your cable's down, and, you know, you call Comcast, and it's a ZenDesk ticket, but you know, then that integrates potentially with the PagerDuty piece, and says, hey, we're, you know, we're working a problem, you know, a back hoe clipped a cable, down your street, and so to take kind of that triage and fix information, and still pump that through to the ZenDesk person who's engaging with the customer, to actually give them a lot more information. So the two are different tracks, but they're really complementary. Absolutely. And that's part of the incident life cycle, is letting your customers know and helping them through customer support, so that the support reps understand what's going on with the systems, and can have an intelligent conversation with a customer so that they're not surprised. Like a customer calls and says you're down. Oh, good to know, no you want to know about that. Which I think Jen also said, that's how most people find out, unfortunately. Another thing that, that struck me was this, this study that you guys have put together about unplanned work, the human impacts of always-on world. You know, we talk a lot in tech about unplanned maintenance, and unplanned downtime of machines, whether it's a computer or military jet, you know, unplanned maintenance is a really destructive thing. I don't think I've ever heard anyone frame it for people, and really to think about, kind of, the unplanned work that gets caused by an alert, a notification, that is so disruptive, and I thought that was a really interesting way to frame the problem, thinking about it from an employee-centric point of view to reduce the nastiness of unplanned work. Absolutely, and that's, it's very related to that journey of going from being reactive, just reacting to these situations, to becoming proactive and being able to predict and address things before they impact the customer. I would say, it's anywhere between 20 to 40, or even higher percent of your time, maybe looking at software engineers, has been on this unplanned work. So, what you want to do is you want to minimize that, you want to make sure that there's a lot of automation in the process, that you know what's going on, that you have visibility, and that the easy things, the repetitive things, are easy to automate, and the system can just do it for you, so that you, you focus on innovating and not on fixing fires left and right. Right, or if you did have to fix a fire, you'd at least get the fire to the right person who's got the right tool to fix it. I was just, you know, we see that all the time, and it's especially in early days, for triage, you know. What's happening, who did it, who's the right people to work on this problem, and you guys are putting a lot of the effort into AI, and modeling, and your 10 years of data history, to get ahead of the curve in assigning that alert, or that triage when it comes across the transom. Yeah, absolutely, and that's another issue, is not having the right ownership, getting people notified when they don't own it, and there's nothing they can do about it. Like, the old ways of sending the alert to everyone and having a hundred people on a call bridge, that just doesn't work anymore. Because they're just sitting there, and they're not going to be productive the next day at work, because they're sitting there all night, just kind of waiting for something to happen. And that's kind of the old way of lack of ownership, just blast it out to everyone, and we have to be a lot more targeted and understand who owns what. And what's, which systems are being impacted. And only getting the right people on the, on the call, as quickly as possible. The other thing that came up, which I thought, you know, probably a lot of people aren't thinking, they only think of the fix-it guy that has to wear the pager. But there's a whole lot of other people that might need to be involved. Before, we talked about the Comcast example, the people interacting with a customer, maybe senior executives need to be informed. Maybe people that are, you know, on the hook for the SLA on some of the softer things. So, the assembly of that team goes, who needs to know what, goes well beyond just the two or three people that are the fix-it people. Right, and that's, that's actually tied to one of our announcements at Summit, Business, our Business Response products. So, it's all about yes, we notify the people who are on call, and are responsible for fixing the problem, you know, the hands on keyboard folks, the technical folks, but we've expanded our workflow solution to also loop in stakeholders, so think like, executives, business owners, people who maybe they run a division, but they're not going to go on call to fix the problem themselves. But they need to know what's going on, they need to know what the impact is, they need to know, is there revenue impact, is there customer impact, is there reputation or brand impact, to the business they're running. Yeah, which is another thing you guys have brought up, which is so important. It's not just about fixing, fixing the stuck server, it is, what is the brand impact, what is the business impact, and it's a much broader conversation which is interesting, to pull it out of, just the poor guy and the pager, waiting for it to buzz, versus now the whole company really being engaged to what's going on. Absolutely, like, connecting the technical, what's happening with the technical services and infrastructure, to what is the impact on the business if something goes wrong. And how much, like, are you actually losing revenue? There's certain businesses like e-commerce, where you could actually measure your revenue loss on a per minute or per five minute basis. Right, pretty important. Yeah. All right, Alex, so you talked about the IPO as a milestone, it's fading in the, it's fading in the rearview mirror. Now you're on the 90 day shop clock, so, you got to keep moving forward. So as you look forward now from your CTO role, what are some of your priorities over the next year or so, that you kind of want to drive this ship? Absolutely, so I think just focusing on making the system smarter, and make it so that you can get to that predictive holy grail, where we can know that you're going to have a big incident before it impacts your customers, so you can actually prevent it, and get ahead of it based on the leading indicators. So if we've seen this pattern before, and last time, it caused, like, an hour of downtime, let's try to catch it early this time so that you can address it before it impacts the customers. So that's one area of investment for us. And the other one I would say is more on the, the real time work outside of managing software systems. So, security, customer support, there's all of these other use cases where people need to know, like, signals are being generated by machines, people need to know what's going on with those signals, and you want to be proactive and preventative around it. Like think a factory with lots and lots of sensors, you don't want to be surprised by something breaking. You want to, like, get proactive about the maintenance of the systems so you don't have that, you know, like, say a multi-day outage at a factory can cost maybe millions of dollars. All right, well, Alex, thanks a lot again. Congratulations on the journey. We're enjoying watching it, and we'll continue to watch it evolve. So thanks for coming on. All right, he's Alex, I'm Jeff, you're watching theCUBE. We're at PagerDuty Summit 2019 in downtown San Francisco. Thanks for watching. We'll see you next time.