Latest / IT Infrastructure as a Conversation / Observability without Overload: Lessons from Chronosphere’s Field CTO
Transcript
- 0:00One of the phrases I've heard on this podcast
- 0:02very often is, data is the new oil. But let's
- 0:06be honest, it's more like the new oxygen. Especially
- 0:09with AI, because without data there is no AI.
- 0:13And your business can't function without it.
- 0:16NordLayer, they ensure that your data moves securely
- 0:20between people, apps and offices through encrypted
- 0:24tunnels. Encrypted tunnels and strict access
- 0:28controls. and it keeps sensitive information
- 0:31out of reach from prying eyes, but gives you
- 0:35the full visibility over how your network is
- 0:37used. Now, I've heard many stories on here where
- 0:41a single exposed endpoint quickly turned into
- 0:44a company -wide breach, but Nord layer, they
- 0:48will make sure that that doesn't happen to you,
- 0:50so if... If this interests you, please go to
- 0:53nordlair .com slash tech talks daily. And if
- 0:57you enter the special code tech daily dash 28,
- 1:01you'll also get 28 % off. So please go check
- 1:05that out. Let me know your thoughts, but right
- 1:07now let's get today's guest on. What happens
- 1:12when observability stops being a tool conversation
- 1:15and starts becoming a business conversation?
- 1:19Well today I'm joined by someone who has lived
- 1:21that shift in the real world. His name's Bill
- 1:24the Heinlein and he's the Field CTO at Chronosphere.
- 1:29And before that he spent years inside highly
- 1:32complex environments at United Airlines where
- 1:36telemetry volume, cardinality and scale were
- 1:39much more than just buzzwords. and throughout
- 1:42his career he has seen first hand how a collect
- 1:45everything mindset turns data lakes into data
- 1:48landfills and digital clutter and digital hoarding
- 1:52and how alert fatigue can drain great teams and
- 1:55even a hero culture can quietly burn out your
- 1:59best people. But he does believe observability
- 2:03should be intentional, service -focused and tied
- 2:06to SLOs from the very first sprint. So if your
- 2:11CFO knows observerability for the cost line rather
- 2:14than the value that you're delivering, he's got
- 2:17some thoughts that you're going to want to hear
- 2:19today. So you can expect practical examples,
- 2:22a few scars and war stories from late night pager
- 2:25duty, and most importantly, clear steps that
- 2:28you can use tomorrow. So if you're ready to swap
- 2:31noise for insights and turn your monitoring into
- 2:35a momentum, you're going to love this one. But
- 2:38enough from me. Let's get our guest onto the
- 2:40podcast now. So thank you for joining me on the
- 2:44podcast today. Can you tell everyone listening
- 2:47a little about who you are and what you do? Sure.
- 2:51I'm Bill Heinlein. I'm the field CTO at Chronosphere.
- 2:55We're an observability platform. I have been
- 2:59in the tech industry more time than I care to...
- 3:03to say, but a little more than 25 years, a lot
- 3:06of that time spent in the airline industry. And
- 3:10most recently, before joining Chronosphere, I
- 3:13was Director of Observability at United Airlines.
- 3:18So a lot of large company experience, a lot of
- 3:23battle wounds over the years. So I can go on
- 3:27for a long time about observability and other
- 3:30topics. So happy to be here. And that is exactly
- 3:34why I invited you on the podcast to join me today
- 3:36because every day we take a different subject,
- 3:39demystify it, talk about it in a language everyone
- 3:41can understand, make everyone realize we're all
- 3:44battling with the same problems and how we can
- 3:46overcome them as well. So from all the conversations
- 3:49you've had in your current role, United Airlines,
- 3:51everything throughout your career, what is that
- 3:53number one observability challenge that organizations
- 3:57are grappling with right now? What is it? Well,
- 4:00I mean, cost is the easy answer, but, you know,
- 4:05cost isn't the root cause, right? Cost happens
- 4:08because, you know, over the years, we have all
- 4:13subscribed in one way, shape or form to this
- 4:15collect everything model, right? And it's filled
- 4:19with a little bit of false hope and really creates
- 4:22what I've started calling a data landfill. You
- 4:25know, we wind up instead of a lake, you get a
- 4:27landfill. A lot of data in there, not a lot of
- 4:31value, hard to measure the value. And there's
- 4:35a challenge with measuring the usage and the
- 4:38value of the data that you're collecting. So
- 4:41it's cost, but it's because there is this collect
- 4:46everything strategy rather than a more curated
- 4:48telemetry strategy. And things like logs are
- 4:52some of the biggest offenders. It's really getting
- 4:57down that telemetry to things that are collected
- 5:01with intent and really bringing down the noise
- 5:05to signal ratio as a result. And we've all been
- 5:09on somewhat of a journey over the last few years
- 5:12I find myself The older I get that everything
- 5:15happens in cycles and we've everyone left on
- 5:18-prem data centers. That was old school We're
- 5:20all going to the cloud now these rumors of will
- 5:22coming back again. So I'm curious How do the
- 5:26traditional rules of observability? How do they
- 5:28change in cloud native environments? I would
- 5:31argue that observability has never been traditional.
- 5:34It's always had to evolve. I think I don't I
- 5:38think technology in general is hard to categorize
- 5:41as traditional. If it is, then it's probably
- 5:43something we talk about, you know. years ago
- 5:47that nobody's doing anymore. But in short, we've
- 5:51had to move from predictable telemetry volumes.
- 5:55In the world of monolithic applications and they
- 6:00lived on a few servers in a data center, they
- 6:02admitted a predictable amount of telemetry. You
- 6:05were monitoring a less complex infrastructure
- 6:09and you understood when it needed to scale. You
- 6:14built that in early on by predicting what your
- 6:17scale was going to be, and then you knew what
- 6:20to predict as far as the volume of telemetry
- 6:23you were dealing with. And telemetry wasn't as
- 6:25sophisticated back then, right? We were still
- 6:27in the era of early APM. Some teams were still
- 6:32just managing component monitoring, right? Is
- 6:37the database up? Is the server running? Am I
- 6:39out of disk space? Very, very basic. stuff that
- 6:42nerds like me used to write a lot of scripts
- 6:44in Perl for years about. But then it moved to
- 6:48cloud and everything exploded, right? Because
- 6:51suddenly we have servers that last for seconds
- 6:54while they run and then die after they're done.
- 6:56And we have scale that went from a few servers
- 7:00to tens of thousands, depending on the size of
- 7:04the environment and the scale at which you're
- 7:06operating at the time. And the telemetry that's
- 7:10all there is deeper and the cardinality just
- 7:14explodes. And so it's really what's changed is
- 7:18the volume of telemetry and those more traditional
- 7:23APM tools, the more traditional and early observability
- 7:27tools really struggle with keeping up with the
- 7:30scale of that and doing it in a way that's cost
- 7:33effective. I also think the rise in cloud costs
- 7:38has served as somewhat of a virtual intervention
- 7:40of sorts to face up to our data hoarding ways.
- 7:44From what you've seen here, what issues do the
- 7:47collect everything data collection practices
- 7:50bring when the first step is gaining control
- 7:53or attempting to gain control? Well, I think
- 7:57there's certainly an opportunity for us to do
- 7:59a show on your favorite binge watching network
- 8:04about data hoarding, you know, similar to people
- 8:08hoarding in a storage unit somewhere. But the
- 8:13collect everything, we already mentioned cost,
- 8:15obviously. But, you know, if you collect everything,
- 8:18you're not curating. And the problem is, is that
- 8:21I think we all got very excited about AI and
- 8:25thought that AI could answer everything. But
- 8:28it's still a classic garbage in and garbage out.
- 8:30If you've got telemetry types coming from multiple
- 8:34sources, different observability suppliers, maybe,
- 8:41Lord knows a lot of different infrastructure,
- 8:43whether it be cloud or on -prem, it's all very
- 8:46different. It has varying levels of enrichment.
- 8:49You throw in application logs that anybody can
- 8:52write and that have varying levels of enrichment
- 8:54and may be missing key things that don't matter
- 8:57to the developer who wrote it. but may matter
- 9:00to other people who look at it. So you've got
- 9:03this blood of data here that's really hard to
- 9:08correlate. And I always say that if you can't
- 9:13look at a set of data on some paper and correlate
- 9:17it together, AI isn't going to be able to do
- 9:20that either. It needs something to chew on. And
- 9:23that is where the collect everything starts to
- 9:26bite you. Because you've got all this data that
- 9:29you can't correlate easily. And when you're in
- 9:32the midst of that 3 AM call that we all love
- 9:36so much, that you wind up digging in and following
- 9:42a metric that spikes suddenly. But oh, guess
- 9:44what? It doesn't really. correlate to the problem
- 9:47you're working on. And that, oh, by the way,
- 9:49that looks like that happens every third Thursday
- 9:51for some reason, because somebody's running a
- 9:54job over here and it's unrelated, but you wasted
- 9:5620 minutes. So it's that noise to signal ratio
- 9:59that's killing your MTTR and you're spending
- 10:02a lot of money. And your CFO probably knows what
- 10:05observability is for the wrong reason. The cost
- 10:08instead of the value. Yeah. 100%. And of course,
- 10:12we also throw in operational burden and the dreaded
- 10:16alert fatigue, also sources of burnout and frustration
- 10:20for IT teams. But what real impact does this
- 10:23have on businesses and what can IT leaders do
- 10:26to address it? Something we've been talking about
- 10:29for years, but any way forward here? Yeah, I
- 10:33think and I mean, look, I had big teams. a team
- 10:38of more than 500 at United. And so burnout was
- 10:41something that I worried about every day. You
- 10:44know, because you get great people and through
- 10:48no fault of anybody's, those great people often
- 10:50wind up being the people that that you build
- 10:53into heroes. Yeah. And a hero culture kills your
- 10:57best people. They burn out. And then the next
- 11:00thing you know, they have they've secretly turned
- 11:02on the open to work flag on LinkedIn. Right.
- 11:05And then you've got opportunity cost problems
- 11:09and all sorts of things. But the real impact
- 11:11there is you've hired these engineers, whether
- 11:15they be traditional engineers in the data center
- 11:18or cloud engineers or software engineers, you've
- 11:21hired these brilliant people to make money, to
- 11:25build products that are going to displace your
- 11:27competitors or your market. And this operational
- 11:32burden that happens because on -call exists,
- 11:36and you're not doing it efficiently, takes away
- 11:39that innovation. And so all the capital investment
- 11:42that you're making that your investors are happy
- 11:44that you're making in all this technology to
- 11:47disrupt the marketplace suddenly starts to look
- 11:49more like some cost because you're not getting
- 11:51anywhere with it. But at the same time, your
- 11:54operating costs are on the rise because your
- 11:57teams are spending all this time keeping the
- 11:59engine running, so to speak. Sorry. back to airline
- 12:03metaphors, but to drive or to keep things running.
- 12:08Right. And so they're not focused on innovating
- 12:10and that context switching that they have to
- 12:12do constantly is a real thing, too. I wrote an
- 12:15article and quoted an APA study not too long
- 12:18ago that talks about context switching can have
- 12:22as much as a 40 percent reduction in productivity
- 12:27for an individual because of that. I mean, imagine,
- 12:30right? You're in the zone, you're coding, hopefully
- 12:33you're not vibe coding too much, and you're really
- 12:36into it, you've got a great thing going, and
- 12:38all of a sudden, I'll date myself, the pager
- 12:41goes off, and then you've got to completely stop
- 12:43what you're doing, and then focus over here on
- 12:46code that likely somebody else wrote, but you've
- 12:48got to figure out why it's broken, while you
- 12:50listen to a group of people ask you, is it fixed
- 12:53yet? Then you come down from that and then you
- 12:57go back and you try to pick up where you left
- 12:58off. The business impact, I mean, certainly is
- 13:02burnout for that individual, but from a business
- 13:06perspective, you're not innovating. You're getting
- 13:09left in the dark. You're losing great people.
- 13:12There's a lot of business impact that leaders
- 13:15really have to consider and it's a real thing.
- 13:18That's where observability can either hurt or
- 13:21help you there. you collect everything, you're
- 13:24probably on the pathway to hurt. You're in good
- 13:29company there. I still get flashbacks from those
- 13:32days of that pager going off, man. Don't get
- 13:34me started there. There's a whole world of pain
- 13:36opened up that I've still not dealt with, I think.
- 13:39But seriously, moving beyond tools, what does
- 13:42a strategic shift in today's IT infrastructure
- 13:45management look like at a high level? And what
- 13:48kind of conversations need to happen to kickstart
- 13:51that change? Because it's very difficult to...
- 13:54get all the stakeholders and get everyone in
- 13:56a room and kickstart the changes required and
- 13:58get everybody on board. But what needs to happen
- 14:00here? I think it's just fundamental change at
- 14:03the top, right? And by the top, I don't mean
- 14:06necessarily the CIO or the CDO or the CTO. I
- 14:09mean at the top level of the executives of the
- 14:12organization, including those people. IT has
- 14:17to deliver strategic value. In this world of
- 14:21cloud costs and the cost of IT and the cost of
- 14:25IT labor and all of that, you have to be able
- 14:27to justify the value of your organization because
- 14:30IT organizations that don't provide value, guess
- 14:33what? They get outsourced. Those outsource teams
- 14:36don't provide the great value that is usually
- 14:39hoped for either. That's some cases they do.
- 14:44I think what you have to do is really start with,
- 14:48it's not so much how you manage just infrastructure.
- 14:51I mean, I know that's the primary focus of your
- 14:55conversations on this podcast, but I think it's
- 14:58at a higher level of making sure that your IT
- 15:02organization and specific, your IT leader has
- 15:05a seat at the table and that they are participating
- 15:08in those strategic discussions and advising rather
- 15:12than order taking. Then it's easier to, number
- 15:20one, justify, okay, we're spending a lot on Cloud.
- 15:22Well, we made together this strategic discussion
- 15:25or decision that we're going to invest in Cloud
- 15:27because we need to scale, we need to be able
- 15:31to support things in this region, whatever your
- 15:34justifications are. But you've done that as a
- 15:36collective group rather than going off and trying
- 15:40to deliver what you think the business wants.
- 15:44and then coming back and then being questioned
- 15:46whether you're raising the value or not. And
- 15:49in a world where all these systems are constantly
- 15:52scaling and shifting and I'm curious, how can
- 15:56teams ensure that they're collecting the right
- 15:58data, not collecting everything, collecting the
- 16:00right data, making that shift to make informed
- 16:03decisions without being overwhelmed by the volume?
- 16:06I mean, I know there's an argument that AI will
- 16:08make that volume a little bit easier to digest,
- 16:11but anything you're seeing here? I think I joke
- 16:18sometimes that when people get so excited about
- 16:20AI, I say it's the new gluten -free, right? We
- 16:23put AI on a lot of things. At one point, I think
- 16:27I saw that gluten -free was on a laundry detergent
- 16:31and I scratched my head. I'm like, why does that
- 16:33matter? But I think that we've got to go back
- 16:38to the reality of collecting data with intent.
- 16:43It goes back to business value. If I work with
- 16:47a team in my past, they get wrapped around this
- 16:50and it goes back to they'd want to measure this
- 16:53component and that component instead of thinking
- 16:55about the service they're offering. I'll step
- 16:58back and say, tell me five reasons this feature
- 17:02exists in your product. Why did the product owner
- 17:05come to you and say, I want this to happen, code
- 17:08this? Then what does good look like? And I promise
- 17:12you, if you focus on traffic, error rate, latency,
- 17:17and saturation, just as a starting point, your
- 17:20leaps and bounds ahead of where you'd be if you
- 17:22started to see, is that server up? Is it up now?
- 17:25Was it up 10 minutes ago? And instead, looking
- 17:27at, is my service at the level I hoped? And build
- 17:31an SLO, spoiler alert, at the beginning, instead
- 17:35of at the end. I've been just as guilty as anybody
- 17:38else in this space. Five or six years ago, I
- 17:42started asking myself, why do we wait until after
- 17:46the release to think about an SLO? Why aren't
- 17:50we thinking about that in the first sprint? And
- 17:54then that informs metrics that you might need
- 17:57to measure that SLO. It informs how you can build
- 17:59an error budget. And then you get to a point
- 18:02where you can look at things with value, and
- 18:07you know what you specifically need to collect
- 18:09rather than just throwing everything in a pot
- 18:11and crossing your fingers. And as things grow
- 18:18and as you're scaling, that becomes more and
- 18:20more. It's got to be repeatable. It's got to
- 18:22be frictionless for a developer. Observability
- 18:24has to be a utility for people, not a burden.
- 18:28It has to be a first -class citizen. And I think
- 18:30observability is dealing with the sentiment in
- 18:35most organizations that that cyber had to deal
- 18:39with, you know, 10 years ago when we were really
- 18:43doubling down in organizations on cyber because
- 18:45it was becoming a real significant threat for
- 18:48everybody. Right. Developers had to do it. But
- 18:51it was another thing I got to do. And I got this
- 18:53giant delivery cycle observability sort of in
- 18:55that spot right now. We got to make it a first
- 18:57class citizen, too. And that includes making
- 19:00it frictionless and making it repeatable and
- 19:02doing that by simplifying the approach of how
- 19:05you start monitoring. And we've both jokingly
- 19:09shared stories of our past and that page are
- 19:11going off when we didn't want it to. But I think
- 19:13for many teams, they've almost been conditioned
- 19:16to operate with a reactive mindset after years
- 19:19of firefighting, fixing P1s, et cetera. So what's
- 19:23the best way to foster that cultural shift that's
- 19:25required to move towards a much better and stress
- 19:29-free, proactive, data -driven decision -making?
- 19:32It almost sounds like a utopia, me saying this
- 19:35out loud after years of doing the same. What
- 19:38needs to happen here? I joke regularly that I
- 19:42generally talk in euphoric sense, you know, like,
- 19:44or utopic sense rather than, you know, reality.
- 19:48But it is my opinions are based on reality and
- 19:51repeated mistakes on my own part. But I think
- 19:55for us to get people to shift, they have to understand
- 19:58the power of data and the data that's available
- 20:00to them. I look at a histogram. The power of
- 20:08a histogram that shows you things like performance
- 20:11or error rate from one minute to the next can
- 20:14help you really think about ways to be more predictive.
- 20:18If you're using SLOs, if you've decided you're
- 20:21going to shift your mind and do an SLO first
- 20:25and you do it in the sprint and lo and behold,
- 20:28you go to production, you've got a good SLO and
- 20:30you've got good error budget, even if it's really
- 20:32simple, You're a step ahead and now you're saying
- 20:37okay I don't have to react and wait till somebody
- 20:40calls me because something breaks or maybe I'm
- 20:43a step ahead and and my tools tell me something
- 20:45broke before the customer but So many of the
- 20:50things that I saw that were these big Newsworthy
- 20:54events not just at my former employer, but it
- 20:58in general right Started out as small paper cuts
- 21:03hours, days before. And if there were little
- 21:06data points to tell you, like, hey, you're starting
- 21:09to burn your error budget a lot more than you
- 21:11were. That's the easiest story to tell to get
- 21:15people to think about how to go from it and get
- 21:18away from, well, my app's up. Nobody's calling
- 21:20about it, right? I can log in. It's a shift in
- 21:24mentality and shift to get them to realize that
- 21:28Look, you got woken up at three, but this started
- 21:30at like 10 in the morning while you were having
- 21:32coffee. Like, you know, what if, what if it interrupted
- 21:36your coffee and started to sleep? Yeah. Oh, now
- 21:39that's progress, but a hundred percent. I think.
- 21:43Thanks to pesky, shiny emerging technologies
- 21:46like, yes, AI, agentic AI or whatever, the next
- 21:50big thing is we seem to have a lot of stalled
- 21:54projects, things caught in pilot. So as a result,
- 21:57leadership teams now are putting an extra special
- 22:00caution on all tech projects. Everything's got
- 22:02to be about ROI, measurable difference, quite
- 22:06rightly too. So how can organizations effectively
- 22:09measure the ROI of investing in a new observer?
- 22:12strategy, especially when benefits aren't always
- 22:16tangible at first or may appear that way. And
- 22:19I suspect this is an area you're passionate about
- 22:21too. I am. And it goes back to my line. If your
- 22:27CFO knows what observability is because of the
- 22:30cost rather than the value, you've got a problem.
- 22:33To get you there, to get your organization there,
- 22:36to realize that value, observability should underpin
- 22:39your business objectives. Right. You don't just
- 22:43decide. I mean, maybe if you're a nerd like me
- 22:46and you see what the cool things are, then you
- 22:48might think, yeah, absolutely. We should do observability.
- 22:51And you may be not thinking about how it's tying
- 22:53to objectives at a higher level. Right. It's
- 22:56cool. I had a leader years ago tell me cool is
- 23:00not a business case. It's not. But. If I can
- 23:06tie observability back to, if I'm an e -commerce
- 23:09company, I care about how quickly you can buy
- 23:11widgets, I care about conversion rate, I care
- 23:13about how quickly you go through the funnel,
- 23:17right? And I care about things that interrupt
- 23:19that funnel and make you abandon your shopping
- 23:21cart. Observability helps you get there. If you
- 23:25design observability for that application to
- 23:28say, I care about response time and my services,
- 23:31I care about how long it takes somebody to be
- 23:33presented with the 40 widgets that I sell and
- 23:37pay for it and all of that, then I'm underpinning
- 23:40a business objective that is likely sell more
- 23:42widgets or increase my conversion rate by X percent.
- 23:46If observability underpins your business strategy,
- 23:49it's easier for people to swallow. And then instead
- 23:52of measuring with MTTR, which typically only
- 23:57technology leaders care about, then you're selling
- 24:01it with the idea of, hey, look, we drove better
- 24:05performance, which you, Mr. and Mrs. Product
- 24:07Owner care about because it's driving better
- 24:11conversion rate and therefore better revenue
- 24:12based on what your product is doing. To bring
- 24:18everything that we've talked about here today
- 24:20back to your work at Chronosphere, are you able
- 24:24to share maybe a real world example of how a
- 24:26strategic observability plan has actually helped
- 24:29an organization move from that reactive firefighting
- 24:32model to be more proactive and innovative one
- 24:35and get that over the line? Anything you can
- 24:38share on that? You don't have to mention any
- 24:39names. I think it'd be useful. Well, I'll tell
- 24:43you, in my former life, we were early embracers
- 24:48of APM, right? And we were in a state where there
- 24:51were still a few folks that were saying, you
- 24:53know, nobody's calling, you know, nobody's calling
- 24:56the service desk, nobody is, you know, there's
- 24:59no problem, I can log in kind of a thing. They
- 25:00were still in that mode with some of the teams.
- 25:04Getting people to understand the value of, again,
- 25:09that histogram, getting people to be able to
- 25:11understand that A lot of the problems that keep
- 25:16us on phone calls for hours in the middle of
- 25:19the night started with a paper cut at seven in
- 25:23the morning, and maybe seven in the morning on
- 25:26Wednesday, and it's Sunday morning. Changing
- 25:32that perspective and showing that this data is
- 25:36possible together, it's not difficult together,
- 25:39and then when it's there, It gives you a lot
- 25:42of great things. And one of the best ways is
- 25:46to help people. I call it mean time to innocence.
- 25:52Right. Hey, look, I know it's not my problem
- 25:55and I know it's not because here's data. Right.
- 25:58Here's my service. It's calling this. And oh,
- 26:00guess what? It's your service because I'm calling
- 26:03you. And every other third time I call you, it
- 26:06fails. You're killing me. Right. So we did that.
- 26:12We did that. And I was proud to say that some
- 26:15of the earliest adopters of APM in our organization
- 26:18were the NetOps team members, because they got
- 26:22tired of being blamed for it being the network,
- 26:24because it's always the network, right? That
- 26:26was always a running joke in every organization,
- 26:28I'm sure. But these guys and gals would get on,
- 26:32look at the tools and say, it's not the network.
- 26:35And as a matter of fact, it's this service right
- 26:37here that keeps failing. And it started right
- 26:39when people started calling. So it's, it's, there
- 26:45are a lot of, I can probably think of hours of
- 26:48examples of things like that, both in my current
- 26:51or with my former employer. And as I talk to
- 26:53customers in my current role. And just listening
- 26:56to that story reminds me people were so fiercely
- 26:59defensive back in the day on things like that
- 27:01with, it almost felt like the Spider -Man meme
- 27:04with the network guy pointing at the apps guy
- 27:06and the apps guy pointing at the, you know, everyone's
- 27:08pointing at each other. But if an IT is listening
- 27:12to a conversation today, maybe they want to take
- 27:14one actionable step tomorrow too. begin to improve
- 27:18their observability posture, what would you recommend
- 27:21they do? I appreciate it's not as simple as that,
- 27:24but if they want to take that first baby step,
- 27:26what should they do? One of my favorite questions
- 27:28when I walk into a customer as we're talking
- 27:30to them is, why did you buy observability and
- 27:33how is it supporting your business goal? Yeah.
- 27:37I mean, and if the answer is, I don't know, or
- 27:41they stumble for it, then they probably just
- 27:44need to take a step back and think about it.
- 27:49So many people jumped into observability because
- 27:51they needed something, and it was an evolving
- 27:56technology and evolving tool. APM promised some
- 28:01really amazing things easily, and as the cloud
- 28:04changed things, that got more difficult in it.
- 28:08diminished a little bit and people had to work
- 28:10harder to get it. And so it's even more important
- 28:13to have that strategy. And that strategy starts
- 28:15with, why am I doing this? I'm doing this so
- 28:18I can sell more widgets. I know I'm selling more
- 28:21widgets because I'm driving better conversion.
- 28:24I'm driving better conversion because my performance
- 28:26is better. And I know that if I change my performance
- 28:29by X number of milliseconds, I drive my conversion
- 28:33up by X percentage. I think it's really sitting
- 28:39back and asking yourself, why am I doing this?
- 28:41Because somebody is going to ask you. Don't wait
- 28:44till your CFO asks you. And I think that is a
- 28:47powerful moment to end on today. But anyone we've
- 28:50inspired today, maybe they want to take that
- 28:52first baby step into improving that observability
- 28:55posture. Where can they find you or your team
- 28:58and more information on anything we talked about
- 29:01today? Where should they go? Certainly, you can
- 29:04find me on LinkedIn, and I'm always, as you can
- 29:07tell, happy and passionate to talk about observability.
- 29:10And part of my role as the CTO at Chronosphere
- 29:14is not to sell, but to guide and help folks with
- 29:19how to do things like an observability strategy.
- 29:21So you can also find me there. We're at chronosphere
- 29:24.io. And again, we're a cloud -native observability
- 29:28company focused on giving you control of your
- 29:30data. rather than drowning in it and not understanding
- 29:35what your telemetry is doing for you. Love that.
- 29:38So I will add links to absolutely everything.
- 29:41Make it easy for people to find. Thank you. We
- 29:43covered so much today from observability challenges
- 29:46that organizations are grappling with, collecting
- 29:49the right data to make informed decisions. As
- 29:53an XIT guy measuring the ROI of investing in
- 29:55a new observability strategy, the old belts and
- 29:58braces approach of you can only improve what
- 30:01you measure. So great to hear. Right. Just thank
- 30:03you for shining a light on this today. Thank
- 30:06you. Thanks for the time. Now, before we wrap,
- 30:09what will you change this week to collect with
- 30:11intent rather than just collecting everything?
- 30:15I think Bill left us with a simple challenge
- 30:17that almost doubles as a compass. Start with
- 30:20why your product exists. Define what good looks
- 30:24like. Set those SLOs early and let your choices
- 30:28guide the data that you're gathering. And this
- 30:31is how observability can earn a seat at the strategy
- 30:34table and how teams reclaim time for building,
- 30:37not just wasting their days firefighting. So
- 30:41if this sparked an idea, please connect with
- 30:43Bill on LinkedIn, check out their website. And
- 30:45if you've got anything you'd like to share with
- 30:47me, techtalksnetwork .com, please let me know.
- 30:50But that's it for today. So thank you for listening
- 30:53as always and I'll speak with you all again tomorrow