Easy Agile Podcast Ep.16 Enabling high performing agile teams with Adaptavist

In this insightful episode, we dive deep into one of the most common frustrations in engineering and dev teams: retrospectives that fail to drive meaningful change. Join Jaclyn Smith, Senior Product Manager at Easy Agile, and Shane Raubenheimer, Agile Technical Consultant at Adaptavist, as they unpack why retrospectives often become checkbox exercises and share practical strategies for transforming them into powerful engines of continuous improvement.

Want to put these insights into practice? We hosted a live, hands-on retro action workshop to show you exactly how to transform your retrospectives with practical tools and techniques you can implement immediately.

Key topics covered:

• Common retrospective anti-patterns and why teams become disengaged
• The critical importance of treating action items as "first-class citizens"
• How to surface recurring themes and environmental issues beyond team control
• Practical strategies for breaking down overwhelming improvement initiatives
• The need for leadership buy-in and organizational support for retrospective outcomes
• Moving from "doing agile" to "being agile" through effective reflection and action

This conversation is packed with insights for making your retrospectives more impactful and driving real organizational change.

About our guests

Jaclyn Smith is a Senior Product Manager at Easy Agile, where she leads the Easy Agile TeamRhythm product that helps teams realize the full benefits of their practices. With over five years of experience as both an in-house and consulting agile coach, Jaclyn has worked across diverse industries helping teams improve their ways of working. At Easy Agile, she focuses on empowering teams to break down work effectively, estimate accurately, and most importantly, take meaningful action to continuously improve their delivery and collaboration.

Shane Raubenheimer is an Agile Technical Consultant at Adaptavist, a global family of companies that combines teamwork, technology, and processes to help businesses excel. Adaptavist specializes in agile consulting, helping organizations deliver customer value through agile health checks, coaching, assessments, and implementing agile at scale. Shane brings extensive experience working across multiple industries—from petrochemical to IT, digital television, and food industries—applying agile philosophy to solve complex organizational challenges. His expertise spans both the technical and cultural aspects of agile transformation.

Transcript

This transcript has been lightly edited for clarity and readability while maintaining the authentic conversation flow.

Opening and introductions

Jaclyn Smith: Hi everyone, and welcome back to the Easy Agile Podcast. Today I'm talking to Shane Raubenheimer, who's with us from Adaptavist. Today we're talking about why your retrospectives keep failing and how to finally fix them. Shane, you and I have spent a fair amount of time together exploring the topic of retros, haven't we? Do you want to tell us a little bit about yourself first?

Shane Raubenheimer: Yeah, hello everyone. I'm Shane Raubenheimer from Adaptavist. I am an agile coach and technical consultant, and along with Jaclyn, we've had loads of conversations around why retros don't work and how they just become tick-box exercises. Hopefully we're going to demystify some of that today.

Jaclyn Smith: Excellent. What's your background, Shane? What kind of companies have you worked with?

Shane Raubenheimer: I've been privileged enough to work across multiple industries—everything from petrochemical to IT, to digital television, food industry. All different types of applied work, but with the agile philosophy.

Jaclyn Smith: Excellent, a big broad range. I should introduce myself as well. My name is Jaclyn. I am a Senior Product Manager here at Easy Agile, and I look after our Team Rhythm product, which helps teams realize the benefits of being agile. I stumbled there because our whole purpose at Easy Agile is to enable our customers to realize the benefits of being agile.

My product focuses on team and teamwork, and teamwork happens at every level as we know. So helping our customers break down work and estimate work, reflect—which is what we're talking about today—and most importantly, take action to improve their ways of working. I am an agile coach by trade as well as a product manager, and spent about five years in a heap of different industries, both as a consultant like you Shane, and as an in-house coach as well.

The core problem: When retrospectives become checkbox exercises

Jaclyn Smith: All right, let's jump in. My first question for you Shane—I hear a lot that teams get a bit bored with retros, or they face recurring issues in their retrospectives. Is that your experience? Tell me about what you've seen.

Shane Raubenheimer: Absolutely. I think often what should be a positive rollup and action of a sequence of work turns out to normally become a checkbox exercise. There's a lot of latency in the things that get uncovered and discussed, and they just tend to perpetually roll over. It almost becomes a checkbox exercise from what I've seen, rather than the mechanism to actively change what is happening within the team—but more importantly, from influences outside the team.

I think that's where retros fail, because often the team does not have the capability to do any kind of upward or downstream problem solving. They tend to just mull about different ways to ease the issues within the team by pivoting the issues rather than solving them.

I think that's where retros fail, because often the team does not have the capability to do any kind of upward or downstream problem solving. They tend to just mull about different ways to ease the issues within the team by pivoting the issues rather than solving them.

Jaclyn Smith: Yeah, I would agree. Something that I see regularly too is because they become that checkbox, teams get really bored of them. They do them because they're part of their sprint, part of their work, but they're not engaged in them anymore. It's just this thing that they have to do.

It also can promote a tendency to just look at what's recently happened and within their sphere of influence to solve. Whereas I think a lot of the issues that sometimes pop up are things that leadership need to help teams resolve, or they need help to solve. It can end up with them really focusing on "Oh well, there's this one bit in how we do our code reviews, we've got control over that, we'll try to fix that." Or as you say, the same recurring issues come up and they don't seem to get fixed—they're just the same complaints every time.

Shane Raubenheimer: Absolutely. You find ways that you put a band-aid on them just so you can get through to the next phase. I think the problem with that is the impact that broader issues have on teams is never completely solvable within that space, and it's no one else's mandate necessarily to do it. When an issue is relatable to a team, exposing why it's not a team-specific issue and it's more environmental or potentially process-driven—that's the bit that I feel keeps getting missed.

When an issue is relatable to a team, exposing why it's not a team-specific issue and it's more environmental or potentially process-driven—that's the bit that I feel keeps getting missed.

The pressure problem and overwhelming solutions

Jaclyn Smith: Yeah, I think so too. The other thing you just sparked for me—the recurring issue—I think that also happens when the team are under pressure and they don't feel like they have the time to solve the problems. They just need to get into the next sprint, they need to get the next bit of work done. Or maybe that thing that they need to solve is actually a larger thing—it's not something small that they can just change.

They need to rethink things like testing strategies. If that's not working for you, and it's not just about fixing a few flaky tests, but you need to re-look at how you're approaching testing—it seems overwhelming and a bit too big.

Shane Raubenheimer: Absolutely. Often environmental issues are ignored in favor of what you've been mandated to do. You almost retrofit the thing as best you can because it's an environmental issue. But finding ways to expose that as a broader-based issue—I think that should be the only output, especially if it's environmental and not team-based.

The problem of forgotten action items

Jaclyn Smith: Something I've also seen recently is that teams will come up with great ideas of things that they could do. As I said before, sometimes they're under pressure and they don't feel they have the capacity to make those changes. Sometimes those actions get talked about, everyone thinks it's a wonderful idea, and then they just get forgotten about. Teams end up with this big long backlog of wonderful experiments and things that they could have tried that have just been out of sight, out of mind. Have you seen much of that yourself?

Shane Raubenheimer: Plenty. Yes, and often teams err on the side of what's expected of them rather than innovate or optimize. I think that's really where explaining the retrospective concept to people outside fully-stacked or insular teams is the point here. You need, very much like in change management, somebody outside the constructs of teams to almost champion that directive—the same way as you would do lobbying for money or transformation. It needs to be taken more seriously and incorporated into not just teams being mini-factories supporting a whole.

You transform at a company level, you change-manage at a company level. So you should action retrospective influences in the same way. Naturally you get team-level ones, and that's normally where retrospectives do go well because it's the art of the possible and what you're mandated to do. I think bridging the gap between what we can fix ourselves and who can help us expose it is a big thing.

I see so much great work going to waste because it simply isn't part of the day job, or should be but isn't.

You transform at a company level, you change-manage at a company level. So you should action retrospective influences in the same way.

Making action items first-class citizens

Jaclyn Smith: Yeah, absolutely. I know particularly in the pre-Covid times when we were doing a lot of retros in person, or mostly in person with stickies on walls, I also found even if we took a snapshot of the action column, it would still end up on a Confluence board or something somewhere and get forgotten about. Then the next retro comes around and you sort of feel like you're starting fresh and just looking at the last sprint again. You're like, "Oh yeah, someone raised that last retro, but we still didn't do anything about that."

Shane Raubenheimer: I think Product Owners, Scrum Masters, or any versions of those kinds of roles need to treat environmental change or anti-pattern change as seriously as they treat grooming work—the actual work itself. Because it doesn't matter how good you are if the impediments that are outside of your control are not managed or treated with the same kind of importance as the actual work you're doing. That'll never change, it'll just perpetuate. Sooner or later you hit critical mass. There's no scenario where your predictability or velocity gets better if these things are inherent to an environment you can't control.

Product Owners, Scrum Masters, or any versions of those kinds of roles need to treat environmental change or anti-pattern change as seriously as they treat grooming work—the actual work itself.

Jaclyn Smith: Yeah, that's true. We've talked about action items being first-class citizens and how we help teams do that for that exact reason. Because a retro is helpful to build relationships and empathy amongst the team for what's happening for each of them and feel a sense of community within their team. But the real change comes from these incremental changes that are made—the conversations that spark the important things to do to make those changes to improve how the team works.

That action component is really the critical part, or maybe one of two critical parts of a retro. I feel like sometimes it's the forgotten child of the retro. Everyone focuses a lot on engaging people in getting their ideas out, and there's not as much time spent on the action items and what's going to be done or changed as a result.

Beyond team-level retrospectives

Shane Raubenheimer: Absolutely, consistently. I think it's symptomatic potentially of how retros are perceived. They're perceived as an inward-facing, insular reevaluation of what a team is doing. But I've always thought, in the same way you have the concept of team of teams, or if you're in a scaled environment like PI planning, I feel retrospectives need the same treatment or need to be invited to the VIP section to become part of that.

Because retrospectives—yes, they're insular or introspective—but they need to be exposed at the same kind of level as things like managing your releases or training or QA, and they're not.

Jaclyn Smith: Yeah, I think like a lot of things, they've fallen foul of the sometimes contentious "agile" word. People tend to think, "Oh retros, it's just one of those agile ceremonies or agile things that you do." The purpose of them can get really lost in that, and how useful they can be in creating change. At the end of the day, it's about improving the business outcomes. That's why all of these things are in place—you want to improve how well you work together so that you can get to the outcome quicker.

At the end of the day, it's about improving the business outcomes. That's why all of these things are in place—you want to improve how well you work together so that you can get to the outcome quicker.

Shane Raubenheimer: Absolutely. Outcome being the operative word, not successfully deploying code or...

Jaclyn Smith: Or ticking the retro box, successfully having a retro.

Shane Raubenheimer: Yeah, exactly. Being doing agile instead of being agile, right?

Expanding the scope of retrospectives

Jaclyn Smith: One hundred percent. It also strikes me that there is still a tendency for retros to be only at a team level and only a reflection of the most recent period of time. So particularly if a team are doing Scrum or some version of Scrum with sprints, to look back over just the most recent period. I think sometimes the two things—the intent of a retro but also the prime directive of the retro—gets lost.

In terms of intent, you can run a retro about anything. Think about a post-mortem when you have an incident and everyone gets together to discuss what happened and how we prevent that in the future. I think people forget that you can have a retro and look at your system of work, and even hone in on something like "How are we estimating? Are we doing that well? Do we need to improve how we're doing that?" Take one portion of what you're working on and interrogate it.

You can run a retro about anything. I think people forget that you can have a retro and look at your system of work, and even hone in on something like "How are we estimating? Are we doing that well? Do we need to improve how we're doing that?" Take one portion of what you're working on and interrogate it.

Understanding anti-patterns

Shane Raubenheimer: Absolutely. You just default to "what looks good, what can we change, what did we do, what should we stop or start doing?" That's great and all, but without some kind of trended analysis over a period of time, you might just be resurfacing issues that have been there all along. I think that's where the concept or the lack of understanding of anti-patterns comes in, because you're measuring something that's happened again rather than measuring or quantifying why is it happening at all.

I think that's the big mistake of retros—it's almost like an iterative band-aid.

I think that's the big mistake of retros—it's almost like an iterative band-aid.

Jaclyn Smith: Yeah. Tell me a little bit more about some of the anti-patterns that you have seen or how they come into play.

Shane Raubenheimer: One of them we've just touched on—I think the buzzword for it is the cargo cult culture for agile. That's just cookie-cutting agile, doing agile because you have to instead of being agile. Literally making things like your stand-up or your review or even planning just becomes "okay, well we've got to do this, so we've ticked the box and we're following through."

Not understanding the boundaries of what your method is—whether you like playing "wagile" or whether you're waterfall sometimes, agile at other times, and you mistake that variability as your agility. But instead, you don't actually have an identity. You're course-correcting blindly based on what's proportionate to what kind of fire you've got in your way.

Another big anti-pattern is not understanding the concept of what a team culture means and why it's important to have a team goal or a working agreement for your team. Almost your internal contracting. We do it as employees, right?

I think a lot of other anti-patterns come in where something's exposed within a team process, and because it's not interrogated or cross-referenced across your broader base of teams, it's not even recognized as a symptom. It is just a static issue. For me, that's a real anti-pattern in a lot of ways—lack of directive around what to do with retrospectives externally as well as internally. That's simply not a thing.

A lot of other anti-patterns come in where something's exposed within a team process, and because it's not interrogated or cross-referenced across your broader base of teams, it's not even recognized as a symptom. It is just a static issue. For me, that's a real anti-pattern in a lot of ways—lack of directive around what to do with retrospectives externally as well as internally.

Jaclyn Smith: Yeah, I think that's a good call-out for anyone watching or listening. If you're not familiar with anti-patterns, they're common but ineffective responses to recurring problems. They may seem helpful initially to solve an immediate problem, but they ultimately lead to negative outcomes.

Shane, what you just spoke about there with retrospectives—an example of that is that the team feel disengaged with retrospectives and they're not getting anything useful out of it, or change isn't resulting from the retrospectives. So the solution is to not hold them as frequently, or to stop doing them, or not do them at different levels or at different times. That's a really good example of an anti-pattern. It does appear to fix the problem, but longer term it causes more problems than it solves.

Another one that I see is with breaking down work. The idea that spending time together to understand and gain a shared understanding of the work and the outcome that you need takes a lot of time, and breaking down that work and getting aligned on how that work is going to break down on paper can look like quite an investment. But it's also saving time at the other end, reducing risk, reducing duplication and rework to get a better outcome quicker. You shift the time spent—development contracts because you've spent a little bit more time discovering and understanding what you're doing.

A common anti-pattern that I see there is "we spent way too long looking at this, so we're going to not do discovery in the same way anymore," or "one person's going to look at that and break it down."

The budget analogy

Shane Raubenheimer: I always liken it to your budget. The retrospective is always the nice shiny holiday—it's always the first to go.

I always liken it to your budget. The retrospective is always the nice shiny holiday—it's always the first to go.

Jaclyn Smith: It's the contractor.

Shane Raubenheimer: Yeah. It's almost like exposing stuff that everybody allegedly knows to each other is almost seen as counterintuitive because "we're just talking about stuff we all know." It often gets conflated into "okay, we'll just do that in planning." But the reality is the concept of planning and how you amend what you've done in the retrospective—that's a huge anti-pattern because flattening those structures from a ceremonies perspective is what teams tend to do because of your point of "well, we're running out of daylight for doing actual development."

But it's hitting your head against the wall repeatedly and hoping for a different outcome without actually implying a different outcome. Use a different wall even. I think it's because people are so disillusioned with retrospectives. I firmly believe it's not an internal issue. I believe if the voices are being heard at a budgeting level or at a management level, it will change the whole concept of the retrospective.

Solution 1: Getting leadership buy-in

Jaclyn Smith: I like it, and that's a good thread to move on to. So what do we do about it? How do we help change this? What are some of the practical tips that people can deploy?

Shane Raubenheimer: A big practical tip—and this is going to sound like an obvious one—is actual and sincere buy-in. What I mean by that is, as a shareholder, if I am basing your performance and your effectiveness on the quality and output of the work that you're promising me, then I should be taking the issues that you're having that are repeating more seriously.

Because if you're course-correcting for five, six, or seven sprints and you're still not getting this increasing, predictable velocity, and if it's not your team size or your attitude, it's got to be something else. I often relate that to it being environmental.

Buying into the outputs for change the same way as you would into keeping everyone honest, managing budgets, and chasing deadlines—it should all be part of the same thing. They should all be sitting at the VIP table, and I think that's a big one.

Buying into the outputs for change the same way as you would into keeping everyone honest, managing budgets, and chasing deadlines—it should all be part of the same thing. They should all be sitting at the VIP table.

Solution 2: Making patterns visible

Jaclyn Smith: I think so too. Something that occurs to me, and it goes back to what we were talking about right at the beginning, is sometimes identifying that there's a pattern there and that the same thing keeps coming up isn't actually visible, and that's part of the problem, right?

I know some things we've been doing in Easy Agile TeamRhythm around that recently, attempting to help teams with this. We've recently started surfacing all incomplete action items in retrospectives so people can see that big long list. Because they can convert their action items to Jira items or work items, they can also see where they've just been sitting and languishing in the backlog forever and a day and never been planned for anything to be done about them.

We've recently started surfacing all incomplete action items in retrospectives so people can see that big long list. Because they can convert their action items to Jira items or work items, they can also see where they've just been sitting and languishing in the backlog forever and a day and never been planned for anything to be done about them.

We've added a few features to sort and that kind of thing. Coming in the future—and we've been asked about this a lot—is "what about themes? What about things that are bubbling up?" So that's definitely on our radar that will be helpful.

I think that understanding that something has been raised—a problem getting support from another team, or with a broken tool or an outdated tool that needs to be replaced in the dev tooling or something like that—if that's been popping up time and time again and you don't know about it, then even as the leader of that team, you don't have the ammunition to then say "Look, this is how much it's slowed us down."

I think we live in such a data world now. If those actions are also where the evidence is that this is what needs to change and this is where the barriers are...

Solution 3: The power of trend analysis

Shane Raubenheimer: Certainly. I agree. Touching on the trend analytics approach—we do trend analysis on everything except what isn't happening or what is actually going wrong, because we just track the fallout of said lack of application. We don't actually trend or theme, to your point.

We do trend analysis on everything except what isn't happening or what is actually going wrong, because we just track the fallout of said lack of application.

We theme everything when we plan, yet somehow we don't categorize performance issues as an example. If everybody's having a performance issue, that's the theme. We almost need to categorize or expose themes that are outward-facing, not just inward-facing. Because it's well and good saying "well, our automated testing system doesn't work"—what does that mean? Why doesn't it work?

I think it should inspire external investigation. When you do a master data cleanup, you don't just say "well, most of it looks good, let's just put it all in the new space." You literally interrogate it at its most definitive and lowest level. So why not do the same with theming and trending environmental issues that you could actually investigate, and that could become a new initiative that would be driven by a new team that didn't even know it was a thing?

Jaclyn Smith: Yeah, and you're also gathering data at that point to evidence the problem rather than "oh, it's a pain point that keeps coming up." It is, but it gives you the opportunity to quantify that pain point a little bit as well. I think that is sometimes really hard to do when you're talking about developer experience or team member experience. Even outside of product engineering teams, there are things in the employee experience that affect the ability for that delivery—whatever you're delivering—to run smoothly. You want to make that as slick as possible, and that's how you get the faster outcomes.

Solution 4: The human factor

Shane Raubenheimer: Absolutely. You can never underestimate the human factor as well. If everything I'm doing and every member of my team is doing is to the best of not just their capability, but to the best of the ability in what they have available to them, you become jaded, you become frustrated. Because if you're hitting your head against the same issue regardless of how often you're pivoting, that can be very disillusioning, especially if it's not been taken as seriously as your work output.

If everything I'm doing and every member of my team is doing is to the best of not just their capability, but to the best of the ability in what they have available to them, you become jaded, you become frustrated.

We run a week late for a customer delivery or a customer project, and we start complaining about things like money, budget overspend, over-utilization. But identifying systematic or environmental issues that you can actually quantify should be treated in exactly the same way. I feel very strongly about this.

Solution 5: Breaking down overwhelming action items

Jaclyn Smith: We tend to nerd out about this stuff, Shane, and you're in good company. You've also reminded me—we've put together a bit of a workshop to help teams and people understand how to get the most out of their retrospectives, not just in terms of making them engaging, but fundamentally how to leverage actions to make them meaningful and impactful.

We've spoken a lot about the incremental change that is the critical factor when it is something that's within the team's control or closely to the team's control. That's how you get that expansion of impact—the slow incremental change. We've talked about sometimes those action items seem overwhelming and too big. What's your advice if that's the scenario for a team? What do you see happen and what can they do?

Shane Raubenheimer: I would suggest following the mantra of "if a story is too big, you don't understand enough about it yet, or it's not broken down far enough." Incremental change should be treated in exactly the same way. The "eat the elephant one bite at a time" analogy. If it's insurmountable, identify a portion of it that will make it a degree less insurmountable next time, and so on and so forth.

If we're iterating work delivery, problem-solving should be done in rapid iteration as well. That's my view.

Jaclyn Smith: I like it.

The "eat the elephant one bite at a time" analogy. If it's insurmountable, identify a portion of it that will make it a degree less insurmountable next time, and so on and so forth. If we're iterating work delivery, problem-solving should be done in rapid iteration as well.

Wrapping up: What's next?

Jaclyn Smith: I think we're almost wrapping up in terms of time. What can people expect from us if they join our webinar on July 10th, I believe it is, where we dive and nerd out even more about this topic, Shane?

Shane Raubenheimer: I think the benefit of the webinar is going to be a practical showing of what we're waxing lyrical about. It's easy to speak and evangelize, but I think from the webinar we'll show turning our concepts into actual actions that you can eyeball and see the results of.

With our approach that we took to our workshop, I think people will very quickly get the feeling of "this is dealing with cause and effect in a cause and effect way." So practical—to put that in one sentence, an active showing or demonstration of how to quantify and actually do what we've been waxing lyrical about.

the benefit of the webinar is going to be a practical showing of what we're waxing lyrical about. It's easy to speak and evangelize, but I think from the webinar we'll show turning our concepts into actual actions that you can eyeball and see the results of.

‍Jaclyn Smith: Excellent. That was a lovely summation, Shane. If anyone is interested in joining, we urge you to do so. You can hear us talking more about that but get some practical help as well. There is a link to the registration page in the description below.

I think that's about all we have time for today. But Shane, as always, it's been amazing and lovely to chat to you and hear your thoughts on a pocket of the agile world and helping teams.

Shane Raubenheimer: Yeah, it's always great engaging with you. I always enjoy our times together, and it's been my pleasure. I live for this kind of thing.

Jaclyn Smith: It's wonderful! Excellent. Well, I will see you on the 10th, and hopefully we'll see everyone else as well.

Shane Raubenheimer: Perfect. Yeah, looking forward to it.

Jaclyn Smith: Thanks.

Ready to end the frustration of ineffective retrospectives?

Jaclyn Smith and Shane Raubenheimer also hosted a live, hands-on webinar designed to turn retrospectives into powerful engines for continuous improvement.

In this highly interactive session, they talked about how teams can:

• Uncover why retrospectives get stuck in repetitive cycles
• Clearly capture and assign actionable insights
• Identify and avoid common retrospective pitfalls and anti-patterns
• Get hands-on experience with Easy Agile TeamRhythm to streamline retrospective actions
• Practical tools, techniques, and clear next steps to immediately enhance retrospectives and drive meaningful team improvements.

👉 Watch the recording here.

"Having gone through a cloud migration at Splunk, Greg share's some insightful key learnings, challenges and opportunities" - Chloe Hall

Greg Warner has been involved with the Atlassian ecosystem since 2006 and is a frequent speaker at Atlassian events. Greg has worked as a senior consultant for a solution partner, supported Jira and Confluence at Amazon, and in his current role at Splunk, executed a cloud migration to Atlassian Enterprise Cloud for over 10,000 of his colleagues.

In this episode, Greg and Chloe discuss the cloud migration journey:

📌 The mental shift to cloud migration and how to think beyond the technical side

📌 How to navigate the journey without a roadmap to follow

📌 The four pillars to success for your cloud migration journey

📌 Finding the right time to migrate & thinking about future opportunities    beyond your migration

📌 The unexpected value that can come from a cloud migration

+ more!

📲 Subscribe/Listen on your favourite podcasting app.

Thanks, Greg and Chloe!

Transcript

Chloe Hall:

Hey everyone and welcome back to the Easy Agile Podcast. So I'm Chloe, Marketing Coordinator at Easy Agile, and I'll be your host for today's episode. So before we begin, we'd like to acknowledge the traditional custodians of the land from which I am recording today, the Wodiwodi people of the Dharawal-speaking nation and pay our respects to elders past, present, and emerging. We extend that same respect to all Aboriginal and to Australia Islander peoples who are tuning in today.

Chloe Hall:

So we have a very exciting guest on the podcast today. This guest has been involved with the Atlassian ecosystem since 2006 and is a frequent speaker at Atlassian events. He has worked as a senior consultant for a solution partner, supported Jira and Confluence at Amazon and at his current role at Splunk, executed a cloud migration to Atlassian Enterprise Cloud for over 10,000 colleagues. So welcome to the Easy Agile podcast, Greg Warner.

Chloe Hall:

How are you?

Greg Warner:

Good, and thank you for having me.

Chloe Hall:

No worries. It's great to have you here today.

Greg Warner:

This is one of my favorite topics. We talk about cloud migration and yeah, I hope I can explain why.

Chloe Hall:

Yes, that's exactly what we want for you because I remember when we met at Team 22, you were just so passionate about cloud migration and had so many insights to share and I was very intrigued as well.

Greg Warner:

To give it a bit background about myself.

Chloe Hall:

Yeah.

Greg Warner:

I haven't always been a cloud person. So you mentioned before about being involved since 2006. I was involved early days with when Jira had the several different flavors of standard and professional, when you'd order an enterprise license for Atlassian and they'd send you a shirt. That was one of the difference between one of the licenses. So based a lot in the server versions, over many years. I looked at the cloud as being the poorer cousin, if you like.

Greg Warner:

I'd been to several Atlassian summits and later Team events where there was always things of what was happening in cloud but not necessarily server. I participated in writing exam questions for Atlassian certification program for both server and DC. For me, in the last 18 months, two years now, to make this fundamental shift from being certainly a proponent of what we do doing on server in DC to now absolutely cloud first and that is the definite direction that we as a company have chosen and certainly why I'm so passionate about speaking to other enterprise customers about their cloud migration journey.

Chloe Hall:

Wow. So what do you think it was that you were like, okay, let's migrate to the cloud, as you were so involved in the server DC part of it? What was it that grabbed your attention?

Greg Warner:

I joined Splunk in 2019 and it wasn't all roses in regards to how we maintained Jira and Confluence. It wasn't uncommon to have outages that would last hours. For two systems that were just so critical to our business operations to have that, I was kind of dumbfounded but I thought, hey, I've been here before. I have seen this. And so it was a slow methodical approach to root cause our problems, get us to a version that was in long-term support, and then take a breather.

Greg Warner:

Once we got to that point where we didn't have outages, we kind of think of what the future would be. And for me, that future was exactly what I'd done before, what I'd done at Amazon, which is where we would move all of our on-prem infrastructure, Jira, Confluence, and Crowd to public cloud, whether it would be a AWS or GCP, something of that flavor. I'd done that before. I knew how we were going to do that to the extent that I'd even held meetings in my team about how we were going to stand up the infrastructure, what the design was going to be.

Greg Warner:

But there was probably one pivotal conversation that was with our CIO and it was in one of those, just passing by, and he's like, "Greg, I've seen the plans and the funding requests." He's like, "But have you considered Atlassian Cloud?" Now, the immediate personal reaction to me was like, we are not going to do that because I'd seen the iterations. I'd seen it over time. I'd worked for a solution partner. I'd worked with customers in cloud, never really thought we could be enterprise-ready. So my immediate reaction was not going to do that. I said, "I'm not going to answer that question right now." I said, "I don't know enough to give you an answer."

Greg Warner:

And I'm absolutely glad I did that because I would've put a foot in mu mouth had I given the immediate response that was... So yeah, I took that question, went and did some analysis, spoke to our technical account manager at the time, and really looked at what had been going on and where was cloud today? Where was it in its maturity? And the actual monumental thing for me was that I think it's actually ready. People make excuses for why they can't do it, but there are a bunch of reasons why you should. And if we look at us as a company, with our own products that we are moving our own customers to cloud, and we are using cloud services, like Google Workspace and Zoom and a variety of SaaS applications. What was so different about what we did in engineering that couldn't go to cloud? And that was like, okay, I think the CIO was actually asking me a much bigger question here.

Greg Warner:

So the result of that was yes, we decided that it was the right time for Splunk to move. And that is a monumental shift. And I know there's a lot of Jira admins out there that are like, if you do this, you're putting your own jobs at risk. The answer is no, you're not. And even within my team, when we had we'd discussed this, there was emotional connection to maintaining on-premise infrastructure and were we giving our own jobs away if we do this? There's all those... No.

Greg Warner:

And there have actually been two people in my team that got actually promoted through the work of our cloud migration that otherwise wouldn't have because they could demonstrate the skills. But that's kind of like the backstory about how we decided to go to cloud. And I think as we are thinking about it, there is a mental shift first. Before you even go down the technical path about how you would do it, change your own mind so that it's open so that you're ready for it as well.

Chloe Hall:

Yeah, I love that. It's so good. And I think just the fact that you didn't respond to your CIO, did you say that?

Greg Warner:

Yep.

Chloe Hall:

That you didn't respond to your CIO straight away and you weren't like, "No, I don't want to do that." You actually stepped away, took that time to do your research, and think maybe cloud is the better option for Splunk, which is just so great and really created that mental shift in yourself. So when you say that your employees, like everyone kind of has that beef that, oh, we're going to lose our job if we move from on-prem to cloud and those employees ended up getting promoted. How did their roles change?

Greg Warner:

When we moved from on-prem to cloud, you no longer have to maintain the plumbing, right?

Chloe Hall:

Yeah.

Greg Warner:

You no longer have to maintain all the plumbing that's supporting Jira, Confluence, BitBucket, whatever is going to move. Now we thought that was the piece that's actually providing value to the organization. And it wasn't until we went to cloud, we actually realized it wasn't. Like what we can do now is different. And that's what my team has done. They've up-leveled.

Greg Warner:

So in the times since we moved from Jira, Confluence on-prem to cloud, we now get involved a lot more with the business analysis and understanding what our project teams want. So when someone from engineering is requesting something that has an integration or a workflow, we've got more time to spend on that than are we going to upgrade? Are we on the current feature release? Is there a bug we have to close? Log for J as a prime example where the extent of where we covered was logging a call with the Atlassian enterprise support and then telling us, "Yep, it's done."

Greg Warner:

Whereas other colleagues within the ecosystem that I spoke to spent a week dealing with that, right? Dealing with patching and upgrades. So the value for our team in the work we do has shifted up. We've also done Jira advanced roadmaps in that time. So we've been able to provide things we would've never got to because we're too busy to the plumbing, to the extent now that we have a very small footprint of on-prem that remains and that's primarily FedRAMP and IO5. It's not quite certified yet. It's going to get there. So we have a very small footprint and I'm the one who has to do the upgrades and now you look at it like, oh my god, that's going to be this couple-week tasks we going to do where I could do all this other better work that's waiting for us in cloud. You don't realize it until you have it removed how much you used to do.

Greg Warner:

And so we used to do two upgrades of Jira year and two upgrades of Confluence a year. We put that down to about a month's work of each. By the time you do all of your testing and you're staging and then do that. So you're really looking at four months of the year you were spending doing upgrades. We don't have that anymore. It's completely gone. And so now we make sure that we do things cloud first. We don't bring across behaviors that we were doing on-prem into cloud. So that's probably one thing we learned was that don't implement server DC in cloud.

Chloe Hall:

Yeah, that's so great. It seems like it's opened up a lot more opportunity for you as well. So I think something that I kind of want to look into and understand a bit more is that people focus a lot on the technical aspect of the cloud migration. What other aspects do you think need to be considered?

Greg Warner:

Certainly people. I mentioned at the very front here the mental mindset and that really started with my team, to get their mind around how we're going to do this cloud migration. There isn't necessarily yet a roadmap that says these are all the steps need to take to get ready for your cloud migration. So we had to invent some of those and one of those two was, what did we want to get out of the cloud migration?

Greg Warner:

I speak to other Atlassian customers. You talk about they're running a project, the project is the cloud migration, the start and the end is the cloud migration day. No, completely wrong. The cloud migration actually has a beginning, a middle, and an end. What you're talking about here, about this first changes is in the beginning, and that should be we're moving to cloud because it should be fundamentally better than what we have today.

Greg Warner:

If it's not better, there's no value in doing the activity. So we started with a vision and that vision was that all of the core things had to work from day one and they had to work better. So create issue, edit issue, up to issue, that just needs to work. There should be no argument whether it does or does not. That needs to work and work better. Create a page, edit a page, share a page. That stuff needs to work in Confluence without any problems. We also need to make sure that there are people in the organization who this could be a fundamental change of how they work, depending on how much they work with Jira and Confluence. So appreciating that there is some change management and some communications that needs to be ready as you do your cloud migration to ensure that your vision is going to work, but also acknowledging you will break some things. You're not going to be able to do a cloud migration and shift you from A to B without nothing.

Greg Warner:

It will go wrong. So we were aware of that and for that, what I would always tell people was that we're really fixed on the vision of making it sure it's better than it was today, but flexible on the details, how we get there. We will probably find different ways as we go along because things will change. Cloud changes itself. You'll discover things you didn't know before. There was a Jira admin that made a decision 10 years ago, you now found that. So yeah, very, very fixed on that vision that day one that we had to have this unboxing experience that when people got to use Jira and Conference Cloud for the first time, they could see why we'd spent so much effort to make sure it was polished and things just worked. And as you went a bit further out, there might be things to do with apps that might not be quite the same.

Greg Warner:

That's okay. And then further out, things you just ultimately can't control. And for that, we had 76 integrations of teams that had written automations from all over the company. We're never going to get to find out what they do, but we knew that some of those would probably break. And so just dealing with some change control and allowing those people to know this is coming, what the rest endpoints will be, how to set up their API keys. We did a lot of that, but we did have one integration that broke and that integration broke because the entire team was on PTO or leave that week. We can't avoid that one. But it was good to see other teams actually jumped in because they'd been involved in updating theirs to go help fix that. So that was okay. We had one integration that we really gave the white glove support to and that was for... We have a Salesforce to Jira integration that's a revenue-generating integration.

Greg Warner:

We gave that a lot of attention to make sure that just worked. But the 76 others, we provided a runbook. The runbook was essentially teams, you do things like this. So they knew how to change and update to the new system. But yeah, certainly the beginning, middle and end. The beginning is all those shifts that you're going to have to change and probably some history about design decisions. The middle is in fact your cloud migration and the end, middle to the end is everything you do with it afterwards. So that's where the real value comes from in your cloud migration. It's once you're in, what can we do with it?

Greg Warner:

And we are towards the end of that now. There have been things that I couldn't have planned for that people have done. So we did your advanced roadmaps, saving the forest there, but also we're encouraging our staff to extend the platform. That used to be really difficult and we've worked with Atlassian to understand what should that look like? And we've settled on using it Atlassian Forge. And so now we have our first app this week, in UAT, in Atlassian Cloud to solve business problems that we have. That's a custom Atlassian Forge app. And we're encouraging our engineers to build those and so they can extend and get that real value through the cloud migration.

Chloe Hall:

Yeah, wow. You've come so far and it's nice to hear that you're towards the end of it and all the opportunities are coming with it and you're seeing all the value. It's all paying off as well. I think I just want to go back to that moment where you talk about there isn't essentially a roadmap outlay. There isn't someone or something to follow where it says this is where you need to start. These are the steps to cloud migration. And I think a lot of people, that's what they fear. They're like, we're not sure exactly where to start. We're not sure what roadmap we'll follow. How do you navigate that in a way?

Greg Warner:

So I get back to that when I talked about the vision. We said we're fixing the vision flexible details. Early on when we signed for cloud migration, it was in the first week after we'd signed for it, that same CIO asked me, "Greg, what's our date? When are we moving? Because you've sold me that this is so much better. Where's the action? When are we get this?" And we took a good six weeks after we signed to actually understand the tooling that's available. So for Jira, there's really two options. There's the Jira site import and the Jira cloud migration assistant. And on Confluence side, there's one that's called the Confluence cloud migration assistant. Better kind of understand how those technologies work. And for a couple weeks there, my team actually considered if we did the migration ourself, we could probably save the company a bunch of money and we would own it.

Greg Warner:

We would know how this thing worked. We got about four weeks in and decided that was a terrible idea. Do not do that. Any enterprise customers I talk about that say we're going to do it ourselves, do not do that. Do not do that. And part of the reason is that there's really four pillars to success for your cloud migration. Jira migration, Confluence migration, apps, and users. And we did not know how to do apps and users and we probably could have gotten away with Confluence and Jira. But we said, look, this is something that we actually need to have a partner involved. And so we did ask for partners to provide their way of doing it, knowing what they knew about us. And we did provide as much detail as we can. We had two partners actually provided completely different methodologies how to get there.

Greg Warner:

So this is that flexible on the details, but we really had to make a decision on what worked for us. So when it really came down to Jira, would we do a big bang approach and just switch it over in the course of a weekend or did we want to do cohort by cohort over time? And we decided for us, because we are a 24/7 organization that's supporting our customers, doing the big bang switchover, that was the best way to do it. So that's one of the reasons we chose the partner we did. But that partner didn't necessarily have a roadmap of where they want to go. But we did then explain what we want to get out of this. That was the first thing, was about it needs to happen on a weekend. So that then filters down what your choices are. The ecosystem apps part is really important to make sure that one, there may have been apps installed in your system that have been there for 10 years and you're not sure why they're there anymore because it was four Jira admins ago.

Greg Warner:

Nobody knows what's there. But if they don't have a cloud migration pathway, you really should consider they're probably going to hit their end because there is no equivalent. So you can rule them out. Identify the ones that do have a business process with them. And for that, Salesforce for us, we had to find a cloud-first connect that would work. So that meant that we knew that was going forward. But really, I think the key thing that we invented that we didn't know about was that we created this thing called an App Burn Down. And that's where we looked at all the apps we had. We had about 40 apps. We said, okay, which ones are not going to go to cloud? Which ones don't have a migration pathway? Which ones are going to replace something else? And so we started to remove apps over the course of about three months.

Greg Warner:

So people would see that we're starting to get away from on-prem design decisions and old ways of doing things. But we also said, but once we get to cloud, this is the pathway out of it. So that we said, look, we're going to turn this app off but you're going to get this one instead, which is the cloud-first app. So people could see how we're going to make the jump over the river to get there. But it meant that we would, over time, identify apps that weren't used. If we turned them off and nothing happened, it's fine. But also we did come across some where they were critical to a business use. And so if we didn't have an answer for those yet, it gave us time to find one. And with your user base, typically it's your colleagues, that's going to be your most critical customers. They're going to ask, okay, you're turning it off. When do I get the functionality back?

Greg Warner:

And by doing that App Burn Down over time, that does buy you time to then have that answer. So it's a much easier conversation than I'm simply turning off functionality, I don't have an answer for you yet. There are things like that. It wasn't necessarily a roadmap, but working with a solution partner is absolutely the right way to go. Don't try and do it yourself. They also work with Atlassian and they have far better reach into getting some of these answers than you can possibly ever have. And I have on at least three different occasions where our solution partner did go and speak directly with an ecosystem partner to find out what's the path forward. How can we make this work? So it is good. The migration is really a three-way collaboration between yourself, your solution partner, and Atlassian. And you all have the same goals. You want to get to cloud and it does work really well.

Chloe Hall:

Wow. Yeah. So sounds like hope everyone got that advice. Definitely don't take this on your own. Reach out to solution partner. And I really like how you said you went to two different solution partners and you found out what their ideas were, which ways they wanted to take you, so you could kind of explore your options, work out what was the best route for Splunk. And it's worked very well for you as well. Having that support I think as well. Yeah. Sorry, you go.

Greg Warner:

The choice of the partner is really important and it's probably one of the earliest decisions that we made to get that one right. And I remember several times thinking about, have we got the right people on board? Did we speak to... And it was an interview process to the extent that when we had our final day after we'd been working with Atlassian and with our partner for six months, one month after our migration was completed and we're all done, we had one final Zoom call with all of us and took a photo and did that. But it kind of felt like a breakup, to be honest, because we'd been in each other's faces for six months and working. We're now all saying goodbye. We might not see each other. It was like the weirdest feeling. But it did work. And so yeah, it is a real fundamental choice.

Greg Warner:

Just take the time, make sure they understand what we want to do, make sure you understand how they're going to do it. But yeah, if we have done it ourselves, we would've got ourselves all caught up in knots, wouldn't have been a successful migration or so. I'm a technical guy. I want to solve it. I want to be like... But I think the actual right answer was no, you don't need to know how this works 100% because you're going to do this hopefully just once. And so focus on the real business value things about dealing with stakeholders and the change and making design decisions that are really important for you because you're going to own those probably the next decade rather than worrying about how do I get my data from A to Z?

Chloe Hall:

Yeah. It definitely would've felt like a breakup for you because you would've been working side by side for so long, dealing with so much. Are you still in contact with them or...

Greg Warner:

Yeah, we had this fundamental thing we always said is we're always, if there's a problem, we're always cautiously optimistic, we're going to solve it. We did engineering challenges that we went through, but I did say right early on is, the ecosystem is only big and we're all going to bump into each other at some point. So yeah, let's make sure that we're still friends at the end of this. And I didn't realize how important that was until later when I was in New York for Christmas and I arranged to meet the project manager that worked for us. She lives in New York, so how about I meet you so... So we met each other at the hotel and she's like, "I have never met a customer outside of work to do this." Yeah, I gave the story about it felt like a breakup, but she did say that at the beginning you said we'll be friends after.

Greg Warner:

Yeah it is because it can be really hard. I've been on the consultant side where you kind of have to have some hard conversations and sometimes... You want to make sure that everyone understands the problem. You're trying to make it better so that at the end of it, you can still be friends like that. That is the thing. There probably will be engagements later on that you might need them again. So you want to make sure that you have your choice of best in breed partner to choose from. You have those relationships. They understand what you want to choose. So yeah, it is really important to choose the right partner. Don't necessarily based on price but choose the partner that's going to work for you, understands what you're trying to get out of your cloud migration and they'll be there in the future when you need them for another cloud migration or a much more gnarly project. Try and be friends at the end of it.

Chloe Hall:

And definitely it's good that you have that friendship now because they have that understanding about your business and what you want and the value of it. So if you do need help again, it's a lot easier to bring them on board straight away. So now that you've performed a cloud migration and you're coming towards the end of it, do you look at the process any differently to when you were at the very beginning?

Greg Warner:

Yeah, I thought we were just executing a data migration just yeah, on-prem to cloud.

Chloe Hall:

Yeah.

Greg Warner:

Pretty straightforward, nothing big. I was pleasantly surprised as we're making some of these decisions as we went along, that it was more than that. There were business processes that we could improve. There was the beginning, the middle, and end. I didn't realize that until actually after the end. So when we did our cloud migration, it was actually the week before Thanksgiving in the US. It was November 19. And even that decision was made in just going for a walk at lunchtime. When should we really do this? And I kind of came down again, spoke to my project manager and said, "How about we do this in the cloud migration the week before Thanksgiving?" Because 50% of our workforce is located in the US and a large proportion of that will be on leave or PTO before.

Greg Warner:

So by doing it over a weekend before then we're ensuring that... Like when you open a new restaurant. You don't want to have all of your tables full on the first night. We knew that we were going to have everybody using Jira and Confluence day one after a migration because we're going to break some stuff. They actually turned out to be really exceptionally good idea. And I encouraged people to find... Look at your data and work out when is low time to do this? I've been involved in Jira and Confluence for a long time and just thought it's task tracker and it's a wiki. There's nothing there that I don't really know about. But one of the decisions we made was actually that when we completed the data migration and it was ready to go, I always said if we waited, do we get a better result? And the answer was no.

Greg Warner:

We should make this available to people now. And so we opened it up on a Sunday morning in the US, which was starting to be business hours in Australia. We started making teams aware that they can now go ahead and use Jira and Confluence. And it was the feedback that we immediately got from those teams that were starting to use Jira service management in cloud for the first time, about, "Wow, this is so much better than it was on-prem." And people said, "I can actually see the attention to detail you've made on fields and descriptions and the changes you've made." And it started to impact people's workday that this was better than it was. I didn't expect that to come back. And so I have a montage that we share with the team of all these Slack messages from people saying, "This is really good. This is much better than we had before."

Greg Warner:

What I didn't also realize is that when we moved from on-prem to cloud is the data that we had became more usable and accessible. Hadn't planned that. It seems obvious now, but when we put it in cloud and it has all the security controls around it and now no longer has the requirements of things like VPN to get access to it, people could build new things to use it to be able to interact with your issues, to interact with pages. And so we started with 76 integrations and over space of three months now we had this big jump in the first three months up to about a hundred something and now we're going to Forge And what it means is people who have had this need to be able to get to the data can now get to it. I didn't see that coming. I just thought we were just server cloud. But yeah, having a more accessible has led to improvements in the way that our teams are working but also how they use it in other applications that just simply wasn't available before.

Chloe Hall:

Yeah. Wow. That's great. And it's good that you were able to receive that feedback straight away from the teams that you had in Australia. I think that's really good and it sounds like it's created such a good opportunity for you at Splunk as well now that you're on cloud.

Greg Warner:

Yeah, it's certainly a business leader that can propel you forward and I eagerly come in now and look at what are other teams going to do with it. And so when we had the first team that said they want to build a Forge app, I'm like, Sure. We should not discourage that at all. Extend the platform. That's why we spent the money and time to do it. What can you do with it now? And we did certainly make Atlassian aware on the product side, like how we're using it and where we'd like to see improvements. If you look at the server DC comparison, I used to be that person that would look at the new features in cloud and ask that question about, when is that new feature coming to on-prem? To going to being that customer who's now, I have that feature today, right? And I'm using it because we don't wait for it.

Greg Warner:

So you mentioned about things you didn't plan from the roadmap. There are design decisions that I talk to enterprise customers that I need to make aware of about. One of them is to do with release tracks. In enterprise cloud, you can choose to bunch up the change to cloud and then they get released periodically every two weeks, every month. When I looked at that, came back to one of our principles about don't implement server in cloud, why would we do that? Atlassian has far more data points on whether this works for customers at scale than we do. So why would we hold back functionality? So as a result we don't do release tracks. We let all of the new functionality get delivered to us as Atlassian sees fit. And the result of that is our own engineering staff, our own support staff who use Jira, get the notifications about new products and features and this is fantastic.

Greg Warner:

Again, why would we implement server, which is where you would bunch up all your changes and then go forward? The other thing too about our cloud migration journey is don't be blinked that you're just doing a cloud migration today and then the project ends. There are things you need to be thinking about as you go along, but what's the impact in the future? So for us, we have multiple sites. Enterprise customer have multiple sites. So there are design decisions that we've made so that we can, in the future, do cloud to cloud migration. You will move sites. Your organization could be bought or could be buying companies. So you do mergers and acquisitions. And so as part of that, we have some runbooks now that talk about using the cloud-to-cloud tooling so we can move a Jira project from a site here to a site there, how we'd move users here and users there.

Greg Warner:

And that actually came about through the assistance with our TAM, not focusing just always on the cloud migration date but also what's that look like six months later? What's it look 12 months later? So that you don't perform your cloud migration and then lock yourself in a corner that later on now I have to unwind something. I had the opportunity to fix it. So yeah, I do encourage migration customers to also think six months, 12 months beyond their cloud migration. But what could also happen and then speak to your solution partner about design decisions today that could affect you in the future.

Chloe Hall:

Yeah. So you definitely need to be thinking future-focus when you're doing this cloud migration. I know you've addressed a lot of the opportunities that came out of the cloud migration. Was there anything else that was an unexpected value that came from it that you wanted to share?

Greg Warner:

The other value is make it more accessible. We have seen people use it in different places that we hadn't thought about. So some of the things that we were doing before, we had to have a company-owned asset to get on the VPN and just things like that. That actually restricted people in where they could do work. Whereas now we've, as long as you've got a computer or mobile device connected to the Internet, absolutely you can use a mobile device support, you can get access to it. Approvals that used to be done on a computer are now done on a mobile device. Those things. But I think the integrations has been probably been the one thing I'm most... We're not the catalyst. We kind of pushed it along but seeing people get real use out of it and using the data for other purposes. We have seen people build some microservices that use the data from Jira that we couldn't do before. Again, you're just unlocking that potential by making it more usable and accessible.

Chloe Hall:

After going through the whole migration journey and, like you said, you're coming towards the end of it, what were the things that stood out to you that you're like, okay, they didn't go so well? Maybe if I was to do this again, how would I do this better next time?

Greg Warner:

So I get back to that day one unboxing experience. You know you want to give it that best experience. And we delivered that for people in Australia and APAC as we opened it and they got to use Jira for the first time and it worked fine. And that is mainly the result of a lot of emphasis on the Jira piece because we said, we know this is going to be hard. It's got workflows, issue schemes, notifications schemes. This is going to be hard.

Greg Warner:

So we started that one really early and then probably about 60% down through our migration journey, we started on Confluence. We thought how hard can Confluence be. It's a bunch of spaces and pages. It can't be that hard. We actually hit some migration challenges with the engineering tooling with Confluence, which meant that the Confluence UAT was delayed. The Jira UAT was fantastic. Ran for a month. We found some problems, got fixed, got answers. We were really confident that was going to be fine.

Greg Warner:

And then we hit this Confluence piece. We're like, wow, this is going to be a challenge. And there was at least one time I could think of. It was a Saturday morning at breakfast where our solution partner sent me a Slack message about, I think we've got a problem here with some tooling. What are we going to do? Towards the middle of the day, I was kind of scratching my head. This could be a real blocker. We actually worked with Atlassian, came up with the engineering solution, cleared that out. That was good to see, like in the space of 12 to 24 hours, there was a solution. But what it meant was that it delayed the Confluence UAT and it made a week. And there was something we found to do with the new Confluence editor and third-party apps right at the end of that week. And we had to really negotiate with our stakeholders to make this go ahead.

Greg Warner:

Because again, if we'd waited, we'd get a better result. No, we really should go. We know that there's this problem. It's not system-wide but it affects a small group people. So we did it. But for about a hundred people they have this really bad Confluence experience because of this thing. And so for me, I couldn't deliver on that thing I promised, which was a day one experience that was going to be better than what it had before.

Greg Warner:

Now we did work with Atlassian and app vendors to get some mitigation so it wasn't as bad on day five. It wasn't day one but it wasn't perfect. But I would certainly encourage people to make sure that you do treat Jira and Confluence with as much importance as each other. They do go together. When I did our cloud migration, we did it on a weekend and I remember coming back after dropping my kids at school on Tuesday and sitting in the car park. I was like, wow, we actually pulled that off.

Greg Warner:

If we'd propose to the company to move your company email system and your finance system on a weekend, the answer would be no because it's too big a hat. But what we'd said is we're going to move all of our Atlassian stack in a weekend, which really is two big systems, Jira and Confluence. So if I had the time again, we would've started Confluence much, much earlier and then we wouldn't have the need to rush it at the end. And that really did result in a bad day one experience for those people. We have worked with Atlassian since then. We're getting that resolved. We know other Atlassian guys have the same problem. I would start early and don't underestimate the complexity that could happen. There will be some things outside of your control.

Greg Warner:

I talk about this Confluence problem and the migration tooling, which is actually do at scale. Not every customer will see it. We saw it, I conducted customer interviews when we were doing our solution partner decision and the customer actually told me this. Like I should have started Confluence because we had this problem, we wasted some time, and we did it. I even have my notes. But it wasn't until later, same problem, you even had the answer and they told you and you still waited. So I'm spending a few minutes on this podcast talking about it because it happened to me. It's probably going to happen to the next person. So if I could do one thing and that is just encourage you to start it earlier. You're going to end up with a much, much better migration and hopefully can deliver on that day one experience that I couldn't do.

Chloe Hall:

Yeah, no I'm so glad that you've shared that with the Easy Agile audience as well because now they know and hopefully the same mistake won't keep getting repeated. Well, Greg, my final question for you today, and I don't know if you want that to be your answer, but I think it's really good just for the audience, if there's one key takeaway that they can go away with them today from the podcast, what would be that one piece of advice for everyone listening to start their migration journey?

Greg Warner:

The first thing to do is to prioritize it. So if you're an Atlassian customer that's using on-prem Jira or Confluence and you don't have a timeline and you don't have a priority to your cloud migration, start there. Open up the task, which is start to investigate Atlassian Cloud and choose a date. Because yeah, there will come a situation down the track where you might be asked by your CIO and so it's better to have an answer prepared already. I would encourage people to start to look at it because it is the future. If you look across the industry, people are moving to SaaS. It's really a question. Do you want to maintain and be that customer wondering when that feature's coming to cloud or do you want to be that customer in cloud who has it today? We have seen a monumental shift to when we moved to cloud in functionality, availability, all the good things that cloud delivers. And it's one of the biggest promoter... The person that used to write exam questions for servers now saying go to cloud.

Greg Warner:

Absolutely. So when I've spoken to other enterprise customers, particularly at Team, I said like, when do you plan your cloud migration? I was like, wow, we're going to start it in three years. I'm like, three years? You need to go back to the office next week and start like 12 months because yeah you will... There is absolutely a competitive advantage to doing it. And it's not just me being now as biggest cloud opponents. We see it, we see it every day and for me, this is one of the most influential projects I've been involved in with Atlassian since 2006. This one here is going to have a long-lasting effect at Splunk for a long time and I'm happy to speak to yourself at Easy Agile and others about it and here at their cloud journey because I want to go to Team next year. I want to make sure we have these conversations in the whole way about, I got that one thing. It's either I started my Confluence migration earlier or I actually put in a timeline of when we should start our cloud migrations.

Chloe Hall:

Yeah, beautiful. That is some great advice to take away, Greg. And so honestly, thank you so much for coming on the podcast today. You have provided some brilliant insights, takeaways, and also because there is no roadmap, I feel like your guidance is so good for those who are looking to start their cloud migration. Yeah. We really appreciate you sharing your knowledge.

Greg Warner:

All right. Thanks for having me on. Thank you for listening.

Chloe Hall:

No worries.

On this episode of The Easy Agile Podcast, tune in to hear developers Angad, Jared, Jess and Jordan, as they share their thoughts on observability.  

Wollongong has a thriving and supportive tech community and in this episode we have brought together some of our locally based Developers from Siligong Valley for a round table chat on all things observability.

💥 What is observability?
💥 How can you improve observability?
💥 What's the end goal?

‍

‍

‍

"This was a great episode to be a part of! Jess and Jordan shared some really interesting points on the newest tech buzzword - observability""

Be sure to subscribe, enjoy the episode 🎧

‍

‍

‍

Transcript

Jared Kells:

Welcome everybody to the Easy Agile podcast. My name's Jared Kells, and I'm a developer here at Easy Agile. Before we begin, Easy Agile would like to acknowledge the traditional custodians of the land from which we broadcast today, the Wodiwodi people of the Dharawal nation, and pay our respects to elders past, present and emerging, and extend that same respect to any aboriginal people listening with us today.

Jared Kells:

So today's podcast is a bit of a technical one. It says on my run sheet here that we're here to talk about some hot topics for engineers in the IT sector. How exciting that we've got a couple of primarily front end engineers and Angad and I are going to share some front end technical stuff and Jess and Jordan are going to be talking a bit about observability. So we'll start by introductions. So I'll pass it over to Jess.

Jess Belliveau:

Cool. Thanks Jared. Thanks for having me one as well. So yeah, my name's Jess Belliveau. I work for Apptio as an infrastructure engineer. Yeah, Jordan?

Jordan Simonovski:

I'm Jordan Simonovski. I work as a systems engineer in the observability team at Atlassian. I'm a bit of a jack of all trades, tech wise. But yeah, working on building out some pretty beefy systems to handle all of our data at Atlassian at the moment. So, that's fun.

Angad Sethi:

Hello everyone. I'm Angad. I'm working for Easy Agile as a software dev. Nothing fancy like you guys.

Jared Kells:

Nothing fancy!

Jess Belliveau:

Don't sell yourself short.

Jared Kells:

Yeah, I'll say. Yeah, so my name's Jared, and yeah, senior developer at Easy Agile, working on our apps. So mainly, I work on programs and road maps. And yeah, they're front end JavaScript heavy apps. So that's where our experience is. I've heard about this thing called observability, which I think is just logs and stuff, right?

Jess Belliveau:

Yeah, yeah. That's it, we'll wrap up!

Jared Kells:

Podcast over! Tell us about observability.

Jess Belliveau:

Yeah okay, I'll, yeah. Well, I thought first I'd do a little thing of why observability, why we talk about this and sort of for people listening, how we got here. We had a little chat before we started recording to try and feel out something that might interest a broader audience that maybe people don't know a lot about. And there's a lot of movements in the broad IT scope, I guess, that you could talk about. There's so many different things now that are just blowing up. Observability is something that's been a hot topic for a couple of years now. And it's something that's a core part of my job and Jordan's job as well. So it's something easy for us to talk about and it's something that you can give an introduction to without getting too technical. So we don't want to get down. This is something that you can go really deep into the weeds, so we picked it as something that hopefully we can explain to you both at a level that might interest the people at home listening as well.

Jess Belliveau:

Jordan and I figured out these four bullet points that we wanted to cover, and maybe I can do the little overview of that, and then I can make Jordan cover the first bullet point, just throw him straight under the bus.

Jordan Simonovski:

Okay!

Jess Belliveau:

So we thought we'd try and describe to you, first of all, what is observability. Because that's a pretty, the term doesn't give you much of what it is. It gives you a little hint, but it'll be good to base line set what are we talking about when we say what is observability. And then why would a development team want observability? Why would a company want observability? Sort of high level, what sort of benefits you get out of it and who may need it, which is a big thing. You can get caught up in these industry hot buzz words and commit to stuff that you might not need, or that sort of stuff.

Jared Kells:

Yep.

Jordan Simonovski:

Yep.

Jess Belliveau:

We thought we'd talk about some easy wins that you get with observability. So some of the real basic stuff you can try and get, and what advantages you get from it. And then we just thought because we're no going to try and get too deep, we could just give a few pointers to some websites and some YouTube talks for further reading that people want to do, and go from there. So yeah, Jordan you want to-

Jared Kells:

Sounds good.

Jess Belliveau:

Yeah. I hopefully, hopefully. We'll see how this goes! And I guess if you guys have questions as well, that's something we should, if there's stuff that you think we don't cover or that you want to know more, ask away.

Jordan Simonovski:

I guess to start with observability, it's a topic I get really excited about, because as someone that's been involved in the dev ops and SRE space for so long, observability's come along and promises to close the loop or close a feedback loop on software delivery. And it feels like it's something we don't really have at the moment. And I get that observability maybe sounds new and shiny, but I think the term itself exists to maybe differentiate itself from what's currently out there. A lot of us working in tech know about monitoring and the loading and things like that. And I think they serve their own purpose and they're not in any way obsolete either. Things like traditional monitoring tools. But observability's come along as a way to understand, I think, the overwhelmingly complex systems that we're building at the moment. A lot of companies are probably moving towards some kind of complicated distributed systems architecture, microservices, other buzz words.

Jordan Simonovski:

But even for things like a traditional kind of monolith. Observability really serves to help us ask new questions from our systems. So the way it tends to get explained is monitoring exits for our known unknowns. With seniority comes the ability to predict, almost, in what way your systems will fail. So you'll know. The longer you're in the industry, you know this, like a Java server fails in x, y, z amount of ways, so we should probably monitor our JVM heap, or whatever it is.

Jared Kells:

I was going to say that!

Jordan Simonovski:

I'll try not to get too much into-

Jared Kells:

Runs out of memory!

Jordan Simonovski:

Yeah. So that's something that you're expecting to fail at some point. And that's something that you can consider a known unknown. But then, the promise of observability is that we should be shipping enough data to be able to ask new questions. So the way it tends to get talked about, you see, it's an unknown unknown of our system, that we want to find out about and ask new questions from. And that's where I think observability gets introduced, to answer these questions. Is that a good enough answer? You want me to go any further into detail about this stuff? I can talk all day about this.

Jared Kells:

Is it like a [crosstalk 00:08:05]. So just to repeat it back to you, see if I've understood. Is it kind of like if I've got a, traditionally with a Java app, I might log memories. It's because I know JVM's run out of memory and that's a thing that I monitor, but observability is more broad, like going almost over the top with what you monitor and log so that you can-

Jordan Simonovski:

Yeah. And I wouldn't necessarily say it's going over the top. I think it's maybe adding a bit more context to your data. So if any of you have worked with traces before, observability is very similar to the way traces work and just builds on top of the premise of traces, I guess. So you're creating these events, and these events are different transactions that could be happening in your applications, usually submitting some kind of request. And with that request, you can add a whole bunch of context to it. You can add which server this might be running on, which time zone. All of these additional and all the exciters. You can throw in user agency into there if you want to. The idea of observability is that you're not necessarily constrained by high cardinality data. High cardinality data being data sets that can change quite largely, in terms of the kinds of data they represent, or the combinations of data sets that you could have.

Jordan Simonovski:

So if you want shipping metrics on something, on a per user basis and you want to look at how different users are affected by things, that would be considered a high cardinality metric. And a lot of the time it's not something that traditional monitoring companies or metric providers can really give you as a service. That's where you'll start paying insanely huge bills on things like Datadog or whatever it is, because they're now being considered new metrics. Whereas observability, we try and store our data and query it in a way that we can store pretty vast data sets and say, "Cool. We have errors coming from these kinds of users." And you can start to build up correlations on certain things there. You can find out that users from a particular time zone or a particular device would only be experiencing that error. And from there, you can start building up, I think, better ways of understanding how a particular change might have broken things. Or some particular edge cases that you otherwise couldn't pick up on with something like CPU or memory monitoring.

Angad Sethi:

Would it be fair to say-

Jared Kells:

Yeah. It's [crosstalk 00:11:02].

Angad Sethi:

Oh, sorry Jared.

Jared Kells:

No you can-

Angad Sethi:

Would it be fair to say that, so, observability is basically a set of principles or a way to find the unknown unknowns?

Jordan Simonovski:

Yeah.

Angad Sethi:

Oh.

Jess Belliveau:

And better equip you to find, one of the things I find is a lot of people think, you get caught up in thinking observability is a thing that you can deploy and have and tick a box, but I like your choice of word of it being a set of principles or best practices. It's sort of giving you some guidance around these, having good logging coming out of your application. So structured logs. So you're always getting the same log format that you can look at. Tracing, which Jordan talked a little bit about. So giving you that ability to follow how a user is interacting with all the different microservices and possibly seeing where things are going wrong, and metrics as well. So the good thing with metrics is we're turning things a bit around and trying to make an application, instead of doing, and I don't want to get too technical, black box monitoring, where we're on the outside, trying to peer in with probes and checks like that. But the idea with metrics is the application is actually emitting these metrics to inform us what state it is in, thereby making it more observable.

Jess Belliveau:

Yeah, I like your choice of words there, Angad, that it's like these practices, this sort of guide of where to go, which probably leads into this next point of why would a team want to implement it. If you want to start again, Jordan?

Jordan Simonovski:

Yeah, I can start. And I'll give you a bit more time to speak as well, Jess in this one. I won't rant as much.

Jess Belliveau:

Oh, I didn't sign up for that!

Jordan Simonovski:

I think why teams would want it is because, it really depends on your organization and, I guess, the size of the teams you're working in. Most of the time, I would probably say you don't want to build observability yourself in house. It is something that you can, observability capabilities themselves, you won't achieve it just by buying a thing, like you can't buy dev ops, you can't buy Agile, you can't buy observability either.

Jared Kells:

Hang on, hang on. It says on my run sheet to promote Easy Agile, so that sounds like a good segue-

Jess Belliveau:

Unless you want to buy it. If you do want to buy Agile, the [crosstalk 00:13:55] in the marketplace.

Jared Kells:

Yeah, sorry, sorry, yeah! Go on.

Jordan Simonovski:

You can buy tools that make your life a lot easier, and there are a lot of things out there already which do stuff for people and do surface really interesting data that people might want to look at. I think there are a couple of start ups like LightStep and Honeycomb, which give you a really intuitive way of understanding your data in production. But why you would need this kind of stuff is that you want to know the state of your systems at any given point in time, and to build, I guess, good operational hygiene and good production excellence, I guess as Liz Fong-Jones would put it, is you need to be able to close that feedback loop. We have a whole bunch of tools already. So we have CICD systems in place. We have feature flags now, which help us, I guess, decouple deployments from releases. You can deploy code without actually releasing code, and you can actually give that power to your PM's now if you want to, with feature flags, which is great.

Jordan Simonovski:

But what you can also do now is completely close this loop, and as you're deploying an application, you can say, "I want to canary this deployment. I want to deploy this to 10% of my users, maybe users who are opted in for Beta releases or something of our application, and you can actually look at how that's performing before you release it to a wider audience. So it does make deployments a lot safer. It does give you a better understanding of how you're affecting users as well. And there are a whole bunch of tools that you can use to determine this stuff as well. So if you're looking at how a lot of companies are doing SRE at the moment, or understanding what reliable looks like for their applications, you have things like SLO's in place as well. And SLO's-

Jared Kells:

What's an SLO?

Jordan Simonovski:

They're all tied to user experiences. So you're saying, "Can my user perform this particular interaction?" And if you can effectively measure that and know how users are being affected by the changes you're making, you can easily make decisions around whether or not you continue shipping features or if you drop everything and work on reliability to make sure your users aren't affected. So it's this very user centric approach to doing things. I think in terms of closing the loop, observability gives us that data to say, "Yes, this is how users are being affected. This is how, I guess the 99th percentile of our users are fine, but we have 1% who are having adverse issues with our application." And you can really pinpoint stuff from there and say, "Cool. Users with this particular browser or this particular, or where we've deployed this app to," let's say if you have a global deployment of some kind, you've deployed to an island first, because you don't really care what happens to them. You can say, "Oh, we've actually broken stuff for them." And you can roll it back before you impact 100% of your users.

Jared Kells:

Yeah. I liked what you said about the test. I forgot the acronym, but actually testing the end user behavior. That's kind of exciting to me, because we have all these metrics that are a bit useless. They're cool, "Oh, it's using 1% CPU like it always is, now I don't really care," but can a user open up the app and drag an issue around? It's like-

Jess Belliveau:

Yeah, that's a really great example, right?

Jared Kells:

That's what I really care about.

Jess Belliveau:

The 1% CPU thing, you could look at a CPU usage graph and see a deployment, and the CPU usage doesn't change. Is everything healthy or not? You don't know, whereas if you're getting that deeper level info of the user interactions, you could be using 1% CPU to serve HTTP500 errors to the 80% of the customer base, sort of thing.

Angad Sethi:

How do you do that? The SLO's bit, how do you know a user can log in and drag an issue?

Jordan Simonovski:

Yeah. I think that would come with good instrumenting-

Angad Sethi:

Good question?

Jordan Simonovski:

Yeah, it comes down to actually keeping observability in mind when you are developing new features, the same way you would think about logging a particular thing in your code as you're writing, or writing test for your code, as you're writing code as well. You want to think about how you can instrument something and how you can understand how this particular feature is working in production. Because I think as a lot of Agile and dev ops principles are telling us now is that we do want our applications in production. And as developers, our responsibilities don't end when we deploy something. Our responsibility as a developer ends when we've provided value to the business. And you need a way of understanding that you're actually doing that. And that's where, I guess, you do nee do think about observability with a lot of this stuff, and actually measuring your success metrics. So if you do know that your application is successful if your user can log in and drag stuff around, then that's exactly what you want to measure.

Jared Kells:

I think that we have to build-

Jordan Simonovski:

Yeah?

Jared Kells:

Oh, sorry Jordan.

Jordan Simonovski:

No, you go.

Jared Kells:

I was just going to say we have to build our apps with integration testing in mind already. So doing browser based tests around new features. So it would be about building features with that and the same thing in mind but for testing and production.

Jess Belliveau:

Yeah and the actual how, the actual writing code part, there's this really great project, the open telemetry project, which provides all these sort of API's and SDK's that developers can consume, and it's vendor agnostic. So when you talk about the how, like, "How do I do this? How do I instrument things?" Or, "How do I emit metrics?" They provide all these helpful libraries and includes that you can have, because the last thing you want to do is have to roll this custom solution, because you're then just adding to your technical debt. You're trying to make things easier, but you're then relying on, "Well I need to keep Jared Kells employed, because he wrote our log in engine and no one else knows how it works.

Jess Belliveau:

And then the other thing that comes to mind with something like open telemetry as well, and we talked a bit about Datadog. So Datadog is a SaaS vendor that specializes in observability. And you would push your metrics and your logs and your traces to them and they give you a UI to display. If you choose something that's vendor agnostic, let's just use the example of Easy Agile. Let's say they start Datadog and then in six months time, we don't want to use Datadog anymore, we want to use SignalFx or whatever the Splunk one is now.

Jordan Simonovski:

I think NorthX.

Jess Belliveau:

Yeah. You can change your end point, push your same metrics and all that sort of stuff, maybe with a few little tweaks, but the idea is you don't want to tie in to a single thing.

Jordan Simonovski:

Your data structures remain the same.

Jess Belliveau:

Yeah. So that you could almost do it seamlessly without the developers knowing. There's even companies in the past that I think have pushed to multiple vendors. So you could be consuming vendor A and then you want to do a proof of concept with vendor B to see what the experience is like and you just push your data there as well.

Jared Kells:

Yeah. I think our coupling to Datadog will be I all the dashboards and stuff that we've made. It's not so much the data.

Jess Belliveau:

Yeah. That's sort of the big up sell, right. It's how you interact. That's where they want to get their hooks in, is making it easier for you to interpret that data and manipulate it to meet your needs and that sort of stuff.

Jordan Simonovski:

Observability suggests dashboards, right?

Jess Belliveau:

Yeah, perhaps. You used this term as well, Jordan, "production excellence." And when we talk about who needs observability, I was thinking a bit about that while you were talking. And for me, production excellence, or in Apptio we call it production readiness, operational readiness and that sort of stuff is like we want to deploy something to production like what sort of best practices do we want to have in place before we do that? And I think observability is a real great idea, because it's helping you in the future. You don't know what problems you're going to have down the line, but you're equipping your teams to be able to respond to those problems easily. Whereas, we've all probably been there, we've deployed code of production and we have no observability, we have a huge outage. What went wrong? Well, no one knows, but we know this is the fix, and it's hard to learn from that, or you have to learn from that I guess, and protect the user against future stuff, yeah.

Jess Belliveau:

When I think easy wins for observability, the first thing that really comes to mind is this whole idea of structured logging, which is really this idea that your application is you're logging, first of all. Quite important as a baseline starting point, but then you have a structured log format which lets you programmatically pass the logs as well. If you go back in time, maybe logging just looked like plain text with a line, with a timestamp, an error message. Whatever the developer decided to write to the standard out, or to the error file or something like that. Now I think there's a general move to having JSON, an actual formatted blob with that known structure so you can look into it. Tracing's probably not an easy win. That's a little bit harder. You can implement it with open telemetry and libraries and stuff. Requires a bit more understanding of your code base, I guess, and where you want tracing to fire, and that sort of stuff, parsing context through, things like that.

Jordan Simonovski:

I think Atlassian, when you probably just want to know that everything is okay. At a fairly superficial level. Maybe you just want to do some kind of up time on a trend. And then as, I guess, your code might get more complex or your product gets a bit more complex, you can start adding things in there. But I think actually knowing or surfacing the things you know might break. Those would probably be your quickest wins.

Jess Belliveau:

Well, let's mention some things for further reading. If you want to go get the whole picture of the whole, real observability started to get a lot of movement out of the Google SRE book from a few years ago. The Google SRE stuff covers the whole gamut of their soak reliability engineering practice, and observability is a portion of that, there's some great chapters on that. O'Reilly has an observability book, I think, just dedicated to observability now.

Jordan Simonovski:

I think that's still in early release, if people want to google chapters.

Jess Belliveau:

The open telemetry stuff, we'll drop a link to that I think that's really handy to know.

Angad Sethi:

From [inaudible 00:26:12], which is my perspective, as a developer, say I wanted to introduce cornflake use Datadog at Easy Agile. Not very familiar, I'm not very comfortable with it. I know how to navigate, but what's a quick way for me to get started on introducing observability? Sorry to lock my direct job or at my workplace.

Jordan Simonovski:

I would lean, I could be biased here. Jess correct me or give your opinion on this, I would lean heavily towards SLO's for this. And you can have a quick read in the SRE-

Jess Belliveau:

What does SLO stand for, Jordan?

Jordan Simonovski:

Okay, sorry. Buzz words! SLO is a service level objective, not to be confused with service level agreement. An agreement itself is contractual and you can pay people money if you do breach those. An SLO is something you set in your team and you have a target of reliability, because we are getting to the point where we understand that all systems at any point in time are in some kind of degraded state. And yeah, reliability isn't necessarily binary, it's not unreliable or reliable. Most of the time, it's mostly reliable and this gives us a better shared language, I guess. And you can have a read in the SRE handbook by Google, which is free online, which gives you a pretty good understanding of Datadog.

Jordan Simonovski:

I think the last time I used it had a SLO offering. But I think like I was mentioning earlier, you set an SLO on particular functionalities or features of your application. You're saying, "My user can do this 99% of the time," or whatever other reliability target you might want to set. I wouldn't recommend five nines of reliability. You'll probably burn yourself out trying to get there. And you have this target set for yourself. And you know exactly what you're measuring, you're measuring particular types of functionality. And you know when you do breach these, users are being affected. And that's where you can actually start thinking about observability. You can think about, "What other features are we implementing that we can start to measure?" Or, "What user facing things are we implementing that we can start to measure?"

Jordan Simonovski:

Other things you could probably look at are, I think they're all covered in the book anyway, data freshness in a way. You want to make sure the data users are being displayed is relatively fresh. You don't want them looking at stale data, so you can look at measuring things like that as well. But you can pretty much break it down into most functionalities of a website. It's no longer like a ping check, that you're just saying, "Yes, HTTP, okay. My application is fine." You're saying, "My users are actually being affected by things not working." And you can start measuring things from there. And that should give you a better understanding, or a better idea, at least, of where to start with what you want to measure and ow you want to measure it. That would be my opinion on where to get started with this if you do want to introduce it.

Jared Kells:

We're going to talk a little bit about state and how with some of these, like our very front end heavy applications that we're building, so the applications we build just basically run inside the browser and the traditional state as you would think about it, is just pulling a very simple API that writes some things into the database with some authentication, and that sort of stuff. So in terms of reliability of the services, it's really reliable. Those tiny API's just never have problems, because it's just so simple. And well, they've got plenty of monitoring around it. But all our state is actually, when you say, "Observe the state of the system," for the most part, that's state in a browser. And how do we get observability into that?

Jess Belliveau:

A big thing is really, there's not one thing fits all as well. When we talk about the SLO stuff as well, it's understanding what is important to not so much maybe your company but your team as well. If you're delivering this product, what's important to you specifically? So one SLO that might work for me at Apptio probably isn't going to work for Easy Agile. This is really pushing my knowledge, as well, of front end stuff, but when we say we want to observe the state as well, we don't necessarily mean specifically just the state. You could want to understand with each one of those API's when it's firing, what the request response time is for that API firing. So that might be an important metric. So you can start to see if one of those APIs is introducing latency, and so your user experience is degraded. Like, "Hey when we were on release three, when users were interacting with our service here, it would respond in this percentile latency. We've done a release and since then, now we're seeing it's now in this percentile. Have we degraded performance performance?" Users might not be complaining, but that could be something that the team then can look into, add to a sprint. Hey, I'm using Agile terms now. Watch out!

Jared Kells:

That's a really good example, Jess. Performance issues for us are typically not an API that's performing poorly. It's something in this very complicated front end application is not running in the same order as it used to, or there's some complex interaction we didn't think of, so it's requesting more data than expected. The APIs are returning. They're never slow, for the most part, but we have performance regressions that we may not know about without seeing them or investigating them. The observability is really at the individual user's browser level. That makes sense? I want to know how long did it take for this particular interaction to happen.

Jess Belliveau:

Yeah. I've never done that sort of side of things. As well, the other thing I guess, you could potentially be impacted in as well as then, you're dealing with end user manifestations as well. You could perceive-

Jared Kells:

Yeah sure.

Jess Belliveau:

... Greater performance on their laptop or something, or their ISP or that sort of stuff. It'd be really hard to make sure you're not getting noise from that sort of thing as well.

Jordan Simonovski:

Yeah. There are tools like Sentry, I guess, which do exist to give you a bit more of an understanding what's happening on your front end. The way Sentry tends to work with JavaScript, is you'll upload a minified map of your JS to Sentry, deploy your code and then if something does break or work in a fairly unexpected way, that tends to get surfaced with Sentry will tell you exactly which line this kind of stuff is happening on, and it's a really cool tool for that company stuff. I don't know if it'd give you the right type of insights, I think, in terms of performance or-

Jared Kells:

Yeah, we use a similar tool and it does work for crashes and that sort of thing. And on the observability front, we log actions like state mutations in side the front end, not the actual state change, but just labels that represent that you updated an issue summary or you clicked this button, that sort of thing, and we send those with our crash reports. And it's super helpful having that sort of observability. So I think I know what you guys are talking about. But I'm just [crosstalk 00:35:25], yeah.

Jess Belliveau:

Yeah, that's almost like, I guess, a form of tracing. For me and Jordan, when we talk about tracing, we might be thinking about 12 different microservices sitting in AWS that are all interacting, whereas you're more shifting that. That's sort of all stuff in the browser interacting and just having that history of this is what the user did and how they've ended up-

Jared Kells:

In that state.

Jess Belliveau:

In that state, yeah.

Jordan Simonovski:

I guess even if you don't have a lot of microservices, if you're talking about particular, like you're saying for the most part your API requests are fine but sometimes you have particularly large payloads-

Jared Kells:

We actually have to monitor, I don't know, maybe you can help with this, we actually should be monitoring maybe who we're integrating with. It's actually much more likely that we'll have a performance issue on a Xero API rather than... We don't see it, the browser sees it as well, which is-

Jordan Simonovski:

Yeah, and tracing does solve all of those regressions for you. Most tracing libraries, like if you're running Node apps or whatever on your backend. I can just tell you about Node, because I probably have the most experience writing Node stuff. You pretty much just drop in Didi trace, which is a Datadog library for tracing into your backend and your hook itself into all of, I think, the common libraries that you'll tend to work with, I think. Like if you're working for express or for a lot of just HADP libraries, as well as a few AWS services, it will kind of hook itself into that. And you can actually pinpoint. It will kind of show you on this pretty cool service map exactly which services you're interacting with and where you might be experiencing a regression. And I think traces do serve to surface that information, which is cool. So that could be something worth investigating.

Jess Belliveau:

It's funny. This is a little bit unrelated to observability, but you've just made me think a bit more about how you're saying you're reliant on third party providers as well. And something I think that's really important that sometimes gets missed is so many of us today are relying on third party providers, like AWS is a huge thing. A lot of people writing apps that require AWS services. And I think a lot of the time, people just assume AWS or Jira or whatever, is 100% up time, always available. And they don't write their code in such a way that deals with failures. And I think it's super important. So many times now I've seen people using the AWS API and they don't implement exponential back off. And so they're basically trying to hit the AWS API, it fails or they might get throttled, for example, and then they just go into a fail state and throw an error to the user. But you could potentially improve that user experience, have a retry mechanism automatically built in and that sort of stuff. It doesn't really tie into the observability thing, but it's something.

Jared Kells:

And the users don't care, right? No one cares if it's an AWS problem. It's your problem, right, your app is too slow.

Jess Belliveau:

Well, they're using your app. Exactly right. It reflects on you sort of thing, so it's in your interest to guard against an upstream failure, or at least inform the user when it's that case. Yeah.

Jared Kells:

Well, I think we're going to have to call it, this podcast, because it was an hour ago. We had instructed max 45 minutes.

Jess Belliveau:

We could just keep going. We might need a part two! Maybe we can request [cross talk 00:39:21].

Jared Kells:

Maybe! Yeah.

Jess Belliveau:

Or we'll just start our own podcast! Yeah.

Angad Sethi:

So what were your biggest learnings today, given it's been Angad and I are just learning about observability, Angad what was your biggest learning today about observability? My biggest learning was that observability does not equal Datadog. No, sorry! It was just very fascinating to learn about quantifying the known unknowns. I don't know if that's a good takeaway, but...

Jess Belliveau:

Any takeaway is a good takeaway! What about you, Jared?

Jared Kells:

I think, because I we were going to talk about state management, and part of it was how we have this ability, at the moment to, the way our front ends are architected, we can capture the state of the app and get a customer to send us their state, basically. And we can load it into our app and just see exactly how it was, just the way our state's designed. But what might be even cooler is to build maybe some observability into that front end for support. I'm thinking instead of just having, we have this button to send us out your support information that sends us a bunch of the state, but instead of console logging to the browser log, we could be console logging, logging in our front end somewhere that when they click, "send support information," our customers should be sending us the actions that they performed.

Jared Kells:

Like, "Hey there's a bug, send us your support information." It doesn't have to be a third party service collecting this observability stuff. We could just build into our... So that's what I'm thinking about.

Jess Belliveau:

Yeah, for sure. It'll probably be a lot less intrusive, as well, as some of the third party stuff that I've seen around.

Jared Kells:

Yeah. It's pretty hard with some of these integrations, especially if you're developing apps that get run behind a firewall.

Jess Belliveau:

Yeah

Jared Kells:

You can't just talk to some of these third parties. So yeah, it's cool though. It's really interesting.

Jess Belliveau:

Well, I hope someone out there listening has learned something, and Jordan and I will send some links through, and we can add them, hopefully, to the show notes or something so people can do some more reading and...

Jared Kells:

All thanks!

Jess Belliveau:

Thanks for having us, yeah.

Jared Kells:

Thanks all for your time, and thanks everybody for listening.

Jordan Simonovski:

Thanks everyone.

Angad Sethi:

That was [inaudible 00:41:55].

Jess Belliveau:

Tune in next week!