The Power of Observability Culture: Transforming the DevOps & Business World – SKILup Days 2024
Observability is evolving beyond metrics and logs to become a foundational culture within DevOps, transforming teams from reactive to proactive, helping them detect issues before they impact users, and drive innovation at scale. In this session, we’ll explore the “why” and “how” of cultivating an observability culture that empowers DevOps teams to excel in today’s digital-first world.
In this session, we’ll cover:
– The key differences between monitoring and observability, and why this shift matters.
– How an observability-driven culture enhances collaboration and improves response times.
– How observability extends beyond DevOps, impacting ProdOps and strengthening resilience throughout the entire product journey.
Top 3 Takeaways:
– Building a Proactive Culture: Learn how to shift from reactive monitoring to a culture where issues are anticipated, not just responded to.
– Enhanced Collaboration: Understand how observability bridges the gap between development and operations, creating a unified approach to incident management.
– Expansion into Strategic Areas: Discover how observability is pushing beyond DevOps, impacting strategic areas like sales and offering valuable insights across the organization.
Transcript
Hi, I'm Zeki and I'm here to talk about observability. I really love to talk about this topic because I start my career work with observability, and one thing that we would discuss today is the observability integrated with the Q. Let's start.
So our agenda will be for talks. I added in the slide, and let's start to talk about the revolution of the observability. As engineer.
I will start with the definition. What is observability? It's very important we understand that in the engineering world, we have the definition, the observability term start in the engineering field.
But today, the marked, see the observability not only as a technical definition, the marked see observability as a full picture. Okay? And I added here why DevOps rely in observability.
So I think everybody here is technical, and we know the importance to be proactive and find errors, detect the things, uh, before, uh, our page happens. So this is very important in our business. Not only technical spec, let's talk about business.
We are technicals, but if our systems talk, what happened? Our business talk. So if I have a good observability in place, and I will talk about the two and the, uh, the third, if we have a good observability in place, we can reduce the, uh, the time to, uh, resolve the things.
We have a lot of things that we can improve. Okay? So observability, again, technically we have the definition that came, uh, is pro, uh, started in the engineering world, but our, uh, technical, uh, word see observability as the cap, uh, the capability to see all the things that we have in the environment.
We need to understand the data that our machines, our applications, they're provide to us. So the observability are, you compare as a exam that we do, so we can see everything that exists in our body. For example, if we do exam, the, the doctor will request.
And okay, we have the foot pan. Imagine that we are doctors of our applications, our environment, our systems. So, but let's continue observability.
In the past, uh, the companies, uh, they always try to say, oh, observability monitor. No. But today they understand that observability is something so fundamental to to have in place.
That is not a luxury. Okay? We need to invest money, time, resourcing observability.
Because again, if you have outage, imagine that we, you have, uh, application. You are you, you are in the middle of the black Friday. Imagine if this is stop.
So imagine the, uh, the money that will stop to be received in your company. Observability is the capability to see technically what's happening in the business. I'm trying to be a simple here, because technically, if you go on, no observability is composed by, uh, trace logs, um, um, and metrics.
This is technical. We need to, as engineers act that, uh, dev people, we need to start to talk the business language to including, uh, he receive more power. That means money to have more people work with observability and more tools.
So why is that revolution? Because we are in the moment that we are in the digital world. Our work is digital.
Everything is going to the cloud. Everything now is application. They stores everything.
So observability, uh, are the eyes of this new digital world. So we as observability professionals, we are so special here. We we're the eyes, we are, we are helping the business understand what they have in place and the strategy that they need to build.
They are the leadership are using that, this data, our data, okay? And the innovation starts here. So, uh, I thought here about the this, um, beautiful thing that's observability.
Observability, but we have a lot of challenge. Yes, challenge because the word is not perfect. Okay?
So let's talk about the observability challenge now. So if you work with observability, probably you understand what I'm trying to communicate here. In observability world, we have a lot of challenge and I try to classify in three.
So the first one is the technical, okay? We know that observability the tools. Let's talk starting, uh, talk.
Let's start talk. Talk about the tools. We know that we have a lot of tools that can provide, uh, the data.
Can we have a lot of APMs in the market? So we have challenge associated with the level of data that are you extract from the system. Okay?
We have challenge, uh, when we have a big company, when we have different, uh, tools added, you know, however, this team has the use. The two A, however, this team use the two B, how to integrate, you know, um, in more data is better to us. And when we talk about data, we are talk about in some case, extra license, uh, storage, et cetera.
So in the technical specs, we have the challenge of the tool, okay? The, the tools, okay? We have the challenge of the silos because the companies, uh, they are organized in silos.
So how to integrate areas that are totally splitted, but they are using the same, I say, application, okay? Or the infra. So we have this kind of challenge.
We have the challenge of mindset sex, because if you don't provide training, okay, uh, the people maybe can think that observability is the same thing as the monetary. Sure, the two works together, but we need to have the mindset to avoid silos in the monitoring world. This is common.
Okay? Sure. I will not go deep here, but exist a difference between monitoring word only monitoring word and the, the word that combine monitoring observability.
So technical aspects, well, it's a big challenge. The other aspect, that's a challenge, it's money. Okay?
Because sometimes we are so technical that the, the leadership, hmm, why I need, explain why I need to invest in this tool. Sure. Today we have a new generation of cos that understand more technically the reason, but we need to explain that we need to have investment in training because change the mindset.
We need to explain that the tool is not expensive, okay? We need to explain that the tool we collect, the data that we create, our way that we can understand the business. Imagine the simple case of, uh, associated with I commerce.
So if we observe, well the commerce, we can see the how the application, the website, everything's flowing. And if we have out Asia problem, we can say, Hey, leadership, if we, if we don't fix this in one hour, the company will, uh, will not receive this money, stop the business, et cetera. So, uh, always it's good to have a plan, a strategy to present to sta uh, to the stakeholders why they need to invest in tooling and not only tooling, but in people that knows how to work with observability.
Okay? So I talked about technical, um, the money aspects and now future cultural aspects. Why?
Because, uh, in the, I will say old mindset, the people, at least some people that I know, maybe you are in the perfect world, but I'm not in the perfect world. So the people think, okay, I will buy this tool and the tool will do everything. Okay?
We have smart tools. Yes, we have. But um, imagine that you have, uh, a company that you have a lot of teams.
Ah, this team use the, the two, a other team two B, the other team two C. Okay? First, you have a lot of, not in the observability area, you have silence in the company.
So it's hard to consolidate things that are not talking in the business level, organizational level. It's, it's complicated to unify great, uh, a single, uh, visualization. So, and when we implement, uh, observability project, we need to talk with the people.
Hey, team A, team C. Now we have, uh, uh, observability tool. Can you adjust your data?
Can you send your data? You know, you need to go to the people and start to, to demonstrate the value of the observability. Again, sometimes we need to go there and talk, Hey, it's not, it is not the monitoring and, and explain, okay?
We will see the full picture and, and create sometimes the bridge, bridge, the connection between two teams is a little. So again, if, if the future of the company is in silos, sorry, what you say here, you can have the best tool in place. It's very hard.
You create some KPIs that you provide good results to the business. One thing is the technical aspect that yes, we need to have the amazing observability to doing the monitoring of, um, the applications and the entire full cycle, et cetera. But, but if we want to have a good money in our area, we need to, uh, uh, create the future help to create the future.
Okay? So I added here the observability challenge in my, uh, based on my experience, because sometimes we have companies with the best thing with amazing tools, okay? But the, when we start to configure the amazing features in the amazing tool, okay?
You don't have money to pay the tool, okay? The technical team goals, they implement amazing thing, et cetera, but the areas are not integrated enough to, uh, receive in a good way what the observability to our, uh, team is designing. Okay?
Ah, why you need my data. So, so these three things needs to work together, okay? Uh, now I would like to talk here about the observability in the product world.
Word. So, uh, always we think about observability in traditional, in traditional aspects, okay? If I am a dev guy, I use the good, um, observability tools to monitor my CI/CD, blah, blah, blah.
So the culminated sector, but, uh, I'm bringing here a different perspective for some guys. I know that here we have a lot of webinars that will go in a deep level technically, but this something, the observer billing products thats happen. The marketing that it so is interesting, interesting.
So, uh, the first thing, uh, what means products? Okay? Uh, I will try, I try to explain here in simple words, okay?
But is, um, operational excellence of the product, okay? They, they, they want to, to have, uh, a way or, um, output that everything that is in associated with a product is good, is consistent, et cetera. So observability in the scenario, what I'm seeing, observability, uh, is now the heart of products.
You are telling this because work with, uh, products, no, no, no, no. I'm telling this because the behavior of the good products leaders, the engineers that works with products, they are seeing that observability is, um, is, um, how can I say, is that the light that will guide then in Antonio? So I'm adding here, uh, some image from the, uh, the products YouTube channel, okay?
Uh, you can find entire, uh, reference in the product, uh, product chief that com VR is Brazilian side, but is the products and university that the, the people that works with, uh, this products area, what they are talking observability. We have a lot of disciplines when we talk about products, but the observability start to be so important that they are adding in the beginning of a planning when they start a project, project, okay? Why, why they are doing this?
Because everything now is associated with journeys. Journeys, uh, and the observability will provide the data and the insights that the team that are defining create the product. Think about everything needs.
So here it's a phrase that I added. Observability graphs gives products the data and insights required to align the customer experience with the operational goals, driving both performance innovation, think, uh, in this way we have a product. And you need to start to observe what you're create.
Imagine, uh, a new product not make sense. You create the observability definitions, what you observe after the end of your entire product definition. So observability needs to start in the beginning of the, the project.
Okay? I return this slide, and this is based on the experience of the professionals that are working in this world. And as a observability person, I totally agree, how many times you think, oh my God, they create everything, but nobody had an observability to, to know what's happening.
And after that, when you transfer the application or to another area that's responsible for, to, uh, to, how can I say, keep the environment safe, you have problems. So align, not have a bit since the beginning, okay? Uh, in here, uh, I talk, start to talk about the journeys, and that's the point in observability in the products words.
Sure. Ivan, about this framework, this products model, I, uh, is new framework, okay? Is new, but you go to product, we achieve, you can see the pictures, what the market is doing, okay?
And have a different view, okay? But product bring here the top observability, observability. So important that, that in some, um, uh, companies, they are working as a bubble to, not a bubble, but I think to provide, uh, answers to the product delivery team.
To the product operation team to keep the continuous discovery, because think you have a product you deliver, okay? How you can innovate your product if you are not observing. Well, the things, if you don't define the, the KPIs that can help you in, you know, you innovate because in the, in the end, we are talk about innovation, okay?
I'm not reading these slides. You, you have my slides, okay? And this slide is basically what I said.
Observability is became more and more important in the products area, okay? And share, when you create, uh, application, et cetera, you have the traditional, not traditional, the, the dev, uh, cycle, et cetera, um, flowing. So I talked here about the observability, um, revolution because again, sometimes we are so technical and I'm engineer, I assume, I'm so technical that when someone talks about observability, observability came from engineering, blah, blah, blah, that is the capability to see what exists.
Okay? But observability for the marketing now, um, it's the cap. It is the eyes of the, the, the people to see what's happening with the environment.
Applications, okay? I'm simplified, but observability, when I, I talk, oh, I work with observability. Ah, when it, so you know what's happening to the environment, you know what's happening, this application, yes.
And because I know sometimes, okay, but I think your application at sometime needs to, to receive some investment. Why I, I start to talk about this because I know the data, you know? And more and more we are in this tech technical world.
We understand that, oh, you need to buy innovate, okay? Name, this is a leg, okay? This is another conversation.
I talked about the chance, and I talked about observability in the products work. So let's talk, uh, about the part that sometimes the technical people forget that this, think from the technical to the business, okay? Again, if your boss match that your boss don't know nothing about observability, ask, um, I dunno.
Now your name is John. John, why I need observability in my business. Okay?
John, you have data that you can transform in decisions because if you go and explain, no, observe a bit, you provide metrics logs, they tell no, no, no, John will, uh, your boss John will say, no, no, no. But if we use our expertise and we request help from our colleagues in the company and create a good dashboard, we can say that we are supporting the business. And this, I'm talking here ways to do this, but in the end, it's a future that needs to be built in a company to have this collaboration between the areas.
Again, a good observability to IPM that isn't is blah, blah, blah, blah, blah, lead there, blah, blah, blah. And the gardener, they have a lot of features. Open source or a internal dev that was created by the guru in your company, okay?
But if we, we don't have this approach to talk with the boss, okay? The stakeholder, well, sometimes it is complicated to have the investments. And again, when we talk about money, okay, I'm technically, I like to talk about money.
We need to talk about operational cost reduction, okay? We know that good observability tools will help us to detect the problems we know technically you can test do P-O-C-P-O-V if you don't have a observability tool, okay? Will help if we have good observability tools, good observability tools configured by good people.
You have resource optimization. You can detect things that you are not using in the cloud, for example, ah, these are kind of infra that I can, I'm not using. So are you, uh, activate or reduce?
So you have this, this impact that flows, uh, personal goals to the leadership with a good message to investing in observability. The data always will provide us things to improve. Think in the products.
Imagine we returning for the guide that's defined the product. Hey, your products here, but based on in the observability KPIs metrics, what you define the, the product that you created needs to prove this. Man, this is amazing.
Okay? This is a, is more than technical, and observability empowers us to understand, uh, the connection between the business and operationals, et cetera. Sure.
Uh, customer satisfaction, everything that I added in the last, uh, bullet needs to be combined with other things. For example, when we talk about, uh, customer succession, we need to talk about other things, okay? I have the observability tool.
Do, uh, monitor monitoring, observing this part. I have this KPIs and share. If you combine this observability data with, um, a survey with, uh, other KPIs more associated with, uh, with sentimental analysis, you'll be good.
You know, because sometimes the customer are answer surveys, oh, I don't like this application, blah, blah, blah. And okay, it's only the heart of the customer talking. Imagine you combine the heart of the customer talking crying.
Oh, I don't like this application with you, the observability data. Why, uh, I, I say, Paul don't like the, the application always. When the, the customer send a survey, he, he cries.
Okay? So let's see what the experience that PO is having in the application. Observability data we need to combine again, is future, okay?
It's future. You talk with the areas associated. So, uh, observability is not just about how your system works, okay?
I know that we're technical, but the observability word became, uh, I will say a, uh, monster, but in a good expect because, okay, to, I was in a, uh, me, lemme talk one thing with you. I was in a, a secured event and the security guys now observability in the middle. Observability, observability.
Years ago, the security world was totally isolated. Years more than 10 years, okay? I'm not so young.
But understand that point. Everybody start to understand the, the importance of the observability in other areas. So we need to create the connection, the future, talk with the people more than technical.
And I'm finish my presentation, but again, your company is ready to change the future, to work integrated, to transform the, not only the DevOps world, but the business using the power of the observability. That's a question that are, you keep here to finalize my presentation, but again, if you don't go and talk with other areas or forget to change the mindset of your team, well, well, well, well, you'll not, you'll not achieve good results. So mic complete.
My name is je, my contacts are here. So I hope that you this session and see.