In the second session of the 'Unlocking Engineering Productivity' webinar by Typo, host Kovid Batra engages engineering leaders Cesar Rodriguez and Ariel Pérez in a conversation about building high-performing development teams.
Cesar, VP of Engineering at StackGen, shares insights on ingraining curiosity and the significance of documentation and testing. Ariel, Head of Product and Technology at Tinybird, emphasizes the importance of clear communication, collaboration, and the role of AI in enhancing productivity. The panel discusses overcoming common productivity misconceptions, addressing burnout, and implementing effective metrics to drive team performance. Through practical examples and personal anecdotes, the session offers valuable strategies for fostering a productive engineering culture.
Kovid Batra: Hi everyone, welcome to the second webinar session of Unlocking Engineering Productivity by Typo. I’m your host, Kovid, excited to bring you all new webinar series, bringing passionate engineering leaders here to build impactful dev teams and unlocking success. For today’s panel, we have two special guests. Uh, one of them is our Typo champion customer. Uh, he’s VP of Engineering at StackGen. Welcome to the show, Cesar.
Cesar Rodriguez: Hey, Kovid. Thanks for having me.
Kovid Batra: And then we have Ariel, who is a longtime friend and the Head of Product and Technology at Tinybird. Welcome. Welcome to the show, Ariel.
Ariel Pérez: Hey, Kovid. Thank you for having me again. It’s great chatting with you one more time.
Kovid Batra: Same here. Pleasure. Alright, um, so, Cesar has been with us, uh, for almost more than a year now. And he’s a guy who’s passionate about spending quality time with kids, and he’s, uh, into cooking, barbecue, all that we know about him. But, uh, Cesar, there’s anything else that you would like to tell us about yourself so that, uh, the audience knows you a little more, something from your childhood, something from your teenage? This is kind of a ritual of our show.
Cesar Rodriguez: Yeah. So, uh, let me think about this. So one of, one of the things. So something from my childhood. So I had, um, I had the blessing of having my great grandmother alive when I was a kid. And, um, she always gave me all sorts of kinds of food to try. And something she always said to me is, “Hey, don’t say no to me when I’m offering you food.” And that stayed in my brain till.. Now that I’m a grown up, I’m always trying new things. If there’s an opportunity to try something new, I’m always, always want to try it out and see how it, how it is.
Kovid Batra: That’s, that’s really, really interesting. I think, Ariel, , uh, I’m sure you, you also have some something similar from your childhood or teenage which you would like to share that defines who you are today.
Ariel Pérez: Yeah, definitely. Um, you know, thankfully I was, um, I was all, you know, reminded me Cesar. I was also, uh, very lucky to have a great grandmother and a great grandfather, alive, alive and got to interact with them quite a bit. So, you know, I think we know very amazing experiences, remembering, speaking to them. Uh, so anyway, it was great that you mentioned that. Uh, but in terms of what I think about for me, the, the things that from my childhood that I think really, uh, impacted me and helped me think about the person I am today is, um, it was very important for my father who, uh, owned a small business in Washington Heights in New York City, uh, to very early on, um, give us the idea and then I know that in the sense that you’ve got to work, you’ve got to earn things, right? You’ve got to work for things and money just doesn’t suddenly appear. So at least, you know, a key thing there was that, you know, from the time I was 10 years old, I was working with my father on weekends. Um, and you know, obviously, you know, it’s been a few hours working and doing stuff and then like doing other things. But eventually, as I got older and older through my teenage years, I spent a lot more time working there and actually running my father’s business, which is great as a teenager. Um, so when you think about, you know, what that taught me for life. Obviously, there’s the power of like, look, you’ve got to work for things, like nothing’s given to you. But there’s also the value, you know, I learned very early on. Entrepreneurship, you know, how entrepreneurship is hard, why people go follow and go into entrepreneurship. It taught me skills around actual management, managing people, managing accounting, bookkeeping. But the most important thing that it taught me is dealing with people and working with people. It was a retail business, right? So I had to deal with customers day in and day out. So it was a very important piece of understanding customers needs, customers wants, customers problems, and how can I, in my position where I am in my business, serve them and help them and help them achieve their goals. So it was a very key thing, very important skill to learn all before I even went to college.
Kovid Batra: That’s really interesting. I think one, Cesar, uh, has learned some level of curiosity, has ingrained curiosity to try new things. And from your childhood, you got that feeling of building a business, serving customers; that is ingrained in you guys. So I think really, really interesting traits that you have got from your childhood. Uh, great, guys. Thank you so much for this quick sweet intro. Uh, so coming to today’s main section which is about talking, uh, about unlocking engineering productivity. And today’s, uh, specifically today’s theme is around building that data-driven mindset around unlocking this engineering productivity. So before we move on to, uh, and deep dive into experiences that you have had in your leadership journey. First of all, I would like to ask, uh, you guys, when we talk about engineering productivity or developer productivity, what exactly comes to your mind? Like, like, let’s start with a very basic, the fundamental thing. I think Ariel, would you like to take it first?
Ariel Pérez: Absolutely. Um, the first thing that comes to mind is unfortunate. It’s the negative connotation around developer productivity. And that’s primarily because for so long organizations have trying to figure out how do I measure the productivity of these software developers, software engineers, who are one of my most expensive resources, and I hate the word ‘resource’, we’re talking about people, because I need to justify my spend on them. And you know what, they, I don’t know what they do. I don’t understand what they do. And I got to figure out a way to measure them cause I measure everyone else. If you think about the history of doing this, like for a while, we were trying to measure lines of code, right? We know we don’t do that. We’re trying to open, you know, we’re trying to, you know, measure commits. No, we know we don’t do that either. So I think for me, unfortunately, in many ways, the term ‘developer productivity’ brings so many negative associations because of how wrong we’ve gotten it for so long. However, you know, I am not the, I am always the eternal optimist. And I also understand why businesses have been trying to measure this, right? All these things are inputs into the business and you build a business to, you know, deliver value and you want to understand how to optimize those inputs and you know, people and a particular skill set of people, you want to figure out how to best understand, retain the best people, manage the best people and get the most value out of those people. The thing is, we’ve gotten it wrong so many times trying to figure it out, I think, and you know, some of my peers who discuss with me regularly might, you know, bash me for this. I think DORA was one good step in that direction, even though there’s many things that it’s missing. I think it leans very heavily on efficiency, but I’ll stop, you know, I’ll leave that as is. But I believe in the people that are behind it and the people, the research and how they backed it. I think a next iteration SPACE and trying to go to SPACE, moved this closer and tried to figure it out, you know, there’s a lot of qualitative aspects that we need to care about and think about. Um, then McKinsey came and destroyed everything, uh, unfortunately with their one metric to rule it all. And it was, it’s been all hell broke loose. Um, but there’s a realization and a piece that look, we, as, as a, as a, as an industry, as a role, as a type of work that we do, we need to figure out how we define this so that we can, you know, not necessarily justify our existence, but think about, how do we add value to each business? How do we define and figure out a better way to continually measure? How do we add value to a business? So we can optimize for that and continually show that, hey, you actually can’t live without us and we’re actually the most important part of your business. Not to demean any other roles, right? But as software engineers in a world where software is eating the world and it has eaten the world, we are the most important people in the, in there. We’re gonna figure out how do we actually define that value that we deliver. So it’s a problem that we have to tackle. I don’t think we’re there yet. You know, at some point, I think, you know, in this conversation, we’ll talk about the latest, the latest iteration of this, which is the core 4, um, which is, you know, things being talked about now. I think there’s many positive aspects. I still think it’s missing pieces. I think we’re getting closer. But, uh, and it’s a problem we need to solve just not as a hammer or as, as a cudgel to push and drive individual developers to do more and, and do more activity. That’s the key piece that I think I will never accept as a, as a leader thinking about developer productivity.
Kovid Batra: Great, I think that that’s really a good overview of how things are when we talk about productivity. Cesar, do you have a take on that? Uh, what comes to your mind when we talk about engineering and developer productivity?
Cesar Rodriguez: I think, I think what Ariel mentioned resonates a lot with me because, um, I remember when we were first starting in the industry, everything was seen narrowly as how many lines of code can a developer write, how many tickets can they close. But true productivity is about enabling engineers to solve meaningful problems efficiently and ensuring that those problems have business impact. So, so from my perspective, and I like the way that you wrote the title for this talk, like developer (slash) engineering. So, so for me, developer, when I think about developer productivity, that that brings to my mind more like, how are your, what do your individual metrics look like? How efficiently can you write code? How can you resolve issues? How can you contribute to the product lifecycle? And then when you think about engineering metrics, that’s more of a broader view. It’s more about how is your team collaborating together? What are your processes for delivering? How is your system being resilient? Um, and how do you deliver, um, outcomes that are impactful to the business itself? So I think, I think I agree with Ariel. Everything has to be measured in what is the impact that you’re going to have for the business because if you can’t tie that together, then, then, well, I think what you’re measuring is, it’s completely wrong.
Kovid Batra: Yeah, totally. I, I, even I agree to that. And in fact, uh, when we, when we talk about engineering and developer productivity, both, I think engineering productivity encompasses everything. We never say it’s bad to look at individual productivity or developer productivity, but the way we need to look at it is as a wholesome thing and tie it with the impact, not just, uh, measuring specific lines of code or maybe metrics like that. Till that time, it definitely makes sense and it definitely helps measure the real impact, uh, real improvement areas, find out real improvement areas from those KPIs and those metrics that we are looking at. So I think, uh, very well said both of you. Uh, before I jump on to the next piece, uh, one thing that, uh, I’m sure about that you guys have worked with high-performing engineering teams, right? And Ariel, you had a view, like what people really think about it. And I really want to understand the best teams that you have worked with. What’s their perception of, uh, productivity and how they look at, uh, this data-driven approach, uh, while making decisions in the team, looking at productivity or prioritizing anything that comes their way, which, which would need improvement or how is it going? How, how exactly these, uh, high-performing teams operate, any, any experiences that you would like to share?
Ariel Pérez: Uh, Cesar, do you want to start?
Cesar Rodriguez: Sure. Um, so from my perspective, the first thing that I’ve observed on high-performing teams is that is there is great alignment with the individual goals to what the business is trying to achieve. Um, the interests align very well. So people are highly motivated. They’re having fun when they’re working and even on their outside hours, they’re just thinking about how are you going to solve the problem that they’re, they’re working on and, and having fun while doing it. So that’s, that’s one of the first things that I observed. The other thing is that, um, in terms of how do we use data to inform the decisions, um, high-performing teams, they always use, consistently use data to refine processes. Um, they identify blockers early and then they use that to prioritize effectively. So, so I think all ties back to the culture of the team itself. Um, so with high-performing teams, you have a culture that is open, that people are able to speak about issues, even from the lowest level engineer to the highest, most junior engineers, the most highest senior engineer, everyone is treated equally. And when people have that environment, still, where they can share their struggles, their issues and quickly collaborate to solve them, that, that for me is the biggest thing to be, to be high-performing as a team.
Kovid Batra: Makes sense.
Ariel Pérez: Awesome. Um, and, you know, to add to that, uh, you know, I 1000% agree with the things you just mentioned that, you know, a few things came to mind of that, like, you know, like the words that come to mind to describe some of the things that you just said. Uh, like one of them, for example, you know, you think about the, you know, what, what is a, what is special or what do you see in a high-performing team? One key piece is there’s a massive amount of intrinsic motivation going back to like Daniel Pink, right? Those teams feel autonomy. They get to drive decisions. They get to make decisions. They get to, in many ways own their destiny. Mastery is a critical thing. These folks are given the opportunity to improve their craft, become better and better engineers while they’re doing it. It’s not a fight between ‘should I fix this thing’ versus ‘should I build this feature’ since they have autonomy. And the, you know, guide their own and drive their own agenda and, and, and move themselves forward. They also know when to decide, I need to spend more time on building this skill together as a team or not, or we’re going to build this feature; they know how to find that balance between the two. They’re constantly becoming better craftsmen, better engineers, better developers across every dimension and better people who understand customer problems. That’s a critical piece. We often miss in an engineering team. So becoming better at how they are doing what they do. And purpose. They’re aligned with the mission of the company. They understand why we do what we do. They understand what problem we’re solving. They, they understand, um, what we sell, how we sell it, whose problems to solve, how we deliver value and they’re bought in. So all those key things you see in high-performing teams are the major things that make them high-performing.
The other thing sticking more to like data and hardcore data numbers. These are folks that generally are continually improving. They think about what’s not working, what’s working, what should we do more of, what should we do less of, you know, when I, I forgot who said this, but they know how to turn up the good. So whether you run retros, whether you just have a conversation every day, or you just chat about, hey, what was good today, what sucked; you know, they have continuous conversations about what’s working, what’s not working, and they continually refine and adjust. So that’s a key critical thing that I see in high-performing teams. And if I want to like, you know, um, uh, button it up and finish it at the end is high-performing teams collaborate. They don’t cooperate, they collaborate. And that’s a key thing we often miss, which is and the distinction between the two. They work together on their problems, which one of those key things that allows them to like each other, work well with each other, want to go and hang out and play games after work together because they depend on each other. These people are shoulder to shoulder every day, and they work on problems together. That helps them not only know that they can trust each other, they can trust each other, they can depend on each other, but they learn from each other day in and day out. And that’s part of what makes it a fun team to work on because they’re constantly challenging each other, pushing each other because of that collaboration. And to me, collaboration means, you know, two people, three people working on the same problem at the same time, synchronously. It’s not three people separating a problem and going off on their own and then coming back together. You know, basically team-based collaboration, working together in real time versus individual work and pulling it together; that’s another key aspect that I’ve often seen in high-performing teams. Not saying that the other ways, I have not seen them and cannot be in a high-performing team, but more likely and more often than not, I see this in high-performing teams.
Kovid Batra: Perfect. Perfect. Great, guys. And in your journeys, um, there have been, there must have been a lot of experiences, but any counterintuitive things that you have realized later on, maybe after making some mistakes or listening to other people doing something else, are there any things which, which are counterintuitive that you learned over the time about, um, improving your team’s productivity?
Ariel Pérez: Um, I’ll take this one first. Uh, I don’t know if this is counterintuitive, but it’s something you learn as you become a leader. You can’t tell people what to do, especially if they’re high-performing, you’re improving them, even if you know better, you can’t tell them what to do. So unfortunately, you cannot lead by edict. You can do that for a short period of time and get away with it for a short period of time. You know, there’s wartime versus peacetime. People talk about that. But in reality, in many ways, it needs to come from them. It needs to be intrinsic. They’re going to have to be the ones that want to improve in that world, you know, what do you do as a leader? And, you know, I’ve had every time I’ve told them, do this, go do this, and they hated me for it. Even if I was right at the end, then even if it took a while and then they eventually saw it, there was a lot of turmoil, a lot of fights, a lot of issues, and some attrition because of it. Um, even though eventually, like, yes, you were right, it was a bit more painful way, and it was, you know, me and the purpose for the desire, you know, let me go faster. We got to get this done. Um, it needs to come from the team. So I think I definitely learned that it might seem counterintuitive. You’re the boss. You get to tell people to do. It’s like, no, actually, no, that’s not how it works, right? You have to inspire them, guide them, drive them, give them the tools, give them the training, give them the education, give them the desire and need and want for how to get there, have them very involved in what should we do, how do we improve, and you can throw in things, but it needs to come from them. If there were anything else I’d throw into that, it was counterintuitive, as I think about improving engineering productivity was, to me, this idea of that off, you know, as we think about from an accounting perspective, there’s just no way in hell that two engineers working on one problem is better than one. There’s no way that’s more productive. You know, they’re going to get half the work done. That’s, that’s a counterintuitive notion. If you think about, if you think about it, engineers as just mere inputs and resources. But in reality, they’re people, and that software development is a team sport. As a matter of fact, if they work together in real time, two engineers at the same time, or god forbid, three, four, and five, if you’re ensemble programming, you actually find that you get more done. You get more done because things, like they need to get reworked less. Things are of higher quality. The team learns more, learns faster. So at the end of the day, while it might feel slow, slow is smooth and smooth is fast. And they get just get more over time. They get more throughput and more quality and get to deliver more things because they’re spending less time going back and fixing and reworking what they were doing. And the work always continues because no one person slows it down. So that’s the other counterintuitive thing I learned in terms of improving and increasing productivity. It’s like, you cannot look at just productivity, you need to look at productivity, efficiency, and effectiveness if you really want to move forward.
Kovid Batra: Makes sense. I think, uh, in the last few years, uh, being in this industry, I have also developed a liking towards pair programming, and that’s one of the things that align with, align with what you have just said. So I, I’m in for that. Yeah. Uh, great. Cesar, do you have, uh, any, any learnings which were counterintuitive or interesting that you would like to share?
Cesar Rodriguez: Oh, and this goes back to the developer versus engineering, uh, conversation, uh, and question. So productivity and then something that’s counterintuitive is that it doesn’t mean that you’re going to be busy. It doesn’t mean that you’re just going to write your code and finish tickets. It means that, and this is, if there are any developers here listening to this, they’re probably going to hate me. Um, you’re going to take your time to plan. You’re going to take your time to reflect and document and test. Um, and we, like, we’ve seen this even at StackGen last quarter, we focused our, our, our efforts on improving our automated tests. Um, in the beginning, we’re just trying to meet customer demands. We, unfortunately, they didn’t spend much time testing, but last quarter we made a concerted effort, hey, let’s test all of our happy paths, let’s have automated tests for all of that. Um, let’s make sure that we can build everything in our pipelines as best as possible. And our, um, deployment frequency metrics skyrocketed. Um, so those are some of the, uh, some of the counterintuitive things, um, maybe doing the boring stuff, it’s gonna be boring, but it’s gonna speed you up.
Ariel Pérez: Yeah, and I think, you know, if I can add one more thing on that, right, that’s critical that many people forget, you know, not only engineers, as we’re working on things and engineering leadership, but also your business peers; we forget that the cost of software, the initial piece of building it is just a tiny fraction of the cost. It’s that lifetime of iterating, maintaining it, managing, building upon it; that’s where all the cost is. So unfortunately, we often cut the things when we’re trying to cut corners that make that ongoing cost cheaper and you’re, you’re right, at, you know, investing in that testing upfront might seem painful, but it helps you maintain that actual, you know, uh, that reasonable burn for every new feature will cost a reasonable amount, cause if you don’t invest in that, every new feature is more expensive. So you’re actually a whole lot less productive over time if you don’t invest on these things at the beginning.
Cesar Rodriguez: And it, and it affects everything else. If you’re trying to onboard somebody new, it’ll take more time because you didn’t document, you didn’t test. Um, so your cost of onboarding new people is going to be more expensive. Your cost of adding new people, uh, new features is going to be more expensive. So yeah, a hundred percent.
Kovid Batra: Totally. I think, Cesar, documentation and testing, uh, people hate it, but that’s the truth for sure. Great, guys. I think, uh, there is more to learn on the journey and there are a lot more questions that I have and I’m sure audience would also have a lot of questions. So I would request the audience to put in their questions in the comment section right now, because at the end when we are having a Q&A, we’ll have all the questions sorted and we can take all of them one by one. Okay. Um, as I said, like a lot of learning and unlearning is going to happen, but let’s talk about some of, uh, your specific experiences, uh, learn some practical tips from there. So coming to you, Ariel. Uh, you have recently moved into this leadership role at Tinybird. Congratulations, first of all.
Ariel Pérez: Thank you.
Kovid Batra: And, uh, I’m sure this comes with a lot of responsibility when you enter into a new environment. It’s not just a new thing that you’re going to work upon, it’s a whole new set of people. I’m sure you have seen that in your career multiple times. But every time you step in and you’re a new person there, and of course, uh, you’re going as a leader, uh, it could be overwhelming, right? Uh, how do you manage that situation? How do you start off? How do you pull off so that you actually are able to lead, uh, and, and drive that impact which you really want?
Ariel Pérez: Got it. Um, so, uh, the first part is one of, this may sound like fluff, but it really helps, um, in many ways when you have a really big challenge ahead, you know, you have to avoid, you have to figure out how to avoid letting imposter syndrome freeze you. And even if you’ve had a career of success, you know, in many ways, imposter syndrome still creeps up, right? So how do I fight, how do I fight that? It’s one of those things like stand in front of the mirror and really deep breaths and talk about I got this job for a reason, right? I, you know, I, I, they, they’re trusting me for a reason. I got here. I earned this. Here’s my track record. I worked this. Like I deserve to be here. I’m supposed to be here. I think that’s a very critical piece for any new leader, especially if you’re a new leader in a new place, because you have so much novelty left and right. You have to prove yourself and that’s very daunting. So the first piece is you need to figure out how to get yourself out of your own head. And push yourself along and coach yourself, like I’m supposed to be here, right? Once you get that piece, you know down pat, it really helps in many ways helps change your own mindset your own framing. When you’re walking into conversations walking into rooms, there’s a big piece of how, how that confidence shines through. That confidence helps you speak and get your ideas and thoughts out without tripping all over yourself. That confidence helps you not worry about potentially ruffling some feathers and having hard conversations. When you’re in leadership, you have to have hard conversations. It’s really important to have that confidence, obviously without forgetting it, without saying, let me run over everybody, cause that’s not what it means, but it just means you got to get over the piece that freezes you and stops you. So that’s the first piece I think. The second piece is, especially when moving higher and higher into positions of leadership; it’s listening. Listening is the biggest thing you do. You might have a million ideas, hold them back, please hold them back. And that’s really hard for me. It’s so hard cause I’m like, “I see that I can fix that. I can fix that too. I’ve seen that before I can fix it too.” But, you know, you earn more respect by listening and observing. And actually you might learn a few things or two. I’m like, “Oh, that thing I wanted to fix, there’s a reason why it’s the way it is.” Because every place is different. Every place has a different history, a different context, a different culture, and all those things come into play as to why certain decisions were made that might seem contrary to what you would have done. And it helps you understand that context. That context is critical, not only to then figure out the appropriate solution to the problem, but also that time while you’re learning and listening and talking to people, you’re building relationships with people, you’re connecting to people, you’re understanding, you’re understanding the players, understanding who does well, who doesn’t do well, you’re understanding where all the bodies are buried, you’re understanding the strategy, you’re getting a big picture of all the things so that then when it comes time to say now time to implement change, you have a really good setup of who are the people that are gonna help me make the change, who are the people that are going to be challenging, how do I draw a plan to do change management, which is a big important thing. Change management is huge. It’s 90% people. So you need to understand the people and then understand, it also gives you enough time to understand the business strategy, the context, the big problem where you’re going to kind of be more effective at. Here’s why I got hired. Now I’m going to implement the things to help me execute on what I believe is the right strategy based on learning and listening and keeping my mouth shut for the time, right? Now, traditionally, you’ll hear this thing about 90 days. I think the 90 days is overly generous if you’re in a really big team, I think it leans and skews toward big places, slower moving places, um, and, and places that move. That’s it. Bigger places, slower places. When you join a startup environment, we join a smaller company. You need to be able to move faster. You don’t have 90 days to make decisions. You don’t have 90 days. You might have 30 days, right? You want to push that back as far as you can to get an appropriate context. But there’s a bias for action, reasonably so because you’re not guaranteed that the startup is going to be there tomorrow. So you don’t have 90 days, but you definitely don’t want to do it in two weeks and probably not start doing things in a month.
Kovid Batra: Makes sense. Makes sense. So, uh, a follow-up question on that. Uh, when you get into this position, if you are in a startup, let’s say you get 30 to 45 days, but then because of your bias towards action, you pick up initiatives that you would want to lead and create that impact. In your journey at Tinybird, have you picked up something, anything interesting, maybe related to AI or maybe working with different teams that you think would work on your existing code base to revamp it, anything that you have picked up and why?
Ariel Pérez: Yeah, a bunch of stuff. Um, I think when I first joined Tinybird, my first role was field CTO, which is a role that takes the, the, the responsibilities of the CTO and the external facing aspects of them. So I was focused primarily on the market, on customers, on prospects. And as part of that one, you know, one of the first initiatives I had was how do we, uh, operate within the, you know, sales engineering team, who was also reporting to me, and make that much more effective, much more efficient. So a few of the things that we were thinking of there were, um, AI-based solutions and GenAI-based solutions to help us find the information we need earlier, sooner, faster. So that was more like an optimization and efficiency thing in terms of helping us get the answers and clarify and understand and gather requirements from customers and very quickly figure out this is the right demo for you, these are the right features and capabilities for you, here’s what we can do, here’s what we can’t do, to get really effective and efficient at that. When moving into a product role though, and product and engineering role, in terms of the, the latest initiatives that I’ve picked up, like there, there, there, there are two big things in terms of themes. One of them is that Tinybird must always work, which sounds like, yeah, well, duh, obviously it must always work, but there’s a key piece underpinning that. Number one, obviously the, you know, stability and reliability are huge and required for trust from customers wanting to use you as a dev tool. You need to be able to depend on it, but there’s another piece is anything I must do and try to do on the platform, it must fail in a way that I understand and expect so that then I can self service it and fix it. So that idea of Tinybird always works that I’ve been picking up and working on projects is transparency, observability, and the ability for customers to self-service and resolve issues simply by saying, “I need more resources.” And that’s a, you know, it’s a very challenging thing because we’ve got to remove all the errors that have nothing to do with that, all the instability and all the reliability problems so that those are granted. And then remaining should only be issues that, hey, customer, you can solve this by managing your limits. Hey, customer, you can solve this by increasing the cores you’re using. You can solve this by adding more memory and that should be the only thing that remains. So working on a bunch of stuff there on predicting whether something will fail or not, predicting whether something is going to run out of resources or not, very quickly identifying if you’re running out of resources so there’s almost like an SRE monitoring observability aspect to this, but turning that back into a product solution. That’s one side of it. And then the other big pieces will be called a developer’s experience. And that’s something that my, you know, my, my, my peer is working internally on and leading is a lot more about how developers develop today. Developers develop today, well, they always develop locally. They prefer not depending on IO on a network, but developer, every developer, whether they tell you yes or no, is using an AI assistant; every developer, right? Or 99% of developers. So the idea is, how do we weave that into the experience without making it be, you know, a gimmick? How do we weave an AI Copilot into your development experience, your local development experience, your remote development experience, your UI development experience so that you have this expert at your disposal to help you accelerate your development, accelerate your ability to find problems before you ship? And even when you ship, help you find those problems there so you can accelerate those cycles, so you can shorten those lead time, so you can get to productivity and a productive solution faster with less errors and less issues. So that’s one major piece we’re working on there on the embedding AI; and not just AI and LLMs and GenAI, all you think about, even traditional. I say traditional like ML models on understanding and predicting whether something’s going to go wrong. So we’re working on a lot of that kind of stuff to really accelerate the developer, uh, accelerate developer productivity and engineering team productivity, get you to ship some value faster.
Kovid Batra: Makes sense. And I think, uh, when you’re doing this, is there any kind of framework, tooling or processes that you’re using to measure this impact, uh, over, over the journey?
Ariel Pérez: Yeah, um, for this kind of stuff, I lean a lot more toward the outcomes side of the equation, you know, this whole question of outputs versus outcomes. I do agree. John Cutler, very recently, I loved listening to John Cutler. He very recently published something like, look, we can’t just look at outcomes, because unfortunately, outcomes are lagging. We need some leading indicators and we need to look at not only outcomes, but also outputs. We need to look at what goes into here. We need to look at activity, but it can’t be the only thing we’ll look at. So the things I look at is number one, um, very recently I started working with my team to try to create our North Star metric. What is our North Star metric? How do we know that what we’re doing and what we’re solving for is delivering value for our customers? And is that linked to our strategy and our vision? And do we see a link to eventual revenue, right? So all those things, trying to figure out and come up with that, working with my teams, working, looking at our customers, understanding our data, we’ve come up with a North Star metric. We said, great, everything we do should move that number. If that moving, if that number is moving up into the right, we’re doing the right things. Now, looking at that alone is not enough, because especially as engineering teams, I got to work back and say, how efficient are we at trying to figure that out? So there’s, you know, a few of the things that I look at, I obviously look at the DORA metrics. I do look at them because they help us try to figure out sources of issues, right? What’s our lead time? What’s our cycle time? What’s our deployment frequency? What’s our, you know, what, you know, what, what’s our, uh, you know, change failure rate? What’s our mean time to recover? Those are very critical to understand. Are we running as a tip-top shop in terms of engineering? How good are we at shipping the next thing? Because it’s not just shipping things faster; it’s if there’s a problem, I need to fix it really fast. If I want to deliver value and learn, and this is the second piece is critical that many companies fail is, I need to put it out in the hands of customers sooner. That’s the efficiency piece. That’s the outputs. That’s the, you know, are we getting really good at putting it in front of customers, but the second piece that we must need independent of the North Star metric is ‘and what happened’, right? Did it actually improve things? Did it, did it make things worse. So it’s optimizing for that learning loop on what our customers are doing. Do we have.. I’m tracking behavioral analytics pieces where the friction points were funnels. Where are they dropping off? Where are they circling the wheels, right? We’re looking at heat maps. We’re looking at videos and screen shares of like, what did the customer do? Why aren’t they doing what they thought we thought they were going to do? So then now when we learn this, go back to that really awesome DORA numbers, ship again, and let’s see, let’s see, let’s fix on that. So, to me, it’s a comprehensive view on, are we getting really good at shipping? And are we getting really good at shipping the right thing? Mixing both those things driven by the North star metric. Overall, all the stuff we’re doing is the North star moving up into the right.
Kovid Batra: Makes sense. Great. Thanks, Ariel. Uh, this was really, really insightful. Like, from the point you enter as a leader, build that listening capability, have that confidence, uh, driving the initiatives which are right and impactful, and then looking at metrics to ensure that you’re moving in the right direction towards that North Star. I think to sum up, it was, it was really nice and interesting. Cesar, I think coming to your experience, uh, you have also had a good stint at, uh, at StackGen and, uh, you were mentioning about, uh, taking up this transition successfully, uh, which was multi-cloud infrastructure that expanded your engineering team. Uh, right? And I would want to like deep dive into that experience. Uh, you specifically mentioned that, uh, that, uh, transition was really successful, and at that time, you were able to keep the focus, keep the productivity in place. How things went for you, let’s deep dive into that experience of yours.
Cesar Rodriguez: Yeah. So, so from my perspective, the goals that you are going to have for your team are going to be specific to where the business is at, at that point in time. So, for example, StackGen, we started in 2023. Initially, we were a very small number of engineers just trying to solve the initial problem, um, which we’re trying to solve with Stackdn, which is infrastructure from code and easily deploying cloud architecture into, into the cloud environment. Um, so we focus on one cloud provider, one specific problem, with a handful of engineers. And once we started learning from customers, what was working, what was not working, um, and we started being pulled into different directions, we quickly learned that we needed to increase engineering capacity to support additional clouds, to deliver additional features faster. Um, our clients were trying to pull us in different directions. So that required, uh, two things. One is, um, hiring and scaling the team quickly. So now we are, at the moment we’re 22 engineers; so hiring and scaling the engineering team quickly and then enabling new team members to be as productive as possible in day zero. Um, and this is where, this is where the boring, the boring actions come into play. Um, uh, so first of all, making sure that you have enough documentation so somebody can get up and running on day one, um, and they can start doing pull requests on day one. Second of all, making sure that you have, um, clear expectations in terms of quality and what is your happy path, and how can you achieve that. And third, um, is making sure everyone knows what is expected from them in terms of the metrics that we’re looking for and, uh, the quality that we’re looking for in their outcomes. And this is something that we use Typo for. So, for example, we have an international team. We have people in India, Portugal, US East Coast, US West Coast. And one of the things that we were getting stuck early on was our pull requests were getting opened, but then it took a really long time for people to review them, merge them, and take action and get them deployed. So, um, we established a metric, and we did this using Typo, where we were measuring, hey, if you have a pull request open more than 12 hours, let’s create an alert, let’s alert somebody, so that somebody can be on top of that. We don’t want to get somebody stuck for more than a working day, waiting for somebody to review the pull request. And, and the other metric that we look at, which is deployment frequency, we see that an uptick of that. Now that people are not getting stuck, we’re able to have more frictionally, frictionless, um, deployments to our SDLC where people are getting less stuck. We’re seeing collaboration between the team members regardless of their time zone improving. So that’s something actionable that we’ve implemented.
Kovid Batra: So I think you’re doing the boring things well and keeping a good visibility on things, how they’re proceeding, really helped you drive this transition smoothly, and you were able to maintain that productivity in the team. That’s really interesting. But touching on the metrics part again, uh, you mentioned that you were using Typo. Uh, there, there are, uh, various toolings to help you, uh, plan, execute, automate, and reflect things when you are, when you are into a position where as a leader, uh, you have multiple stakeholders to manage. So my question to both of you, actually, uh, when we talk about such toolings, uh, that are there in the market, like Typo, how these tools help you exactly, uh, in each of these phases, or if you’re not using such tools, you must be using some level of metrics, uh, to actually, let’s say you’re planning an initiative, how do you look at numbers? If you’re automating something and executing something, how do you look at numbers and how does this whole tooling piece help you in that? Um, yeah.
Cesar Rodriguez: I think, I think for me, the biggest thing before, uh, using a tool like Typo was it was very hard to have a meaningful conversation on how the engineering team was performing, um, without having hard, hard data and raw data to back it up. So, um, the conversation, if you don’t, if you’re not measuring things, it’s more about feelings and more about anecdotal evidence. But when you have actual data that you can observe, then you can make improvements, and you can measure how, how, how that, how things are going well or going bad and take action on it. So, so that’s the biggest, uh, for me, that’s the biggest benefit for, from my perspective. Um, you have, you can have conversations within your team and then with the rest of the organization, um, and present that in a, in a way that makes sense for everyone.
Kovid Batra: Makes sense. I think that’s the execution part where you really take the advantage of the tool. You mentioned with one example that you had set a goal for your team that okay, if the review time is more than 12 hours, you would raise an alert. So, totally makes sense, that helps you in the execution, making it more smooth, giving you more, uh, action-driven, uh, insights so that you can actually make teams move faster. Uh, Ariel, for you, any, any experiences around that? How do you, uh, use metrics for planning, executing, reflecting?
Ariel Pérez: So I think, you know, one of the things I like doing is I like working from the outside in. By that I mean, first, let me look at the things that directly impact customers, that is visible. There’s so much there on, you know, in terms of trust to customers. There’s also someone’s there on like actual eventual impact. So I lay looking, for example, um, the, it may sound negative, but it’s one of those things you want to track very closely and manage and then learn from is, what’s our incident number? Like, how many incidents do we have? You know, how many P0s? How many P1s? That is a very important metric to trust because I will guarantee you this, if you don’t have that number as an engineering leader, your CEO is going to try to figure out, hey, why are we having so many problems? Why are so many customers angry calling me? So that’s a number you’re going to want to have a very strong pulse on: understand incidents. And then obviously, take that number and try to figure out what’s going on, right? There’s so much behind it. But the first part is understand the number and you want that number to go down over time. Um, obviously, like I said, there’s a North star metric. You’re tracking that. Um, I look at also, which, you know, I don’t lean heavily on these, but they’re still used a lot and they’re still valuable. Things like NPS and CSAT to help you understand how customers are feeling, how customers are thinking. And it allows me to get often when it’s paired with qualitative feedback, even more so because I want to understand the ‘why’ and I’ll dive more into the qualitative piece, how critical is it and how often we forget that piece when we’re chasing metrics and looking for numbers, especially we’re engineers, we want numbers. We need a story and the story, you can’t get the story just from the numbers. So I love the qualitative aspect. And then the third thing I look at is, um, SCIs or failed customer interactions, trying to find friction in the journeys. What are all the times a customer tries to do something and they fail? And you know, you can define that in so many kinds of ways, but capturing that is one of those things you try to figure out. Find failed customer interactions, find where customers are hitting friction points, and let’s figure out which of those are most important to attack. So these things help guide, at the minimum, what do we need to work on as a team? Right? What are the things we need to start focusing on to deliver and build? Like, how do I get initiatives? Obviously, that stuff alone doesn’t turn into initiatives. So the next thing I like ensuring and I drive to figure out what we work on is with all my leaders. And in our organization, we don’t have separate product managers. You know, engineering leaders are product managers. They have to build those product skills because we have such a technical product that we decided to make that decision, not only for efficiency’s sake and stop having two people in every conversation, but also to build up that skill set of ‘I’m building for engineers, and I need to know my engineering product very well, but now let me enable these folks with the frameworks and methodologies, the ideas and the things that help them make product decisions.’ So, when we look at these numbers, we try to look at what are some frameworks and ways to think about what am I going to build? Which of these is going to impact? How much do we think it’s going to impact it? What level of confidence do I have in that? Does that come from the gut? Does that come from several opinions that customers tell us that, is the data telling us that, are competitors doing it? Have we run an experiment? Did we do some UX research? So the different levels of, uh, confidence in I want to do this thing. Cause this thing’s going to move that number. We believe that number is important. The FCI is it through the roof. I want to attack them. This is going to move it. Okay. How sure are you? He’s going to move it. Now, how are we going to measure that? And indeed moved it. Then I worked, so that’s the outside of the onion. Then I work inward and say, great, how good are we at getting at those things? So, uh, there’s two combinations of measures. I pull measures and data from, from GitLab, from GitHub, I look at the deployments that we have. Thankfully, we run a database. We have an OLAP database, so I can run a bunch of metrics off of all this stuff. We collect all this data from all this telemetry from our services, from our deployments, from our providers for all of the systems we use, and then we have these dashboards we built internally to track aggregates, track metrics and track them in real time, because that’s what Tinybird does. So, we use Tinybird to Tinybird while we Tinybird, which is awesome. So I, we’ve built our own back dashboards and mechanisms to track a lot of these metrics and understand a lot of these things. However, there’s a key piece which I haven’t introduced yet, but I have a lot of conversations with a lot of people on, hey, why did this number move? What’s going on? I want to get to the place that we actually introduce surveys. Funny enough, when you talk about the beginning of DORA, even today, DORA says, surveys are the best way to do this. We try to get hard data, but surveys are the best way to get it. For me, surveys really help, um, forget for a second what the numbers are telling me, how do the engineers feel? Because then I get to figure out why do you feel that way? It allows me to dive in. So that’s why I believe the qualitative subjective piece is so important to then bolster the numbers I’m seeing, either A: explain the numbers, or the other way around, when I see a story, I’m like, do the numbers back up that story? The reality is somewhere in the middle, but I use both, both of those to really help me.
Kovid Batra: Makes sense. Makes sense. Great guys. I think, uh, thank you. Thank you so much for sharing such good insights. I’m sure our audience has some questions for us, uh, so we can break in for a minute and, uh, then start the QnA.
Kovid Batra: All right. I think, uh, we have a lot of questions there, but I’m sure we are going to pick a few of them. Let’s start with the first one. That’s from Vishal. Hi Ariel, how do you, how do I decide which metrics to focus while measuring teams productivity and individual metrics? So I think the question is simple, but please go ahead.
Ariel Pérez: Um, I would start with in terms of, I would measure the core four of DORA at the minimum across the team to help me pinpoint where I need to go. I would start with that to help me pinpoint. In terms of which team productivity metrics or individual productivity metrics, I’d be very wary of trying to measure individual productivity metrics, not because we shouldn’t hold individuals accountable for what they do, not because individuals don’t also need to understand, uh, how we think about performance, how we manage that performance, but for individuals, we have to be very careful, especially in software teams. Since it’s a team sport, there’s no individual that is successful on their own, and there’s no individual that fails on their own. So if I were to think, you know, if I were to measure and try to figure out how to identify how this individual is doing, I would, I would look for at least two things. Number one, actual peer feedback. How, how do their peers think about this person? Can they depend on this person? Is this person there when they need him? Is this person causing a lot of problems? Is this person fixing a lot of problems? But I’d also look at the things, to me, for the culture I want to build, I want to measure how often is this person reviewing other people’s PRs? How often is this person sitting with other people, helping unblock them? How often is this person not coding because they’re going and working with someone else to help unblock them? I actually see that as a positive. Most frameworks will ding that person for inactivity. So I try to find the things that don’t measure activity, but are measuring that they’re doing the right things, which is teamwork. They’re actually being effective at working in a team when it comes to individuals.
Kovid Batra: Great. Thanks, Ariel. Uh, next question. That’s for you, Cesar. How easy or hard is the adoption and implementation of SEI tools like Typo? Okay, so you can share your experience, how it worked out for you.
Cesar Rodriguez: So, so two things. So, so when I was evaluating tools, um, I prefer to work with startups like Typo because they’re extremely responsive. If you go to a big company, they’re not going to be as responsive and as helpful as a startup is. They change the product to meet your expectations and they work extremely fast. So that’s the first thing. Um, the hard part of it is not about the technology itself. The technology is easy. The hard part is the people aspect of it. So you have to, if you can implement it early, uh, when your company is growing, that’s better because they’ll, when new team members come in, they already know what are the expectations and what to expect. The other thing is, um, you need to communicate effectively to your team members why are you using this tool, and getting their buy-in for measuring. Some people may not like that you’re going to be measuring their commits, their pull requests, their quality, their activity, but if you have a conversation with, with those people to make them understand the ‘why’ and how can you connect their productivity to the business outcomes, I think that goes far along. And then once you’re, once you’re in place, just listening to your engineers feedback about the tool, working with the vendor to, to modify anything to fit your company’s need, um, a lot of these tools are very cookie cutter in their approach, um, and have a set of, set of capabilities, but teams are made of people and people have different needs. So, so make sure that you capture that feedback, give it to your vendor and work with them to make the tool work for your specific individual teams.
Kovid Batra: Makes sense. Next question. That’s from, uh, Mohd Helmy Ibrahim, uh, Hi Ariel, how to make my senior management and junior implement project management software in their work, tasking to be live update tracking status update?
Ariel Pérez: Um, I, that one, I’m of two minds cause only because I see a lot of organizations who can get really far without actual sophisticated project management tooling. Like they just use, you know, Linear and that’s it. That’s all enough. Other places can’t live without, you know, a super massive, complex Jira solution with all kinds of things and all kinds of bells and whistles and reports. Um, I think the key piece here that’s important and actually it was funny enough. I was literally just having this conversation with my leadership team, my engineering leadership team. It’s this, you know, when it comes to the folks involved is do you want to spend all day asking, answering questions about where is this thing, how is this thing doing, is this thing going to finish, when is it going to finish, or do you want to just get on with your work, right? If you want to just get on with your work and actually do the work rather than talk about the work to other people who don’t understand it, if you want to find out when you want to do it, you need some level of information radiator. Information reader, radiators are critical at the minimum so that other folks can get on the same page, but also if someone comes to you and says, Hey, where is this thing? Look at the information radiator. It’s right there. You, where’s the status on the, it’s on the information radiator. When’s this going to be done? Look at the information radiator, right? That’s the key piece for me is if you don’t want to constantly answer that question, then you will, because people care about the things you’re working on. They want to know when they can sell this thing or they want to know so they can manage their dependencies. You need to have some level, some minimum level of investment of marking status, marking when you think it’s going to be done and marking how it’s going. And that’s a regular piece. Write it down. It’s so much easier to write it down than to answer that question over and over again. And if you write it down in a place that other people can see it and visualize it, even better.
Kovid Batra: Totally makes sense. All right, moving on. Uh, the next question is for Cesar from Saloni. Uh, good to see you here. I have a question around burnout. How do you address burnout or disengagement while pushing for high productivity? Oh, very relevant question, actually.
Cesar Rodriguez: Yeah, so so for this one, I actually use Typo as well. Um, so Typo has this gauge to, um, that tells you based on the data that it’s collecting, whether somebody is working higher than expected or lower than expected. And it gives you an alert saying, hey, this person may be prone to burnout or this person is burning out. Um, so I use that gauge to detect how is the team doing and it’s always about having a conversation with the individual and seeing what’s going on with their lives. There may be, uh, work things that are impacting their productivity. There may be things that are outside of work that are impacting that individual’s productivity. So you have to work around that. Um, we are, uh, it’s all about people in the end, um, and working with them, setting the right expectations and at the same time being accommodating if they’re experiencing burnout.
Kovid Batra: Cool. I think, uh, more than myself, uh, you have promoted Typo a lot today. Great, but glad to know that the tool is really helping you and your team. Yeah. Next question. Uh, this one is again for you, Cesar from Nisha. Uh, how do you encourage accountability without micromanaging your team?
Cesar Rodriguez: I think, I think Ariel answered this question and I take this approach even with my kids. Um, it’s not about telling them what to do. It’s about listening and helping them learn and come to the same conclusion as you’re coming to without forcing your way into it. So, so yeah, you have to listen to everybody, listen to your stakeholders, listen to your team, and then help them and drive a conversation that can point them in the right direction without forcing them or giving them the answer which is, which requires a lot of tact.
Ariel Pérez: One more thing I’ll add to that, right, is, you know, so that folks don’t forget and think that, you know, we’re copping out and saying, hold on, what’s your job as a leader? What are you accountable for? Right? In that part, there’s also like, our job is let them know what’s important. It’s our job to tell them what is the most important thing, what is the most important thing now, what is the most important thing long term, and repeat that ad hominem until they make fun of you for it, but they need to understand what’s the most important, what’s the strategy, so you need to provide context, because there’s a piece of, it’s almost like, it’s unfair, and it’s actually, I think, very, um, it’s a very negative thing to say, go figure it out, without telling them, hold on, figure what out? So that’s a key piece there as well, right? It’s you, you’re accountable as the leader for telling them what’s important, letting them understand why this is important, providing context.
Kovid Batra: Totally. All right. Next one. This one’s for you, Cesar. According to you, what are the most common misconceptions about engineering productivity? How do you address them?
Cesar Rodriguez: So, so I think the, for me, the biggest thing is people try to come with all these new words, DORA, SPACE, uh, whatever latest and greatest thing is. Um, the biggest thing is that, uh, there’s not going to be a cookie cutter approach. You have to take what works from those frameworks to your specific team in your specific situation of your business right now. And then from there, you have to look at the data and adapt as your team and as your business is evolving. So that’s, that’s the biggest. misconception for me. Um, you can take, you can learn a lot from the things that are out there, but always keep in mind that, um, you have to put that into the context of your current situation.
Kovid Batra: I think, uh, Ariel, I would like to hear you on this one too.
Ariel Pérez: Yeah. Uh, definitely. Um, I think for me, one of the most common misconceptions about engineering productivity as a whole is this idea that engineering is like manufacturing. And for so long, we’ve applied so many ideas around, look, engineering is all about shipping more code because just like in a fan of factory, let’s get really good at shipping code and we’re going to be great. That’s how you measure productivity. Ship more code, just like ship more widgets. How many widgets can I ship per, per hour? That’s a great measure of engineering productivity in a factory. It’s a horrible measure of productivity in engineering. And that’s because many people, you know, don’t realize that engineering productivity and engineering in particular, and I’m gonna talk development, as a piece of development, is it’s more R&D than it is like doing things than it’s actual shipping things. Software development is 99% research and development, 1% actually coding the thing. And if they want any more proof of that is if you have an engineer working on something or a team working on something for three weeks and somehow it all disappears and they lose all of it, how long will it take them to recode the same thing? They’ll probably recode the same thing in about a day. So that tells you that most of those three weeks was figuring out the right thing, the right solution, the right piece, and then the last piece was just coding it. So I think for me, that’s the big misconception about engineering productivity, that it has anything to do with manufacturing. No, it has everything to do with R&D. So if we want to understand how to better measure engineering productivity, look at industries where R&D is a very, very heavy piece of what they do. How do they measure productivity? How did they think about productivity of their R&D efforts?
Kovid Batra: Cool. Interesting. All right. I think with that, uh, we come to the end of this session. Before we part, uh, I would like to thank both of you for making this session so interesting, so insightful for all of us. And thanks to the audience for bringing up such nice questions. Uh, so finally, before we part, uh, Ariel, Cesar, anything you would say as parting thoughts?
Ariel Pérez: Cesar, you wanna go first?
Cesar Rodriguez: No, no, um, no, no parting thoughts here. Feel free to, anyone that wants to chat more, feel free to hit me up on LinkedIn. Check out stackgen.com if you want to learn about what we do there.
Ariel Pérez: Awesome. Um, for me, uh, in terms of parting thoughts is; and this is just because how I’ve personally thought about this is, um, I think if you lean on the figuring out what makes people tick and figure, and you’re trying to take your job from the perspective of how do I improve people, how to enrich people’s lives, how do I make them better at what they do every day? If you take it from that perspective, I don’t think you can ever go wrong. If you make your people super happy and engaged and they want to be here and you’re constantly motivating them, building them and growing them, as a consequence, the productivity, the outputs, the outcomes, all that stuff will come. I firmly believe that. I’ve seen it. I firmly believe it. It really, it would be really hard to argue that with some folks, but I firmly believe it. So that’s my parting thoughts, focus on the people and what makes them tick and what makes them work, everything else will fall into place. And if I, you know, just like Cesar, I can’t walk away without plugging Tinybird. Tinybird is, you know, data infrastructure for software teams. You want to go faster, you want to be more productive, you want to ship solutions faster and for the customers, Tinybird is, is built for that. It helps engineering teams build solutions over analytical data faster than anyone else without adding more people. You can keep your team smaller for longer because Tinybird helps you get that efficiency, that productivity out there.
Kovid Batra: Great. Thank you so much guys and all the best for your ventures and for the efforts that you’re doing. Uh, we’ll see you soon again. Thank you.
Cesar Rodriguez: Thanks, Kovid.
Ariel Pérez: Thank you very much. Bye bye.
Cesar Rodriguez: Thank you. Bye!
Implement engineering metrics, gather SDLC insights & set continuous improvement goals in your dev teams.
Try Live Demo
