How to bring startup speed to big organizations
A startup founder joins a 10,000-person company to build AI from scratch and learns how to move fast without losing trust.
Moving fast inside a big company can feel impossible. You’re hired to shake things up - but if you move too quickly, you lose people. This episode dives into that tension. Atlassian’s Head of AI, Nisha Iyer, shares what happens when a startup founder joins a 10,000-person company and tries to build new AI products from the inside. She unpacks learning to slow down, earning trust, and building momentum the right way - without giving up the spark that made you effective in the first place.
Executives and operators will relate to the tension between urgency and patience - and take away ideas for leading change without burning out your team or losing trust.
00:00 — Moving fast meets resistance
04:00 — “Learn the business”: slowing down to speed up
09:00 — The big swing that fell flat — and the lesson in humility
17:00 — Piloting change and earning trust one team at a time
22:00 — Turning internal wins into customer-facing AI
26:00 — Balancing empathy, momentum, and trust
31:00 — Reducing fear, sparking excitement: leading through change
36:00 — What’s next: voice tech, connection, and being more human
Shoutouts:
The First 90 Days — https://www.amazon.com/First-90-Days-Strategies-Expanded/dp/1422188612
Atlassian — https://www.atlassian.com
Rovo — https://www.atlassian.com/software/rovo
Follow Nisha Iyer: https://www.linkedin.com/in/nisha-ranjani-iyer/
Episode links: YouTube, Spotify, Apple Podcasts
Full Transcript
Nisha Iyer (00:00)
I came in with the idea that, one, I need to show everyone my skill set and how fast I can move and how many hats I can wear and how much I can push change.
Understanding that this rapid change causes a lot of tension and anxiety, right? And what am I doing to ensure that I am helping people understand how to work with the change versus fear of the change?
Keith Cowing (00:32)
This is Executives Unplugged. I am Keith Cowing, your host and executive coach. And with me today is Nisha Iyer. Nisha is a startup founder and entrepreneur by background, and she is now Head of AI at Atlassian, where she is launching products internally and externally that have tens of millions of dollars’ worth of impact. And today she will walk us through how to launch zero-to-one products in an entrepreneurial way at large, complex organizations with global impact.
Keith Cowing (01:01)
Nisha, welcome.
Nisha Iyer (01:02)
Thank you. Good to be here.
Keith Cowing (01:04)
Awesome to have you on the podcast. And we were chatting before we got started. You've had a really interesting journey, and you are an entrepreneur. You've gone zero to one at raw startup, super early scale-ups, and now at Atlassian, which is a big multinational public company. And we're talking about what it means to go zero to one and build something and launch it at a startup as an entrepreneur versus at a big company. Maybe we jump in there. What have you learned about the difference between building and going zero to one?
Nisha Iyer (01:28)
Yeah.
Keith Cowing (01:32)
At startups as an entrepreneur versus integrating within a large organization that just has so many more complexities to deal with.
Nisha Iyer (01:40)
Yeah, it's a great place to start. ⁓ Super excited to be here. ⁓ I really launched my career as a founding member of a startup, and then I was able to...
build my own startup and found my passion there, right? Because I love to move fast. I love to just get into the weeds, and I like wearing multiple hats because I get to do everything and move as fast as I want to. And I think that's kind of like the lack of boundaries I had at startups where it was something that people wanted. They wanted someone that would jump in, wear 10 different hats, and make all the decisions. I have always seen that as something that's a good...
good feature, right? Like that's what people look for and ⁓ I had this mentor during my time period at the startup and/or startups, and she worked at a larger company. She was at ⁓ GitHub, and she challenged me. She said, “Hey, I think you'll learn a lot if you go to a big tech company,” and she's like, “If you don't like it, you can leave, but you need to—like, you can always go back to startup world—but go try it out.”
And I was like, okay, this has piqued my interest. ⁓ And so, Atlassian is a company I've really loved. I've loved their products. I've loved the founder mentality. Our original founder is still at Atlassian and...
just like the generation of SaaS and ⁓ of course like I know a lot of people probably won't love this comment, but I actually loved Jira before I went to Atlassian, and I still do like it as, you know, as an engineer and as a techie myself. But ⁓ yeah, I saw an awesome opportunity there, and it was described to me as high ambiguity, a lot of, you know, being able to wear lots of hats and kind of run things, and so I was like—
Keith Cowing (03:12)
Yeah.
Nisha Iyer (03:32)
Yeah, yeah, yeah, this sounds good. ⁓ And I came into Atlassian thinking that because of the way the position was described, I'd be able to operate in the same zero-to-one state. ⁓ And that was definitely not the case. Yeah.
Keith Cowing (03:46)
Let's dig into that a little bit. I've heard
a great description of entrepreneurs as you're either the corporate end of rebel or you're the rebel end of corporate. But either way, it's natural tension when you jump into an organization. And as a startup, you're paid to be bold, to move quickly, to be decisive. And then—
Nisha Iyer (03:56)
Yeah.
Yeah.
Keith Cowing (04:07)
Big companies, there's a lot of different factors. So walk us through maybe your onboarding, what you learned working with the team and how that impacted how you collaborate, how you operate, how you can bring that boldness. Big companies need this right now because they're stuck with fast change, whether they like it or not. They didn't self-select it, but they need to do it. So you need to bring people into the organization that can drive change. But you also have to meet people where they are while moving things forward.
Nisha Iyer (04:20)
Yeah.
Yeah.
I came in with the idea that, one, I need to show everyone my skill set and how fast I can move and how many hats I can wear and how much I can push change. ⁓ And then, two, I thought that that would be what was going to really show my superpower to everyone. ⁓
This is also the first ⁓ remote-only—like, this is the first fully remote onboarding I had, which is also interesting; not the topic today, but like definitely ⁓ a different experience, right? Because I wasn't able to, like, come in, understand the culture, feel the nonverbal cues, understand people ⁓ more intimately.
Keith Cowing (05:12)
Is it also the first
international company that you've worked for? They're based in Australia. There are maybe not a language-complete difference, but there's going to be cultural differences as well and time zones.
Nisha Iyer (05:23)
I hadn't been in a company in a leadership position where I had to work internationally and, yes, different cultures. ⁓
I have to work in all time zones now. Having to work globally, yes, 100%. And that also adds more to the onboarding because there's half the team that you might not meet unless you stay up till 10 p.m. East Coast time to get on calls with Australia. ⁓ A lot of interesting pieces there. And I came into the customer support org ⁓ as, at the time, Head of Analytic Systems.
Became Head of AI.
And my manager, the first thing he suggested—I think like pre-onboarding or like onboarding checklist—was to read The First 90 Days, ⁓ the book, right? And I was like, okay, cool, this is awesome. And so I, like, picked it up. I was like, I'm gonna get—I’m gonna read this. And I started skimming it and I was like, I don't know. Like, I think I know this. I skimmed through it and I was like, I got it. Like, gotta, like, understand the business, learn, like, understand the people. But, you know, my mindset was still like, what can I drive that's gonna really, like, bring me to the top of...
everyone's minds. ⁓ And I remember, like, within this first or second week I pinged my manager on Slack and said, “Hey, like, I'm ready to go. ⁓ What do you need me to—what do you want me to do? Like, where do you want me to focus? What can I, you know, really dig into?” And he just wrote back, “No, learn the business.” And it felt almost like a slap. Like, I was like, what? Like, I'm trying to—like this—
Keith Cowing (06:57)
You're not ready yet.
Nisha Iyer (06:58)
Yeah,
and I was like, why—like, learn the business? And it had been two weeks. I was, like, itching to just get something done, like, deliver something, get—start working on something, and he just wanted me to... You know, what it ended up being, because he wouldn't really give me any kind of, like, direction or where I should be focusing. ⁓
I just talked to people for the first month. I set up calls with leaders across the support org—it was large, like 1,200 people at the time—and ⁓ understood their problems, what kept them up at night, ⁓ understood more about their... ⁓
business unit or their part of the business. And a lot of these were ops people as well because that was a very important piece to understand of the business, because that's where I would want to make improvement too. So it was a big learning curve, I think, ⁓ to slow down. I was like, don't just jump in. Slow down, understand what you're dealing with.
Keith Cowing (07:58)
And at the end of the day, that was a listening tour. What'd you learn from that listening tour?
Nisha Iyer (08:03)
I don't know if I learned how much I needed to pay attention to what everyone was saying until I, you know, kind of did my own thing and fell flat on my face on how I was trying to change things. But when—once something like that happened, which we can go deeper into, I was able to take my steps back and be like,
I now understand all the different angles people are coming from—maybe, like, some of the things they're concerned about, ⁓ and why what I just proposed fell flat on its face. So I think, like, in hindsight—hindsight's always 20/20—I think, in hindsight, it was very helpful because I had the roots; I just, like, immediately didn't know how to lean into them.
Keith Cowing (08:49)
It reminds me a little bit of—I have a friend who got his son golf lessons, and the first lesson with the pro they didn't even touch a club. It was just pure balance and motion and body and feel and feet. And it's like, “Give me the club. I want to hit the ball.” And it's like, no, hold on. Like, we have to get the foundations right. If that foundation is there, it makes all the difference. It's hard to have patience as an entrepreneur.
Nisha Iyer (08:56)
Yeah.
Yeah.
Keith Cowing (09:11)
Usually entrepreneurs are impatient as a feature. Walk me through that psyche a little bit—working with your team, doing the listening tour while you're probably itching to ship code, to close deals. Thirty days is a long time in startup land.
Nisha Iyer (09:23)
My listening tour also included—I inherited a 30-person team. Meeting with each of the data scientists and engineers on my team, understanding what they did. And that's where I was like, my God, there's so much opportunity. I want to get in. And instead I had to take a step back because ⁓ I was a manager of managers. ⁓ And so the teams had their managers to keep them running smooth. So, like, my strategy coming in hard...
with a new strategy was also not the best look. You know, like, I—it was the first time I came in not on the ground building my own team and being a senior leader at a company too.
I realized that, like, as an entrepreneur the next thing you go to is, like, what are the big problem areas? And so that's where I got my excitement. I was like, okay, let me talk to these people, figure out some of the areas of opportunity, and so then when I can go in, I'll have something strong to go after.
Keith Cowing (10:21)
So now let's fast-forward to shipping product. You came in, you are ready to hit the ground running. You did a listening tour—maybe not by your own will; it was forced upon you—with the really important mindset of really learning the culture, learning the business, and you went through that. Now, you got to build some things, you got to ship it, and even that isn't going to come without obstacles. So walk me through some of those early things that you built and got out the door and the obstacles that you ran into...
Nisha Iyer (10:29)
Yeah.
Yeah.
Keith Cowing (10:51)
...along the way and how you navigated that.
Nisha Iyer (10:53)
So what I found—you know, like, the big areas of what kept these support leaders up at night—was inefficiency, operational inefficiency. How is our very, very large team, that we don't want to micromanage but we also want to understand what they're doing—how are they spending their time?
How are we enabling them to be able to do their job the best they can? And one of the key areas of focus that takes a lot of time from each of these people, or took a lot of time, was ⁓ actually writing knowledge, right? Like, they have to write—create the articles. And one was, I started in 2023, so there were already, like, a lot of ⁓ capabilities that you could potentially use AI for.
And I came to discover that there was a lot of time being spent on writing knowledge and a lot of articles...
I talked to leaders, I talked to people on the ground, I talked to some people on the enablement team. So it was getting, like, a wide variety...
and heard a lot of people complaining about, ⁓ like, knowledge management's a hard thing here. There could be so much more gained from knowledge being written well.
And so once I figured that out, then I started going into the data. I was like, okay, how much—what is the actual objective information here? How much are people spending? And I was—my team was data... oh yeah, big time. Yeah, big time. And that's where I got super excited. Because I was like, oh, I found a big problem. Amount of time being spent here, you know, like, I—
Keith Cowing (12:17)
And did that line up? The data reinforce the anecdotal beliefs?
Nisha Iyer (12:31)
—had this whole spreadsheet, calculated it all out, was like, if I was able to automate this and have, like, a couple people full-time on knowledge, we'd be able to save so much time over the ⁓ calendar year, and so much time of supporting engineers. And so it would help everyone because it would give back time to work on customer tickets. We'd be getting back to the customer faster, which is our number one thing because we do have world-class support—customer support—and that's always top of...
our mind. ⁓ And then ⁓ we would, you know, also be able to improve the speed of knowledge creation using AI, and that was like—that's where, like, my excitement came in because I knew we could build automated knowledge, but I knew we had to kind of follow like a... We wouldn't just go from everyone's creating knowledge to ⁓ we want to just use AI to automatically create it. There had to be, like, a step-function path to it.
Keith Cowing (13:27)
So, seems like a sweet spot for AI. Lots of things happening. It can be repetitive once you create the artifact once. It can be reused in a bunch of situations. If you listen to a conversation with somebody who really knows the product, that's been there for 20 years, it's an expert, and you just record it and think about it, you should be able to extract, hey, here's knowledge. So this should be a home run and easy, right? You build this, you launch it, everybody's on board. So, okay, you got through step one, which was—
Nisha Iyer (13:29)
Yes.
Exactly.
Yeah.
Yeah, should be. Yeah, exactly.
Keith Cowing (13:54)
—identifying the pain point, validating it, deciding it's worth fixing because AI is a good fit for it and it drives huge value and is good for customers. Okay, great. Now you got to actually make it happen. What was that like?
Nisha Iyer (14:06)
So I was super excited because I'd done all the things you just said, and I had the data, and I had the talking points, and I knew I was gonna be able to save tens of millions of dollars a year with what I was proposing ⁓ through, you know, work—
actual work hours logged, saved for support engineers, and other ways. And I'd done all these calculations. And so what was my logical next step? Go talk to the Head of Customer Support. And so I set up time with him and no one really said anything to me. They were like, “Yeah, yeah, okay, sure. Go set up time with him,” and it ended up being this huge meeting with a bunch of leaders on it ⁓ but I was just pitching the idea.
And I went in very confident with my data and talking points and went straight into it, said, like, this is what we have, this is the problem—got stopped, you know...
five minutes into it or less; got stopped by him, and he had a bunch of questions. And they were around culture. They were around, like, how's this gonna affect, you know, X, Y, and Z? And how's this gonna make these people feel? And how are you thinking about that? And I hadn't thought about it. My solution wasn't around how am I thinking about the change management and the transformation of a team that currently exists. My solution was how am I taking a painful problem and...
and using technology to improve it in an effective way without thinking about all the different other things that will be affected—AKA culture, change management ⁓ operations.
And so I'm on this call with a bunch of people, realize I haven't really thought through some of these things, start just kind of shooting from the hip on things in this call, and it didn't end up well. It ended up going over, you know, had to be stopped. There was no conclusion reached. My—what I was proposing wasn't really accepted or rejected, but I didn't feel great after the call. And I confirmed, you know, I talked to my manager. He was like,
Like...
“Yep, waste of time. Wasn't a good call.” And that was probably—yeah, I think everyone did, because I talked to a few other people and they're like, “Well, everyone has those calls,” you know? Like, it's a learning experience. I was like... like, could have been warned what was gonna happen. ⁓ And that's when I think a lot started—I feel like it was like a white light, ⁓ kind of like a moment of clarity, right? Like, I was like, my God, now I get it.
Keith Cowing (16:11)
And he probably knew that ahead of time, but he was letting you learn.
Nisha Iyer (16:37)
I need to listen to what everyone's saying before just understanding a problem and, like, you know, bull in a china shop going to attack it. There's a lot more to this. It's a lot more complex. Now I get what, like, my boss kept saying in the beginning. He was like, “This is a very—it's a very complex org.”
Keith Cowing (16:39)
You—
Nisha Iyer (16:56)
And I was like, okay, like, I can—I figured out the problem. You know, like, it was just like, I ⁓ startup mentality. It's like, I can find the problem, figure it out, and everyone's gonna love me. And that's not how it worked. And so I had to take a step back. Yeah, yeah.
Keith Cowing (17:05)
You just went for the gold all at once. You're going to seize it. You're going to get...
it through one bold, swift move and then kind of fell on your face, or fell on your back—one way or the other. And there's a saying that I love that everything is learnable, but not everything is teachable. And it feels like an example of that—you sort of have to go through this sometimes and then the feedback afterwards, the coaching to be able to say, like, okay, now what? So let's get to the now what?
Nisha Iyer (17:12)
Yeah. Yeah, yeah, I did. Yeah.
Yeah.
Yeah.
Yeah, no, I totally agree. Yeah, so the “okay, now what” was, like—I feel like I came to a screeching halt. Like, I literally thought I was gonna come out of that meeting and everyone was gonna be like, “Wow, Nisha, she's so smart. Like, she's proven herself in three months and she got this—like, it's done,” you know? And then I was gonna be on to the next big thing. I think what actually came out of it was like, okay—like, for me—it's like, I need to slow down. I need to think about the people. I need to think about the process. I need to think about how—what do I actually—what would I actually do at a startup?
Roll out a POC and then I scale. And what was I doing here? I was scaling and trying to approach an entire org when I should have been thinking about what is a POC and how can I gain trust? You know, like, I'm new here. Like, how am I building and gaining trust of the people here? Not just by talking to them. I have to actually prove something works and prove my ideas are, you know, alongside their best interest.
And so I ended up talking to one of the four leaders under the person—the guy, the person I pitched to was the Head of Support. And they had four leaders that, like, led each of the business units under him. And I talked to one of them and said, “Hey, are you willing to pilot this?” And I had a much more curated presentation that focused on his pain points, not as much the company's pain points—
what would speak loudly to him, right? Like, how are we gonna make this recommendation system that we built—my team had built—better for his engineers to be able to respond to the tickets faster. And all of that still is based off of improving the speed and the quality of the knowledge we're creating. But the way it was presented was in a different way—it was for how's it gonna benefit him...
and his team, and he said yes, and we piloted it, and it ended up being really successful—and we can go into what happened after ⁓ but yeah, that was—that was a huge lesson for me. And I just, like, always remember one thing my ⁓ then-manager said to me, and I take it into everything today, is “Every moment is an opportunity.”
And when he first said it, I was like, what? But I realized it. Yeah, because I was like, I went into this meeting—bull in a china shop—thinking my little weird presentation that wasn't polished, was just, like, these numbers, was going to convince him. And I didn't take that moment as the opportunity it was. I could have approached it in a very different way.
Keith Cowing (19:48)
Thank—
Nisha Iyer (20:07)
—it would have probably turned out maybe closer to what the outcome was that I was hoping for, but it took a little detour, which was helpful because I learned a lot.
Keith Cowing (20:17)
And I sense a couple of things happening here. One is you said “trust,” which is a huge aspect of—the way that you earn trust is different in different environments. And talking about the second person you went to for the proof of concept—what he cares about and what matters to him. At the end of the day, as a leader, you're changing hearts and minds, and a lot of empathy has to be involved here. It seems like at the beginning, it was a little bit more inside-out: I'm going to analyze this and—
Nisha Iyer (20:44)
Yeah.
Keith Cowing (20:46)
—make it happen and bring it onto the world. And the second attempt seemed much more outside-in of: let me take all of my listening and then tune something so that we can get it started and eventually achieve the same vision and objective, but a different navigation path to get there.
Nisha Iyer (20:51)
Okay.
Yeah.
Yeah, 100%. And I think the big learning piece was that I—like your point about empathy. ⁓
I was unable to put myself in the shoes of the ops side of the house because I was the most unfamiliar with that, right? And I didn't—even though I'd heard them out and heard what was important to them and how they worked, and potentially also was able to understand some of the fears they might've had around AI, around some of the things I was introducing—
I still thought the most important thing was the bottom dollar—like, what was I changing? And, like, some might say, like, yes, it's important, but that—the way to go around doing that isn't necessarily just bulldozing and telling everyone “This is, like, what's right.” And I think, like, that's where that listening tour came in. That's where taking a step back came in. And just—that's where the difference came in, especially with startup mentality versus ⁓
being at a large corporation where there are people that have been there for 10, 15 years that have understood the culture, know the culture, and you have to kind of move your way in, not just bulldoze your way through.
Keith Cowing (22:21)
So you went through all of this, you got the proof of concept, you got some momentum. Now you have momentum that you can build on. Where did you go from there? How did this momentum turn into what came next? What kind of impact AI was able to have as a whole?
Nisha Iyer (22:33)
It was the start of just people embracing the fact that, you know, using AI was actually improving efficiency across the org—both for ⁓ from knowledge creation to bringing that knowledge to support engineers and to improving their overall, you know, like, the overall time for resolution of a ticket to a customer. And so therefore also focusing on how are we improving CSAT. ⁓ And people wanted more of it, all of a...
So, Support is... you know, we hadn't at this point—we did not have any customer-facing AI. So ⁓ if, as a customer, you came into an Atlassian...
product and had a support question, you were directly routed to an agent. And at the same time—and I think there was, like, still some concern around what would it be like if we actually made this customer-facing? Is it good enough? Is it gonna sound like a bot? You know? And of course, like, in the background, AI is just exponentially improving every month. The models are improving, the tones are improving, the outputs improving. And I know that, you know, that—other people knew that, but not a lot of ⁓
the team wasn't sure if they trusted that. This is, like, around a year ago this time, I found out that the...
product team at Atlassian was actually incubating a customer service management product, which was super exciting. So I started working directly with them and they—which now I'm on that team. We had an AI agent that would interact with customers. So it became this thing that kind of came over to—we're releasing a product that's going to do that. We need to try it out internally. And so that's kind of like the merge of how we really got...
to how I really got to be involved in a customer-zero dogfooding initiative with a product that is soon to be going out.
Keith Cowing (24:27)
If you look under the hood for a second between the tooling and the models—on the tooling side, did you build this in-house or did you use vendors or did you use a combination?
Nisha Iyer (24:35)
I built it in-house. Very build heavy.
Keith Cowing (24:35)
Then on the models and then on the model side, did you use open source? Did you use closed-source tools? Did you use a combination? Okay. And you have different models for different scenarios—are you using Claude in some situations and ChatGPT or Llama in other ones?
Nisha Iyer (24:35)
A combination. Yeah.
Yeah, it depends. I mean, there's, like, some orchestration going on. It's changed over time. You—we're always running experiments to see what makes more sense. ⁓ But in the different models piece, I think, like, ⁓ like, you know, without the LLM—LLM is one thing, but then also ⁓ ML models. Like, how are we ⁓ making the decisions to call an LLM? Like, what are the things that happen before that? And the interesting thing about Support ⁓ is also that it's very ⁓—it doesn't need to be orchestrated...
by an LLM, right? There's, like, a lot of, like, very routine decisions that get made. So that's also something that we ⁓ were building into the—like, understanding the pipeline of what happens internally, and most likely externally too. So yeah, I think that ⁓ different types of models—but some in-house, some external.
Keith Cowing (25:49)
And so across all of this shift, what was the biggest thing that you learned about yourself, about taking an entrepreneurial mindset and making things happen at big companies?
Nisha Iyer (26:00)
I learned: keep the mindset, learn the company, take some more time, and treat every moment like an opportunity.
Keith Cowing (26:10)
And take the time. This is a dichotomy. All leadership issues are dichotomies. So you've got urgency and time pressure on one side, and then you have—yet you need to learn the fundamentals. How do you just tactically balance that?
Nisha Iyer (26:24)
I mean, I'm still learning. But I think, you know, I think it's just really understanding the people. Like, empathy is a huge part of this. It's a huge part of leadership. And I think when I came into this, I was—you know, I thought of myself as an empathetic leader because of the way I interacted with my team. You—over the years I've come to understand how to build a strong team, how to ensure the team's connected, how to be a good leader with my team.
I've learned at Atlassian how to empathize with other stakeholders, with leaders, with ⁓ the company's values versus the department's values, and how to ensure that I'm meeting both of those when I'm moving forward with ideas.
Keith Cowing (27:09)
Are there specific habits or rituals that you've built into your schedule to hold you honest?
Nisha Iyer (27:17)
I think just making sure I'm...
paying attention to—there's so many different threads going on, you know, at a large company. And as you know, you know, we live in Confluence. And so there's a newsfeed ⁓ in Confluence where you can kind of see, like, the biggest updates—what's going on in the company. ⁓ If you're not paying attention to that, you can miss out on a lot. And also, like, another thing I think that has helped me be more effective is connecting the dots for other leaders within the company.
Of, like, “You know this was going on, you know, in this department,” or “Do you know we could lean in here?”
And really embracing the team mentality of it, too. You know, our ticker symbol is TEAM. So it's a big thing here where, you know, we're really connecting with each other, and that can be lost a lot of times. And so I think the things that I've learned at, like, a large company is that it is super complex. And that I have to keep my ears open to what's going on and how I can collaborate versus just reinventing the wheel.
Keith Cowing (28:25)
And it's hard for big companies to move fast. It's not easy. It's also hard for people that are aggressive entrepreneurs and used to moving really quickly to sort of adjust. I think we're in a mode where you used to self-select rapid change by being at a startup or a turnaround, and you used to self-select stability by being in a big company. And now stability is gone. It's a fallacy. Everybody's stuck with rapid change, whether you asked for it or not. And...
What you're describing, I imagine everybody's going through right now—where the big companies have to make rapid change, but they're not necessarily as wired for that. And people coming in are not necessarily wired for that environment. And we have to make it work together. We have to tie these dots. So for other folks that are running AI initiatives, making rapid changes, navigating uncertainty, and blending the entrepreneurial boldness with the respect and trust that comes with building relationships in a big organization...
What advice do you have?
Nisha Iyer (29:23)
I think just understanding—for me, I think it's understanding that this rapid change causes a lot of tension and anxiety, right? And what am I doing to ensure that I am helping people understand how to work with the change versus fear of the change? I know that's, like, super broad and high level, but, like, in Support, for example, it's—how do I help people embrace...
⁓ the new tooling and the new things that are coming in and help them understand how it's going to help them versus be fearful of it and how it might hurt them? And I think when people are able to embrace change...
They're able to move along a lot more smooth. I don't know if it's going to be necessarily that it's gonna be, like, a home run and everyone's gonna be on board and it's just gonna go, you know? I'm—that, then it wouldn't be—it wouldn't be a difficult situation, a problem. Like, you know, I think that's something that as leaders you have to navigate, especially at a larger company, but I think it's just: how do we get people to be excited about the change? And I think with, like, CSM, for example—like, when we did release this AI agent ⁓
we ended up having a lot of excitement around what it could do ⁓ because it was just, like, something new, something that was—ended up, I think you asked about, you know, like, what are the metrics that we use to define success? One is a very industry-standard metric, which is resolution rate. So how many of the tickets are we actually resolving with our AI responses? And that was a lot higher than people expected. And I think feeling that ⁓—just like, you know, a few metrics—help...
lessen anxiety, right? Like, objective information that's helping people understand that this is actually doing something that's helping them ⁓ and how is it helping them.
Keith Cowing (31:16)
You mentioned two big words there: fear and excitement. And these emotions are real, and we are in a moment of peak uncertainty and ambiguity in many ways, and there's a lot of things to be excited about at the same time. So what's one tactical tip for people to help their team have less fear, and one tactical tip to help the team have more excitement?
Nisha Iyer (31:45)
I think, you know, that “less fear” depends—I think it depends so much on the team and the people and the appetite for change. But the tact—I think—
Keith Cowing (31:57)
But it's one thing you did. You get on...
Zoom with people in all these different countries. It's hard for something to come through that's so primal—to get rid of fear. Fear goes deep. It's not a surface-level thing.
Nisha Iyer (32:08)
Yeah.
Yeah. Agree. I think it's showing, you know, like, actually demoing ⁓ what I do...
to ⁓ help myself move fast with AI. Because AI is the elephant in the room. That's where, for me, I've seen a lot of fear. “What am I going to do with my job now that AI is here and it can write a PRD in two minutes?” Or “It can spin up a mock-up in two minutes, and I'm a designer?” I think what I've been able to do is just ⁓
hopefully try to empower the team and be like, “Hey, like, now we can do all of this so much faster and get to a point where we can deliver.” It doesn't mean that—it doesn't mean that, like, anyone is at risk. It's just that we can all move together in a faster cadence. And what I've done—like, yeah, yeah.
Keith Cowing (33:00)
You're modeling that behavior of—
I'll show you how I'm using it, and I am excited about it. And I can't force you to be excited, but I can show you what genuine excitement looks like. And I hope it spreads to the team.
Nisha Iyer (33:06)
Exactly.
Exactly. And, like, a very tactical example of that is that we have a product called Rovo ⁓ which is awesome. ⁓ Shout out to Atlassian. But ⁓ it—it basically is, like, across all the different Atlassian products that you have. So, like, you can go into Rovo Chat on Confluence and be like, “Hey, create me, like, X, Y, and Z Jira tickets,” and it’ll create those tickets in your Jira instance. ⁓ So, anyways, ⁓ really cool. Also a lot of apprehension in using it—just, like—“Does this work?
Is it helpful?” And so, like, something I did early on was—I did a Loom. I, like, recorded myself creating—using Rovo—actually creating a Knowledge Article Creator ⁓ and then deploying it, and it took all of, like, five minutes. And I did it live in the call just to show, you know, one, the ease of what I was creating and that other people could also create these things, and two, ⁓ like, don't be scared because you can also do it—
and this is how easy it is to do it and use these things to empower yourself. And I've actually seen that go ⁓—there's a Slack channel we have called #ai-pm-design-hacks and it's just everyone sending their hacks. ⁓ But I love it because it's—yeah, people send Looms, people send Conf—yeah, but a lot of Looms. “Hey, I did this; this is how quickly I did it; look what I was able to ship.” ⁓ It's encouraging.
Keith Cowing (34:30)
And they send that over video—over Looms. I love that.
Nisha Iyer (34:42)
And it's like, you're not alone, right? You're, like, helping your team also be able to build at that same speed. So. ⁓ Yeah.
Keith Cowing (34:48)
I love the use of video, especially right now...
for sharing things where—I even did it with async office hours. If people have a question, they can just email it to me and I'll record a video and send it back because it's just so much more human. It's actually easier for me than sitting down and thinking through an email. But you get the tone and you get the expression and you get the details, and it's much more trust-building. I think it's much more human at the end of the day. It's a nice way to use technology to be more human, not less.
Nisha Iyer (35:01)
Yeah.
Yeah.
Keith Cowing (35:17)
That's why this podcast is on video—so people can see your face and hear from you as a person, not just, “Here's the three things that you do to crush Theater One at a big company,” right? It's like there's nuance to it. Nuance matters. And I think video is much better at nuance than email.
Nisha Iyer (35:25)
Exactly.
Yeah, 100%. As—my undergrad was actually communications, I remember one of my professors always said, “94% of all communication is nonverbal.” So how are you getting that if you're not on video? Video is another step forward, right? I guess you're not in person, but at least you get to see expression, hear tone.
And yeah, I love being able—I record most of my updates in Loom to my team, and especially async. It just makes it way easier.
Keith Cowing (36:00)
Final question for you.
What are you excited about right now in terms of AI and technology—not specific to Atlassian, but just generally speaking—things that make us more human and improve connection?
Nisha Iyer (36:11)
I think the voice technology that's coming out is awesome. Because it makes you feel even that much more human, right? The ability to talk to AI and feel that you're talking to a person—incredible. And opens a door to so many... ⁓
areas, ⁓ industries that need that person-to-person interaction. Education is one of them. I think it opens ⁓ the door to a lot of underserved communities ⁓ and being able to scale, you know—a place where, like, maybe you would need a teacher-to-student, where you can actually have AI-to-student and be able to ⁓ get some of the things—some of the tutoring, some of the skill sets—to areas and populations that might not necessarily have got...
them before we had this amazing technology that just keeps improving every day, which is crazy. I never thought I would be... I didn't think in 10 years—like, what was 10 years ago? You know, like—LLMs—GPT-2? I don't even know if GPT-2 was out in 2015, but it was just, like—like, being able to complete a sentence, like, with generative AI was amazing. So where we are today is just—it's insane.
Keith Cowing (37:24)
I love it. I'm also very excited about voice. I use it all the time. I'll put context into ChatGPT and then I'll go for a three-mile walk with my dog and just talk to it and say, “Hey, I'm preparing for this—what should I think about? Let's get it, you know, agenda-ready,” et cetera, et cetera. It's like having a chief of staff, and then you get back and it's all right there and you just copy and paste it. And being able to use tech to unleash from the desk is one of my favorite uses of it. I don't love sitting at a computer. I love thinking through things, and voice allows you to do that without being at your desk.
Nisha Iyer (37:27)
Yeah.
Yeah.
Keith Cowing (37:53)
All the time.
Nisha Iyer (37:55)
Yeah, it's like...
something—exactly—Chief of Staff for my brain, which is amazing. Yeah.
Keith Cowing (37:58)
Yeah. A scary thought, but also very amazing. Nisha, thank you...
so much for joining me. This is going to be super important, both for entrepreneurs and big companies, to think about how do you go to zero to one—not just at a startup, but at a big organization? And how do you get out of your own way in a certain aspect of thinking about building trust and being empathy-first and not inside out, but outside in—to really manage that dichotomy of being bold and urgent, moving fast, but also...
respecting the environment that you're in and building the relationships and the trust that will allow you to be successful. Atlassian is obviously lucky to have you, and I was lucky to have you on the show. So thank you.
Nisha Iyer (38:37)
Thank you, Keith. It's been awesome. Thank you so much.
Keith Cowing (38:40)
I hope you enjoyed that. If you did, please share it with a friend who's trying to drive zero-to-one entrepreneurial impact and subscribe to the show. We'll bring you playbooks and raw conversations with founders and builders every week. Until next time, enjoy the ride.
