The Stack Overflow Podcast

For Twilio's CIO, every internal developer is a customer

Episode Summary

We chat with Michelle Grover, the Chief Information Officer at Twilio, about making hard decisions across a large engineering organization.

Episode Notes

You can find Michelle on Twitter here.

You can learn more about building apps with Twilio here.

Our lifeboat badge of the week goes to TryingToLearn for explaining the error that pops up in Python when: you can't assign to literal.

Episode Transcription

Sara Chipps What do you have to think about when making big decisions for an internal organization when you have a public company?

Michelle Grover That you're not gonna get it right. [Michelle laughs] When everyone is your customer, it's not going to be perfect, right.


Ben Popper Hello, everybody, welcome to the Stack Overflow Podcast! I'm Ben Popper.

SC Hellooo! 

BP Hi Sara!

SC Hey, Ben, how's it going?

BP It's going well, how are you?

SC Good, good. I just finished moving my car. Now that I'm back in New York City, I have to move my car to the other side of the street every other day.

BP These are city people problems. I used to have these problems. Now I've got runaway chickens. So you know.

SC Now you have to move the chickens to the other side of the yard.

BP Yeah, alternate side chicken day. 

SC That's great. 

BP So Sara, we have a great guest today. She is the CIO at Twilio. CIO—what is a CIO versus a CTO? I think we'll have to dig into that. 

SC That's what we'll have to learn. 

BP You know, it's one of those things where like, are those two things interchangeable? Or do they have like a history behind them? It's not something I'm familiar with. Have you dug into this at all? 

SC I've definitely worked with both, but I would love to hear it from Michelle. So today we have Michelle Grover, who is the CIO of Twilio, which is a very big company. Welcome, Michelle.

MG Thank you, folks. Thanks for the welcome. By the way, I've had the moving from one side to the other problem. The Bay Area also has that. [Ben & Michelle laugh]

SC Yeah, you guys have alternate side of the street parking? Man, it's nuts.

MG We have street cleaning, which is and they will essentially just make you move it and then they sit and wait for you, by the way. They lie and wait. You can see them halfway down the block waiting for eight o'clock to hit.

SC To give you a ticket? 

MG Yep. Yep. He's usually having his coffee and breakfast. I wave at him.

BP I've played that waiting game as well. Sometimes I would be in the car, you know, on a call sending some Slacks or whatever, or just eyeing each other.

MG Yeah, you're looking at the time, you're like, which of us going to move first? I still have a minute and a half.

BP But Michelle, welcome to the show. Tell us a little bit about what is a Chief Information Officer. And if it's different, how's that different from a Chief Technology Officer?

MG Yeah, so it depends on different companies. But usually the Chief Information Officer has a lot of your traditional teams like tech services. So the folks that set up your endpoints or your computers and help desk and set up JIRA, and all the various tools and some of the RPA tools, things like that, that's usually under the purview of the CIO. Chief Technology Officer usually is more about kind of craftsmanship and software craftsmanship, and artisanship is what I would say. So a lot of times, they have the architects on their teams, they help the developers, you know, craft code that's extensible and scalable. So that would be the difference. I'm more about internal tools. My customers are all the internal customers that work at Twilio, because they need computers. They need to use JIRA, they need licenses. And then a little twist on that for what's a little bit different at Twilio, as opposed to traditional CIO, which is I also have, like the cloud services organization. So I have, you know, AWS, GCP, various cloud services, tools, and things fall on to me, as well as the tech services organization.

BP Ah, okay, that makes sense. So you're sort of managing, you're overseeing the internal IT, making sure everybody has what they need, the hardware and the software endpoints. And then also, right, the cloud services on which so much of what modern software stack does kind of runs.

MG Totally. And then even like, if you think about it, like you have your internal developers, right, and so the internal developers need CI/CD, all of those tools, those internal tools they need in order to do their job. So I have a team that actually supports that as well as deployment and that SRE, and all those kind of traditional DevOps types, roles. So when I say internal customers, I mean, our internal developers, everybody that comes in that needs a computer, pretty much everybody for that.

SC Do you know how large is the engineering team at Twilio?

MG You know, honestly, I don't remember how large—I know how large our organization is, is about 5000 of us now between acquisitions and stuff.

SC I bet at least half of those are engineering or maybe a third of them.

MG Well, it's between Go-to-Market and Sales actually has a pretty decent sized org. And development and r&d, or the, you know, teams that support the developers as well, right? Because you have product managers, program managers, UX, all of those folks were falling under the r&d organization as well. So they probably are pretty close in size.

SC That makes sense. When you think about need or assessing need, and what needs to change for such a big group of developers. How do you identify, you know, what's breaking and what needs your focus?

MG You kind of run around like crazy for a bit. [Michelle laughs] But the way I kind of look at it is my job is to do a certain level of firefighting. But if we're still doing this a few years out, then I haven't done what I want. I want always look for opportunities to automate and make things more sustainable. And even just put visibility in, right, a lot of times things break, and you need to go back and retrospect on it and figure out what broke. But my goal is to make sure that when things break or something goes awry, we're actually tracking it, and figuring out a way to make sure that we find out sooner that it happens before, right. Like, if that's kind of the way I look at it, you first go in and you kind of just run around, putting out as many fires as you have to the biggest emergency is the thing that you focus on. But the balance is also trying to make sure that you focus on the long term ability to make sure that you aren't running around still, that you've actually put some planning in place. Do I need more people? Do I need more systems? Do I need to increase how we do certain things? Do I work with the CTO to talk a bit about how our developers do certain things and how we support that in order to get some of that firefighting out if that makes sense.

BP So Michelle, can you tell us a little about sort of what led you to this role? Have you always been on that side of the house, like you said, sort of internal facing helping people get the work done, as opposed to I think you, you know, made the comparison to the more craftsman like, approach of writing code on this, the tech side?

MG Nope, always been a developer, [Michelle laughs] pretty much not been part of tech services. Last organization, I was SVP there had about 300 folks worldwide. I had a mobile team, I had a full stack teams, back end, front end, DevOps, QE, SRE, various teams there. And I've always been on that side of the house where I've actually done, you know, coding myself, and then kind of grown into leadership roles from that. I've done a lot of deployment related stuff, cloud related stuff, infrastructure and data centers long ago, because I, I have a good 30 years under my belt. So I've done a lot of that. And then what brought me here to this role is actually kind of interesting is that they were looking for a CTO, and at my role that I was at previous, it was a great job, like I'd been there seven years, love the folks I was constantly kind of growing in it. But we finally were, we've been purchased by SAP. And it was finally time for us to kind of become part of their, become a business unit as opposed to a separate entity. So in order to go to kind of next steps, I was already SVP, in order you would go to like a VP or like C-suite in SAP proper. So it was more about like, okay, where do I want to go? What do I want to do? And so I kind of looked around a little bit, and then I sit on a not-for-profit on a board called Techtonica. And their CEO actually works at Twilio. And then I have a really good friend that's also a mentor that used to be our CEO at Concur. And she sits on the board for Twilio. So I knew what Twilio was, because we use it in house at you know, Concur and in TripIt. But you know, like, I didn't know what the culture was like or any of that. So I talked to them and saw, you know, asked what they thought about it, like, is it a great place? Is it crazy? Is it Michelle level crazy? [Ben laughs] You know, you have to be honest about these things where you are, who you are. And they really, everybody's like, really like pumped about it. And my friend was the board member she's, she's, of course, the most accurate. She's like, they're at the teenager stage. So they still got to get over the hump a bit. But you've seen a lot, so you could bring a lot to them. So I went from the CTO role talked a bit to my now boss about that role and a few other roles. There were some senior, some senior vice president roles for a few different business units. And then ultimately, they you know, after kind of heard about my background, because I also had cloud and deployment in my purview. And I did I've had Help Desk as well. They said, Yeah, would you think about being our CIO? And I was like, really? No. I talked a little bit to them about it. And I talked to like our old CTO and CIO at Concur. And they had a similar setup where the cloud services organization, a lot of those, as well as do traditional tech services rolled into one individual, so it's not unheard of. And I'm like, you know, this is fine. It's the marriage of kind of both it's all my customers, but it's a different customer facing role. Right. My other my other roles were all about the customers being very unhappy with us. Well, we did something wrong and being very interesting calls. And now it's internal. And they can get to me sooner. Right? They know how to get to me. So that that's what kind of got me here is a weird kind of windy road. But you know, so far, so far entertaining.

SC That's great. And it sounds like it sounds like as part of your role. You've helped to identify system wide problems in a way that you are proactively solving them for the future. I imagine—so something I know working with lots of software engineers is that they love problem solving. And I bet you get no shortage of ideas for how to solve problems, but how do you identify, okay, this is a solution that I think will work, this is something that will really help the team versus okay that might be might be you know, grinding our gears a little bit in that direction.

MG Some of it is going to be trial and error. Some of it is putting certain things in place. That's where being an engineer and experimenting really helps. So sometimes it's a matter of do I just have to plug this hole for now? And what is for now? For now might be three months to six months, but it's not a year. And so can I solve it by building something? Can we build tools? Can we buy tools? Do we already have tools? A large part of that is my job and especially is unique because Twilio did not have a CIO prior to me. So coming in and kind of defining what that role is, kind of making it my own, because it's a little bit different because it didn't exist. And then a little bit of the difficulty is everyone or not everyone, a lot of people have had a CIO at their previous organization, their mental model of what that looks like, versus what it's probably going to look like for us, it's going to be a little different, right. So some parts of just figuring which problems I can jump in on, I can't take all of them right now, or I wouldn't be successful and kind of figuring out which ones make the most sense for now, which ones need certain tools, which ones will have to wait till 2022, maybe, and we just have stopped gaps. That's a lot of my job. And the best thing I can do to kind of set myself up for success right now is just getting the right people in place. And kind of doing some org changes to to get us really kind of prepared for the future, honestly.

BP So Michelle, I had a question, which is that Twilio is kind of in an interesting place in the world of technology, that it interacts with some some, you know, some very old systems, sort of the telephone networks, you know, the world of SMS. Was that something that you had to learn about when you came on? Was that something you had experience in? And? And what's it like, yeah, I was sort of working at a company where you're at, you know, the cutting edge of what mobile apps are able to do, but also relying in some ways, you know, on a much older network?

MG Yeah, so I many, many, many, many moons ago, I worked for a paging company called PacTel, which was part of Pacific Telesis, which eventually became Air Touch and eventually became varizen. Very far down the line.

BP Oh yeah, I've heard of that company. [Michelle laughs]

MG But have you heard of the other ones? That's a trail that was made for them. So I've worked in Telecom, you know, and that type of mobile, prior to mobile paging, you know, many years ago. So I do have that experience. And a lot of funny enough, a lot of those systems still exist, because that's how telecom is, right, like a lot of just the core systems still exists, they're sometimes a little bit shaky, but they still work. And so coming in to Twilio is really fun, because you get a lot of the new stuff, right, like, you know, Jeff's thing is always, you know, ask your developer, like the fact that we enable developers to do what they need to do without having to write huge amounts of code and actually be able to sit on top of our network is pretty amazing. But down in the guts of it, you still have a lot of that same infrastructure, and a lot of the, you know, FCC regulations, and all those things still exist. So it's kind of playing both worlds, right? You have the cool cutting edge, and they got the hang on, we're not ready for the cutting edge yet. So that you have to, like wait folks, hold your horses, how are you supposed to do this? So it is kind of narrowing both of those too.

SC Twilio is a publicly traded company, I thought that adds a lot of pressure to you, as a lot of the things that you do need to be very transparent. As a lot of your customers are internal customers, like you said, what do you have to think about when making big decisions for an internal organization, when you have a public company?

MG That you're not going to get it right. [Michelle laughs] When everyone is your customer, it's not going to be perfect, right? Because I'm not gonna get anything—if I pick a tool, it's not going to be the tool that everybody wants, there's gonna be tools that people are very comfortable with their systems people have had experience with. And so I try to make sure even when I talk to my team, it's a lot less about making sure everybody is happy, and making sure that we kind of make decisions that everyone can use. If that makes more sense. You know, it's, I can't go for 99% satisfaction, my NPS score probably is not going to be that when I run it, you know, later in the year. There's going to be things we get wrong. But what I try and look for is what did we get wrong? Is it that it's someone really likes this other tool or this type, you know, you have the person wants Mac Book, the other people want chrome. Some want old HP laptops, or different, you know, there's, there's always someone's personal preference, and how do we accommodate that personal preference, but also think about the things that we have to do for the company and kind of the greater good of everyone involved. So it's never gonna be perfect. No one is ever 100% happy with me, but I tried to still listen to the customer. They are my customer. I try to make sure that even if it's not exactly what they want, I'm solving the problems and saving them time so they can focus on their real job, which is not worrying about this computer or if the computer goes down, that's that's really impacting their ability to get their work done.

BP So Michelle, let me ask you, you know, you mentioned you worked in Telecom, you know, you've worked overseeing mobile app development, and now a CIO role. Do you personally have like a favorite area that you like to work on? Or that you do when you're not at work? Whether that be mobile development, or web, back end, maybe do some hardware engineering like Sara in her spare time? Is there something that sort of kind of your favorite area when it comes to computer science or engineering?

MG I still like back end. It's funny, because I can actually draw, but I am not a front end person. It drives me bonkers. Just the idea of JavaScript makes me sad, a lot. It's just not. It's just not my thing. [Michelle laughs] It's just not my thing, though. I appreciate it. 

SC Alright, Michelle, we'll schedule some time later and I'll evangelize JavaScript to you.

MG No, no, everybody's trying, it's not worked. [Michelle laughs] But so I do like back in the most fun I have though, is, so I have a 15 year old son. And I spent time with him. And then you know, he's not in high school, but we used to when he was in middle school, and we weren't doing COVID stuff, we would go I go to the school and volunteer with them. And we play with Raspberry Pi's and set up systems and send just simple systems. And what I think is the thing that as a developer that we lose a lot is the fun part, because we're so you know, your day in and day out is more meetings than it is really coding, you know, I don't code at all now. And so sometimes I just think that and I don't know, if you folks are saying, you just need that reminder of why this was fun in the first place. And just kind of sitting and tinkering with something that doesn't have any repercussions. Right? It's just kind of fun. If I mess it up, I mess it up. But it's fun, and my son and I can laugh at each other and make jokes and do stuff. And it's kind of a nice cathartic way of saying, okay, I'm not going to lose anything. And so at some point I will do when I actually have some free time, I'm going to create an app on our platform for Twilio. But I haven't had the chance yet.

SC Twilio and hardware is very fun.

MG I hear that. I just haven't messed with it yet. Because you know, I keep getting calls for all kinds of other things in my day job.

SC I've worked with a lot of kids and teenagers coding. And I found that when it comes to the building things in a physical space, like hardware, it becomes so much more real to them. 

MG Yeah, for sure. 

SC Because they can see it. Yeah, yeah, very cool that that must be a cool bonding experience.

MG It is just making kind of goofy things, we use a 3d printer. And we printed up this like giant mouse once, just because and put wheels on it and use some popsicle sticks and stuck it to like a Raspberry Pi board. And I don't know what we were doing. It kind of worked for like a whole day. And then it stopped working. But it was so fun.

BP It's all fun and games until he gives you your NPM score and then he might be grounded.

MG And let me tell you, my friend is not thrilled with coding. He knows how he's been coding since he was younger. He knows Python because he has Minecraft. So I you know, we would do that as an exercise and you know would programming mods, but at 15, he just cannot be bothered, like the best I can do is to get him to sit down and kind of play with, you know, the Raspberry Pi's, but getting him to sit and physically code, he looks at me like, "oh, no."

BP Well, maybe the Python will come back around someday.

MG It will, I believe in it.

SC Michelle, as you think of I'm sure there's a lot of people listening to the podcast that you know, have career goals. And being CIO for a company like Twilio would be something amazing that they'd love to do. For those folks that are listening and kind of thinking about this as their path. What would you advise them? What would you tell them to focus on?

MG You know, the funny thing is, and it's funny, because I just did a kind of an internal panel for our black Twilio group and did a fireside chat. And we kind of talked about that. And everyone always says like, Oh, you know, it's such a rare thing to have, you know, women, little on minorities at some point in C-suite. But I never thought about this, though. You know, I was that a little kid that that would aspirations and being in the C-suite. That's not how it works. So I take that to heart and I tell people that the windy road to what you want to do is filled with things you had no idea you wanted to do. And so instead of looking at the end, try things. Like I wouldn't have been I went to school for architecture, I didn't plan to be an engineer. I can code and I could code and I did Basic and Fortran and Assembler and all that in high school. But I didn't want to code it was not ideal and nor was it something I particularly cared about, though I was good at it. And so the fact that I but I would try something new and it would be interesting, like oh, well that's kind of interesting, and it would lead to something else and it would lead to something else and you know, you kind of just got here. And so taking chances is really a part of it, trying something and seeing if you like it, looking for people who do it really well, to advise you that sometimes your peers and sometimes they're folks coming in level below you that know something. So always looking for opportunities to do something new or somebody that does something well and learning from that is the best growth potential that you can get. And honestly, not being so caught up on the title. The title was interesting. But with this title comes a huge amount of work and responsibility. And if I did not enjoy what I did, the title wouldn't mean much.

SC Yeah, that makes that makes a lot of sense. Yep. 

BP And now you can have faith that someday your son will think coding is interesting, because you just said, as a teenager entering college, it didn't interest you anymore. So yeah.

MG See! It might like, yeah, like, this is why I'm like, just I keep telling them like, I don't like whatever you think works for you, kiddo. Mom supports it, you know, give it a shot and try it. I'm totally there for it.


BP Alright. Well, that means is that time of the episode where I shout out a lifeboat badge winner, that's somebody on Stack Overflow who answered a question with a score of negative three, and got it up to a score of 20 or more, awarded one hour ago to trying to learn Python can't assign to literal. So if you're interested, and if you've had the same error, we'll put it in the show notes and give you a little help. I'm Ben Popper, Director of content here at Stack Overflow. You can always find me on Twitter @BenPopper, you can email us And if you like the show, be sure to leave a rating and a review. 

SC That's great. I'm Sara Chipps, Director of Community at Stack Overflow. You can find me @SaraJo.eth. And Michelle, who are you and where can we find you on the internet?

MG I am Michelle Grover, CIO of Twilio and you can find me on, I'm on everything LinkedIn, Twitter, just about anything. [Michelle laughs] But totally reach out to me on LinkedIn. I get folks reach out all the time after these and I do try to connect with them, but just let me know that you heard it on the podcast.