Todd McKinnon co-founded Okta when most around him thought it was a bad idea. And soon, he started to wonder if they were right. Was he cut out to be a startup company builder? But eventually, he proved to everyone (and most importantly to himself) that he was right to believe from the very start.
Todd McKinnon:
I remember during that time really thinking to myself, you've got to believe, even when you don't believe. You're going to fail if you don't believe.
Mike Maples:
That's Todd McKinnon, co-founder and CEO of Okta, someone who went through the valley of despair many times, but came out of it winning big. Today, Okta is a public company with 2,000 employees. This is Mike Maples Jr. of Floodgate, and it's go time with Todd McKinnon.
Mike Maples:
Sometimes you win by doing something that's just obviously useful, and Todd McKinnon and Freddie Carist and Okta did just that, making it easy for you to log into all of your SAS apps from one place. Todd is the rare CEO who went the whole distance from the very beginning.
Mike Maples:
Today, you'll see him on TV for the latest earnings call, but it wasn't always that way. Okta took roughly four years to find product market fit, searching high and low before they did. Along this journey, Todd had to prove not only to the people who backed him and his customers, but even his wife and himself, that he made the right decision to leave a great job at Salesforce. Let's talk to him.
Mike Maples:
Todd McKinnon, thanks for coming on the podcast.
Todd McKinnon:
Thanks for having me.
Mike Maples:
Let's go back to 2009 when you started to take the first steps to create the company that would someday become Okta. At that point, your career, you've been pretty successful. You're running all of engineering at Salesforce. What was going through your mind that would cause you to leave a great job like that in the middle of 2009? Because I don't really remember 2009 being a very hospitable year for startups.
Todd McKinnon:
It started a little bit earlier in 2008. I'd been at Salesforce about five years at the time, and I had a great run there. When I started at Salesforce, the team was very small and I was able to build that team up, and when I was at Salesforce we launched the support cloud. We launched the platform, salesforce.com platform as a service. We really changed that company from a single product into multiple product, and it was really part of what I accomplished there, but I felt a sense of like how much more can I do there?
Todd McKinnon:
I was thinking about what would be next for me, and if you rewind back even before then in the first dotcom bubble, I had been at a company called PeopleSoft. I watched all these entrepreneurs go out there and start these companies and have all the success. A lot of it turned into failure, but it was very early in my career it was something I wanted to do.
Todd McKinnon:
I thought it would be very challenging, very rewarding. This is like pre going through it, my psychology of it all, but I think more relevant was just like in the dotcom bubble where you saw the internet happening, and at that time I stayed at my company, PeopleSoft, and it worked out great for my career. It got me into Salesforce. But you saw what was happening in 2005, 6, 7 around cloud computing.
Todd McKinnon:
You saw the same kind of impact in the same kind of longterm potential and transformation of an ability to create companies like you saw with the internet in '98, '97, so that was really what really spurred me to do it was it was clear to me being inside of Salesforce that this was a big deal in cloud computing.
Todd McKinnon:
When every layer of the IT stack or the technology stack was reinvented to be infrastructure as a service or platform as a service or application as a service, and all the security and all the management would have to be totally reworked. It was a huge platform shift, and that was going to take a decade or more to play out.
Todd McKinnon:
The opportunity for to make big new companies in that kind of transformation was really unprecedented. It was probably not going to happen again in my career. That was really the macro context for wanting to start it, not wanting to miss that opportunity to create a really transformative company in that context.
Mike Maples:
We often like to ask founders, what is your why now? You know, what are the set of circumstances that are coming together to make this the right time for your startup? In the case of Okta, was there a really clear why now, or was this more of just kind of a secular bet on the cloud that you were at the right place at the right time?
Todd McKinnon:
There was a reorg inside of Salesforce, and I was going to get a new boss, and it was fine. I mean, my job was still great. I was still running all of engineering, but it just really kind of pushed me to the point where, as you say, why now? It's like, it was a good reason. This change happened. I wasn't going to get the promotion. It kind of reinforced that maybe my growth there was not what I thought it could be, and that was really what in the moment spurred it. But I had, even before that, I wrote down a business plan. I was thinking about it a lot, and I was thinking about the macro idea of cloud computing. It was kind of those things come together that drove the exact timing.
Mike Maples:
How vivid is that time deciding to leave Salesforce to start your own company?
Todd McKinnon:
I remember very clearly that I finally decided to quit my job and leave Salesforce and we had, my wife and I, Roxanne, we had a six month old, our first child. Julia was six months old. Roxanne was home. She had stopped working for a while when Julia was born, and she would be at home watching the news and watching Lehman Brothers and the world coming to an end.
Todd McKinnon:
Yeah, just is this new president going to be able to save us? What's the treasury going to do? I came home one day and I said, "I'm going to quit my job. I'm going to start a company." She looked at me, and she said, "Are you fucking crazy?" See that's the funny part. That wasn't that funny in the moment.
Todd McKinnon:
With Roxanne, her work ethic and her family's work ethic was like, either you're working at a well known company with a brand name or you're a bum. There's no starting a startup and maybe finding yourself. It's pretty black and white. I talked to her about it, and what I ended up doing is I ended up making a PowerPoint presentation that was presenting the case to her.
Todd McKinnon:
We're a team. We have a daughter. Here's the case to her, why it makes sense. The title of it was, "Why I'm Not Crazy: My plan to quit my job and start a company." What I say in the PowerPoint ... I still have it. I look at it every once in a while.
Todd McKinnon:
These ideas I've outlined to you, which was: Cloud computing is transformative. All the great companies from Microsoft to Oracle were started during a recession. It makes things easier to start a company. Now, I don't know if I really believed that at the time, but it was something I read somewhere, and I said it like it would help convince my wife.
Mike Maples:
It sounds good.
Todd McKinnon:
I did really believe in the cloud computing part. I was convinced that there would be big important companies created out of this transition beyond just a company like Salesforce. If you look back at the PowerPoint today, that all seems very prescient. Oh, he's such a visionary, but what the story sometimes doesn't get told that if you actually look at pages three through nine, it's all basically fallback plans and CYA. Like, I can get another job if this doesn't work out. If it's so bad I can't get a job, I would've been in trouble at Salesforce.
Todd McKinnon:
It's like basically six pages of covering my bases. Anyways, that was probably the hardest part, convincing her to ... because she was smart. She was skeptical of these ideas, and once I was able to convince her and actually resign from Salesforce and transition out of there for a few months, I would say that's probably the hardest part of actually starting.
Mike Maples:
Did you have an idea of what the product would look like or what the customer would really need? How typical was this in terms of getting it right right out of the shoot versus kind of customer development and the Steve Blake style?
Todd McKinnon:
Yeah. One of the things Freddie and I did was we read his book, Four Steps to the Epiphany.
Mike Maples:
Yep.
Todd McKinnon:
The customer development and that stuff. It was pretty influential in how we did our customer interviews and we tried to really find the product market fit. I'm trying to think about, like I remember thinking about is it an existing category we're making cheaper or is it a new category?
Todd McKinnon:
Interestingly enough, I think our business evolved eventually to kind of be both, right?
Mike Maples:
Yeah.
Todd McKinnon:
It was like changing an existing category, but a lot of elements of a new category. That whole mindset of executing, making progress, iterating, getting customer feedback, building that into the culture was pretty influential in the early days.
Mike Maples:
As you and Freddie started working together, were there any major changes in how you approached building the product or talking to customers?
Todd McKinnon:
It was right after we started working together that we decided to pivot away from the systems management idea. We changed to identity and login and access management for cloud applications. I drew out a kind of a mock up of what the product would look like, and it actually, even now it looks very similar, the core concepts of an admin console with people and applications and policies and then a very simple end user experience to log into the applications.
Todd McKinnon:
The reason I mention that is because I think that we were fortunate that we pivoted, quote unquote, but it was very low cost pivot because it was two people and nothing else. Once we got on the core identity and got the insight from talking to all those customers about what a problem this was, when they were going to cloud the identity piece, and how it had the two very positive characteristics. One was that, as a business idea, one was that people had pain.
Todd McKinnon:
They were like, yeah, it all worked well when I had active directory from Windows and everything was Windows, but now that you know we have cloud and people are logging in from all over the place, just logging in is a real pain. People had pain, and they would pay money for it today. We got the sense of that.
Todd McKinnon:
Then the second thing was that we knew that we wanted to build a very big impactful tech company, and then we knew that if you started from this pain of just logging into cloud apps and you built the infrastructure right, you could build a substantial product suite around that that did everything from synchronizing user directories to doing multifactor authentication, and we knew that you could build a big company off that, because I think sometimes companies, I'm sure you've seen this, where they find a pain but it's just kind of nichey, right? It's never going to be really big.
Mike Maples:
For our listeners, trying to understand the indicators of product market fit, did product market fit for Okta come as sort of one big bang revelation or was it more of a gradient that kind of developed over time?
Todd McKinnon:
I would say gradient. Right after I knew we could do a $400,000 a quarter, a million dollar bookings a month, pretty quickly it was like, okay, but how big? Okay, but yeah, could you do $5 million? What's the biggest deal you could do? Could you do a million dollar AR deal with this product?
Todd McKinnon:
It definitely felt like a gradient to me. It's interesting, the concept of what we do is just obviously useful. If you didn't have a product that was obviously useful, what does it do? Well, you log in once and you can get to all your stuff.
Mike Maples:
Yeah.
Todd McKinnon:
Not every product is like that. I don't think Twitter was like that. It might be different in businesses like that, but in a business like ours where it's obviously useful, then it comes down to how useful, and is it useful enough to overcome the cost. It lends itself to being more kind of the gradual shift from being in the valley of death to being super successful.
Mike Maples:
So you knew you had something obviously useful, but was there just a limit to how fast the market could grow? After all, you get guys like Jeffrey Moore talking about the technology adoption lifecycle and just the natural pace of development of these things.
Todd McKinnon:
I think that there just had to be enough cloud, had to be enough CloudOps had to be enough email in the cloud and infrastructure in the cloud so people wouldn't think doing your security in the cloud was crazy. It's good, and it's a little bit more painful to go through. It's great because when you come out the other side, assuming you survive, is you own it. Now you're like the cloud identity person, and it's all the other companies aren't cloud identity. They're either on premise or they're not, and all of a sudden it becomes great.
Todd McKinnon:
I would say when we really thought it was going to work was when Microsoft announced a competitor, because what that meant was that it was a big market, and by that point it was pretty clear we could get customers and make money and have revenue. They don't mess around with markets that aren't big. We always wanted to get into doing something that would be big. We thought it could be big. For us we would have fallen short if we had built something that was small and maybe sold it or something.
Mike Maples:
Is this around the time you start to have your early conversations with VCs?
Todd McKinnon:
In the summer of 2009 we started going out, because remember in the PowerPoint deck, the plan was I'm going to raise some money and I'm going to get some health insurance. That was one of the milestones that I had to meet to get my wife's blessing. We started talking to VCs. We'd been really kind of, for six months we'd been out of the world in terms of financial crisis. We didn't have jobs, we weren't getting laid off, and we weren't really trying to sell anything, so we didn't see the customers.
Todd McKinnon:
Yeah, we'd borrowed an office from Jawbone. Freddie knew the Jawbone founders from college, so we were in an office and it was all great. Then we started to try and raise money. We really came head on into ... venture capitalists were a little bit scared. They were freaking out.
Mike Maples:
Yeah.
Todd McKinnon:
They're like, I don't know. My LPs. I mean, I think we're going to be able to do the capital call and get the money. I mean, I think we're going to be able to raise another fund, but they were a little bit standoffish. We talked to a few, quite a few, and we wanted to do the round convertible debt. YC had kind of talked about doing convertible debt rounds, and that was kind of getting popular, and it seemed like a good idea, so we were going to do the round.
Mike Maples:
That was the rage.
Todd McKinnon:
Yeah, we were going to do the round of convertible debt. I think a lot of people use that as an excuse. Oh yeah, we don't do those convertible rounds, so come back and see us later.
Todd McKinnon:
I learned another valuable lesson, and we have a saying for it, Freddie and I now, that there's a thousand ways to say no, and they all mean no. Don't try to decipher why they did this or what they didn't because you're going to drive yourself crazy, and at the end of the day it doesn't matter because they said no.
Todd McKinnon:
But we met Mark and Ben and Andreessen Horowitz along with Scott Cooper, who I think they were brand new. I mean literally there was, Freddie and I went in to meet him, and it was Mark and Ben and Scott Cooper and a fold-up table. It was an office they were just moving into. I remember Freddie went in, and we sat down, and he looks at the table and kind of smacks the table, and he goes, "Costco?" And they go, "Yeah." He goes, "I know, I have one in my house too."
Todd McKinnon:
I think that they thought the idea was interesting, but I think that more so they were really into the people. Right? You know, founders are listening. This might be helpful. When I was thinking about what companies I could build around cloud computing and then especially when we pivoted to the identity company, there was thoughtfulness about me as the engineering and product person and Freddie as well with his background about why we would be able to do this and have an advantage over other people.
Todd McKinnon:
The reason I mention it is it's not trying to just be self-congratulatory, but the reason I mention it is because I think you do have to figure out some way to get, call it crossing the chasm or through the valley of death, when you're building a company, because it's kind of unnatural to make it, and you have to have some advantage to get the thing rolling, whether it's you can kind of fall back on your reputation and get customers to trust you. I think most companies go through that. I think that fact that we had our reputations were able to help us get through it.
Mike Maples:
What were your goals of those early days, and how did you start to develop a rhythm of getting those first customers?
Todd McKinnon:
In all of our meetings with VCs and then of course Andreessen and Floodgate as well, we were very focused on relatively short term milestone. We're going to have a customer live within six months. It didn't seem like too hard for us, but I think for a lot of the investors we talked, to it was like, oh man. It's like these guys are setting a goal. It's a demonstrable point of proof. That was our thing. We just want to build this thing and get super minimal viable product and get it working within six months and have a customer.
Todd McKinnon:
We were all guns blazing to do this, and we were able to accomplish it, getting really a shell of the drawing of the product that I had drawn. We were able to land this customer by December 31st of that year.
Todd McKinnon:
It was a company called Xactly Software. They gave us an order for $400, and I remember I showed my parents. They were over for the holidays, and I said, "Look, we got our first order, $400." "Oh, that is amazing. I can't believe you did that. That was your expense report for a week at Salesforce."
Todd McKinnon:
Yeah, so we got our first customer, and then we went back so that we did a seed round in the summer of 2009, and then we got our first customer and we went back and we did a $10 million AA, I think in January of 2010.
Mike Maples:
When did you find subtraction? Like when did you get those early customers first starting to go live with the product?
Todd McKinnon:
We ended up spending 2010 getting five customers live. We wanted, Xactly was an order signed, but it wasn't actually live. We spent 2010 actually getting five customers sold and deployed and live. That's probably getting basically what was a mockup that Xactly would buy, that was probably 20% of the work to get five customers live.
Mike Maples:
Did all five of those customers conscientiously want to deploy the product, or were some of them just kicking the tires and maybe even experimenting?
Todd McKinnon:
We charged for it, so that helped. That made them serious, I think. We maybe filtered out some people that weren't serious. They wanted to deploy. I think of the first 10 companies we signed up, they all deployed.
Mike Maples:
Okay.
Todd McKinnon:
So yeah, they had a lot of cloud apps, so they were mostly tech companies. Like there was a company called Xactly Live Ops. LinkedIn was an early customer. The mostly tech companies had a lot of cloud, including really critical apps like email in the cloud that really kind of moved that, in their own minds at least, moved that center of gravity of technology outside their data center.
Mike Maples:
With those early customers, I suppose you didn't really have to convince him that the cloud was going to someday be a thing.
Todd McKinnon: It was pain. There was pain for cloud single sign on. If you would ask them how big could this company be, they probably wouldn't have thought it would be that big. A lot of people didn't, right? A lot of people were like, oh yeah, this'll be cool to make my users happy.
Todd McKinnon:
One of our value props early on was the service would be pre-integrated to whatever they needed. It wasn't like they had to wire it together. It'd be out of the box. It would be pre-integrated to all the applications they wanted to access. We built a lot of technology to make it really easy to build these integrations, but we didn't really have any integrations built, so we would basically just get a customer's like, what apps do you have? All right, those are the ones we're building.
Mike Maples:
That's in our catalog.
Todd McKinnon:
Exactly, yeah. It will be tomorrow.
Mike Maples:
Ultimately, we could talk about the success of Okta today, but what was something that you struggled with that was a challenge that you didn't really expect coming out of Salesforce?
Todd McKinnon:
A lot of my career at Salesforce was made on hiring people, building that team. I mean, I hired tons of people there and really scaled out the engineering and technical teams. At Okta, we didn't want to try to go steal a bunch of people from Salesforce because we thought that they would be a partner eventually. We didn't want to ruffle their feathers, so we didn't really have my network to pull people from, and you just really get a firsthand look of how hard it is to recruit for a startup.
Todd McKinnon:
That was another big, really big shock. I often say that if I had Okta to do over again, one of the things I would do differently is I would just take more time on people instead of maybe being confronted with that difficulty of hiring instead of just kind of taking some shortcuts and hiring some people I probably shouldn't have, I probably would just be patient and wait a little longer and just really build up the first 10 people more slowly. Because it would've taken longer.
Todd McKinnon:
Probably two or three years down the road, we probably would have been better off to ... It's hard to say. I mean, it worked out fine. We did get some advantage of acting fast and with a super sense of urgency, but that's probably a trade off I wouldn't make if I had to do over again.
Mike Maples:
Hiring was a challenge. You mentioned earlier getting through the valley of death. What are some of the other challenges you were facing at that time?
Todd McKinnon:
'10 was getting those five customers live. We're also hiring people. We made a lot of progress in '10. '11 was a really hard year. The reason it was hard is because I think a lot of it, like in life, comes down to expectations. In '10 we were new and everything was great and we were starting out and the future was all bright and we're the new kid on the block and so forth.
Todd McKinnon:
By the time '11 came around, we'd been around a while. Been like around two years. We've been around a while to have expectations be high, whether it was the overly optimistic plan we've put forward to the investors or it was our own expectations. But we really hadn't been around long enough to do stuff. It's only a really short amount of time, like two and a half years or something, or a year and a half, so '11 was tough.
Todd McKinnon:
That was one we had to really grind out. In '11 we were trying to close our series B, and we put a plan in front of the investors for that quarter to do like 300 grand in bookings. We delivered 40 grand. I think we were a little bit reticent to talk to big companies. We were trying to find smaller companies because we think they would be easier to sell to. We were kind of stumbling around figuring out who to sell to.
Todd McKinnon:
It was hard. It was like, missing sales targets, employees starting to feel like maybe there's not a cloud. This is just a niche. How big can this company be? I remember during that time really thinking to myself, you got to believe, even when you don't believe, right? It's not going to work if you psych yourself out and get down on yourself.
Todd McKinnon:
You're not going to be able to leave the company or you're not going to be able to do the job to get through this. It's impossible. Given that as a condition, because you're going to fail if you don't believe, usually in your job, if you're working at a company, your success and the markers of your success are kind of out there.
Todd McKinnon:
The company has metrics and you have metrics, and you're kind of like peers in the organization. You can see how you're doing, but when you start a company, you're searching for that or that's missing. What you start doing is you start looking at other companies, and you start saying, well, when did that company get funded?
Todd McKinnon:
Everyone you talk to is full of shit. They're like, "Oh, we're killing it. Oh yeah, yeah. I mean, we're doing, you know ... it's not that great. We do like multimillion dollars of ACV every month." We're going like, what? It's daunting. It's daunting when you're going through it, and you've got to figure out how to keep pushing forward even if you don't think it's going to work and you think you're wasting your time. It's like, I should be doing something else. For us, I think it was, there's companies that it lasts longer for. I don't even think we had it as bad as other companies do.
Mike Maples:
When did things start actually looking up or did they keep getting worse for a little while longer?
Todd McKinnon:
I think that was the low point. Then in Q3 I think the number was like 400 grand, and we did 410. The vibe at the board and with the management team was just ecstatic. I was like $10,000 over the number.
Mike Maples:
Yeah.
Todd McKinnon:
It was like, this might work.
Mike Maples:
Yep.
Todd McKinnon:
Yeah, this might work. I think it was not one thing. It wasn't like we changed our sales methodology or we went after a different target market per se. I think it was just, we were a little bit more aggressive about bigger companies that could pay a little more.
Todd McKinnon:
We were really reticent in the early days to sell the big enterprises because I didn't want to turn it into a company that was building features for one big customer. I didn't want to get the roadmap kind of railroaded by one big customer, and also part of it was just Salesforce.
Todd McKinnon:
Salesforce had a lot of success with SMBs. That seemed like a successful company to follow, even though the products are very different. We beat the number in Q3, and then we had a great Q4, and then the next year, early in the next year in '12 we added a couple executives to the team. One was Adam Aarons who you helped recruit and then Hector Aguilar, who still is with us today and is president of all technology.
Mike Maples:
So far we've talked about the product and how you grew the company, but I'm also curious about culture. Did you do anything intentional to define the culture of the company and just kind of a sense of identity for your employees and coworkers?
Todd McKinnon:
Transparency is ... A lot of companies say we're a really transparent culture. We share everything. We're definitely like that, but the purpose of it was to get people sharing in the problems and owning the solutions. That was the mindset. The mindset was if we can just share this problem ... Listen, we're missing our customer number, but we want to get to 50 customers, and you guys got to figure out how to do it.
Mike Maples:
Were there any lessons from those early days about just thinking of yourself as a CEO and how to become a better leader?
Todd McKinnon:
I think for me as a leader, there was a lot of learning and going from where I thought I had to kind of bear the weight of it all on my shoulders to kind of sharing it more with the team. It's a little bit counterintuitive, but it actually empowers people. It doesn't freak them out. It empowers them. They're like, this is great. I mean, not great, but like, "Oh, we've got all these problems and I've got to help solve them."
Todd McKinnon:
I think there's something there just in general about leadership and building a company in a team that it will own it like their own and be fired up about the objective. You've got to treat them a certain way. You've got to include them. You have to trust them. You have to make it feel like it's their thing, right?
Todd McKinnon:
It's like they own it. They own the outcome. They're going to benefit from it. Get them really emotionally involved. I think that's really important. Even still to this day, I make sure that whenever I talk to a new employee or something or talk to someone I'm interviewing at Okta, I tell them that.
Todd McKinnon:
I say, "We're not always perfect, but we really, really are working to make sure that we have this huge advantage over everyone else, especially bigger companies that we'll make this feel like it's your company, and you'll be motivated to do amazing things, and given the flexibility and especially when you talk about some of the companies, the big companies we compete against now, they can't do that."
Mike Maples:
Somewhere out there, there's a founder wondering if they have the right idea in the first place, probably deciding to believe even if they don't necessarily believe. Is there anything you'd like to pass on to those folks?
Todd McKinnon:
One thing I would say is the people that ... I'm not talking about myself, but you don't have to be ... it's not like these people are superhuman. They just did it, and they were in the right place, and everyone gets lucky. I certainly got lucky. I mean, it can be stressful because you've got to rely on a little bit of luck and timing, but it can be liberating because you have to rely on luck and timing.
Mike Maples:
Well, for what it's worth, it's been a pleasure working with you. Thanks for letting us be along for the ride, and thanks for hanging out with me today on our podcast.
Todd McKinnon:
Thank you very much. Enjoyed being here.