Persona — Solving Identity Verification (Research Radio)

An episode of Research Radio, Contrary Research’s podcast, hosted by Kyle with Rick Song, co-founder and CEO of Persona. Recorded July 19th, 2023 and published November 9th, 2023. About 45 minutes. Listen on Apple Podcasts.

Summary

Persona is an automated identity-verification platform: machine learning, computer vision and algorithmic checks for KYC compliance, age verification and fraud prevention, with customers building their own identity workflows. Rick and Kyle cover:

• The origin. Rick and his co-founder Charles came out of Square and Dropbox, where they saw businesses stitching together an ensemble of identity tools while vendors pitched a single silver bullet. Investors at the seed told them the space was commoditized. Rick’s lesson: the harder question is whether a space is consolidated, not whether it is commoditized. Payments and messaging both looked commoditized before Stripe and Twilio consolidated them.
• The thesis. There is no one-size-fits-all identity. Identity has a lifecycle (sign-up, account recovery, cash-out), and the data has to be managed and orchestrated. Those pillars haven’t changed; the tactics shift as new technology arrives. Rick cites Amazon and AWS as a company that bet on the unchanging customer needs while adopting new technology.
• The platform. Three years earlier Persona bet on making every product a standalone building block, so customers can enter through verification, the Graph product, workflows, data collection or KYB. Rick says two of the five largest companies in the world run significant parts of their business through Persona.
• Consolidation without lock-in. Rick wants consolidation but not centralization, pointing to GitLab’s approach and Cloudflare competing within AWS’s world. Kyle draws the parallel to open-source vs. closed AI (Llama 2).
• What scares him. Not failure but success: a world where one private entity controls everyone’s identity. Persona is designed as infrastructure, so the business a person chooses stays the one making the identity decision.
• Physical vs. digital identity. The biggest fragmentation: Reddit or Gmail shouldn’t care how many accounts you have, but delivering alcohol, getting a degree or opening a bank account has physical consequences that sign-in-with-Google doesn’t solve.
• Advice for founders. Understand yourself and your belief system, since it shapes the culture, and embrace the journey rather than fixating on the destination.

Kyle’s notes

• Did podcast with Rick Song at Persona
  • The Hits Business
    • Matt Levine made this point that Adam Neumann is one of the few founders has proven they can deploy (and destroy) $10B+ of capital
    • Michael Bay is the Adam Neumann of Hollywood’s hit’s business

Transcript

[00:00:00] The biggest fragmentation we’re seeing in identity today is an increasing kind of a realization that the idea of a physical identity and the idea of a digital identity are fundamentally different from each other.

[00:00:18] Welcome to Research Radio, the official podcast of Contrary Research. Contrary Research is the best starting place to understand any private tech company. In each episode, we dive deep into the most important conversations and companies in technology. This show is your first step to understanding any startup. I’m your host, Kyle Harrison, general partner at Contrary. For more info and to read our full research reports, visit research.contrary.com. Myself, guests, and Contrary may have financial interests in the companies discussed. This podcast is for informational purposes only and should not be taken as investment advice. As always, do your own research before making any financial decisions. Today, we’re talking about Persona, a company tackling the difficult problem of managing online identity.

Our lives are increasingly intertwined with the online world. Five billion of us are connected to the internet, with each of us spending an average of nearly seven hours a day online. From work and education to shopping and entertainment, our digital footprints are expanding at a pretty unprecedented rate. But every time you order food or apply for a loan online, there’s a business behind that transaction that has to figure out if you are who you say you are. Especially when they’re sending you money, that’s not something they can really afford to screw up. Persona wants to be the identity verification engine that makes more of your online life possible.

Founded in 2018, Persona is an automated identity verification service which uses machine learning, computer vision, and algorithmic checks. For businesses who are trying to figure out who exactly you are, Persona handles everything from know your customer or KYC compliance to age verification and fraud prevention. And customers can customize their own identity workflows and create reports to keep track of their efforts. So for this week’s episode, we were joined by Rick Song, co-founder and CEO of Persona. He walked us through the platform, how it works and how it wants to solve online identity. Please enjoy. Rick, awesome. Thank you so much for joining me. I’m super excited to get to chat with you.

Likewise, Kyle. Thank you so much for having me today. I use you as a story all the time when I talk to earlier stage founders, where when you’re sort of thinking in this problem discovery phase, you know, opportunity canvassing. I love the fact that you look at what you built with Persona, and it’s very much a lived experience for you, right? With you at Square and Charles, your co-founder at Dropbox. And so maybe unpack that a little bit. What was this experience you had? And what made you realize that it was a big enough problem to say, hey, there’s not good enough solutions out there that I would just want to go buy.

This is something that I need to go create based on the experience that I’ve had. What was that discovery process like for you? I’ll be particularly candid on this. I think in the earlier days, I think there were more revisionist history of how we came about this. But, you know, these days we’re a bit larger, and I can be a little bit more candid in the history of it all. I love it. The truth on these ends, I didn’t think too much about the opportunity size. In fact, like when we had first gone off and started Persona, and actually even well into the history of it.

I think the consensus opinion about this space was that it’s commoditized. That was actually one of the number one concerns. When we were talking with investors for our seed round, one of the most common things was, this isn’t a new space and isn’t it largely commoditized? How is there innovation left? I have opinions about that, but my background is engineer. So in all honesty, at that point in time, I was just learning what TAM meant. I had no idea. These days now I know actually the acronym. I don’t think I still fully understand it, but at least know what the letters stand for. That wasn’t kind of what had crossed our mind.

Actually, when we were preparing our first seed deck, like I think one of the friends who was helping us was like, hey, you need to like talk about the TAM of this. And I’m like, what the heck? How am I supposed to figure this out, right? And then you’re reading random McKinsey reports trying to like add up numbers and be like, yeah, why not? This makes sense to some degree. Compound angle growth rate or something. And yeah, it makes sense. At that time though, in terms of like, why we kind of like thought to solve this idea, it wasn’t that the idea was big enough as much as we like this general problem set.

And the disconnect that we were seeing was that at every single business, at least at Square for certain, a lot of the other businesses out there, what we hear vendors talk to us about was that there’s this one-size-fits-all for identity, that there’s going to be some holy grail, you know, silver bullet towards every solution was that silver bullet too. This type of way in which you think about identity, you will like solve all your identity problems out there, right? And at Square, on the other hand, we were using an entire ensemble of solutions. And not to mention customization, mixing, matching, thinking about how the journey would increasingly move from being this, let’s collect everything of the world about this individual and add as much friction towards this more progressive journey in which you gradually collect data, make decisions real time, and modify that.

And there’s this increasing disconnect between how I think we in the industry were thinking about us as customers versus I think about how a lot of vendors were talking about how everyone’s going to be like this next silver bullet towards solving identity. I was pretty passionate about this idea. At the very least, I liked it enough such that I spent many years kind of within the general identity space. One other thing that we were seeing a lot of at that time, too, was this idea that there wasn’t identity for different perspectives was just going to be different. Like we always think about like identity in the same way at that time as authentication, which like, oh, we’re going to use like one set of credentials and like, you know, this idea of like there’s going to be a, again, a silver bullet towards it.

What we were seeing instead in the industry is that if you were for caviar, for delivery, versus if you were signing up for merchant onboarding, versus if you’re signing for cash out, the requirements were different for every one of these. So the early genesis actually was much more about like, well, this would be an idea that I wouldn’t mind working on for a bit longer. We used to make rather grim and dark jokes. One joke that we used to say was like, well, you know, at least if this startup goes nowhere, I can go back to Square and be like, hey, at least I worked on something that’s relevant to what I was working on before.

So, you know, let’s call it a wash. Sure, up the resume. Yeah, yeah. Yeah, yeah. I don’t have to dance around this topic of, I just waste away a year of my life trying to do a startup or something. It would be a little bit easier to go back towards it. But that was actually a lot of the genesis. And it was interesting because we hadn’t thought so much about the opportunity size, actually. And in fact, so much of the messaging at the time, and a lot of the external messaging that we had gotten, was that the opportunity was very small. That the space was commoditized, that there wasn’t much room for innovation.

And we kind of just pressed through it. I joke about it at this point. I will say, actually, in the early days, I was very dismayed that everyone had felt so negatively towards us. Five years later, I think identity has become one of the most exciting spaces out there. And at the very least, I’m very grateful that it seems like the consensus opinion about this space has changed. It’s an interesting sort of lesson to learn where it’s like there are a lot of people who beam counters is like the meanest way to say it or whatever. There’s probably a meaner way to say it. That is a way to sort of simplify it.

But these people who they really know what TAM means, they really know how to think about it. And they’re, they’re sort of microanalyzing all this stuff for you. You’re just living an experience where you know that this is a problem and you’re interested in how to solve it. And so you tackle it. And then over time, it’s sort of, you realize that actually like the reason you’re having this problem is because it’s not a solved problem. But I want to unpack that a little before we dive into persona and how the product has evolved. You think about how the space has changed. You talked about how a lot of people’s concern is that a function of it is commoditized. when I think about the space, I think that there are elements of this sort of stack where there are a ton of vendors who are doing the same thing.

Everybody’s race to the bottom on one piece of the puzzle. It seems like what you guys have unlocked is that you’re trying to be this sort of central switchboard control center, if you will, for everything that’s going on in the space, which makes you kind of an index of all of the things that need to happen in identity. That’s my understanding from the outside in, but how has the space evolved and how have you kind of evolved persona to fit the moment? There still has to be. And on this talk of kind of like monetization, and this was not things I was thinking about when we had first started Persona, but like, I always like to kind of think about like almost the meta analysis of kind of like what we’ve learned about, you know, starts throughout it all.

This idea, actually, this interesting kind of thing about how most spaces online does consolidate. And actually, I think the much more challenging thing is whether a space is consolidated or not, not whether it’s commoditized or not. When a space is consolidated, it’s very difficult to break in. but generally speaking i think a lot of folks mistake the two one for another right so a classic example of this would be um strike in the early days of striving just in payments in general there was a lot of talk at that time that uh payments has commoditized this is obviously many many many years ago but like i remember actually because at that time i just come out the valley there was a perspective at that time that what stripe was doing was a very like commoditized construct there wasn’t much value he had in that because of how commoditized this it will be impossible for that space to consolidate.

Fast forward to today, it’s actually very fascinating. This space has 100% consolidated. There are very few mega players right now in payments. There’s still a handful, but it has largely consolidated amongst, I would say like four or five players and really just one or two at this point for the new era of payment technology. And these days I have a perspective now, even for say a Twilio, which like there was a perspective, I think like mainstream opinion that maybe five, 10 years ago, that that space was rather insipid And effectively, like SMS and just online communications would be largely commoditizable. And anyone could kind of like offer an offering of an API that allows you to send text messages today.

Fast forward, clearly has consolidated around Twilio. Yeah. And these days, I think a lot about how pretty much any space in which you see a lot of hyper competition, where there’s a consensus opinion of commoditization. There probably actually is, if someone is able to crack the equation, a way to consolidate that space. because the fact that there’s so much competition, but still no consolidation means that at the very least when the final space is online, that there is an opportunity for consolidation. On that front end, in terms of how it ties kind of into personas, like kind of strategy, I think our strategy has evolved quite a bit, but at the same time been the same.

The core thesis has always been this idea that there’s no one size fits all towards identity. The identity is like this, like every evolving kind of thing. One thing that made that very clear was that, let’s say 15 years ago, the de facto way most people prove themselves online and who they are online is through social security numbers. These days, I think most people would all have to understand that social security numbers have largely been leaked, at least for US identities. After that, there was this whole period in which people would ask like back of envelope questions where it’d be like, what’s your mother’s maiden name?

Or what was your first street address when you were like five or something like that, right? And now that has kind of like gone out of vogue. And there’s this idea at that time, which like identity is like the way in which we prove ourselves online. It’s like this ever evolving thing because the tie between your physical and digital self is a probabilistic tie. During the era, you know, the boom of kind of like a crypto web three, there’s a lot of perspective, like, can we digitize like all aspects of identity? And like, the reality was that that became a little bit more complex. I think a lot of folks, I think there’s still a lot of innovation within that side of things.

But it wasn’t just as simple as, let me just digitize everything online, because sooner or later, that credential now can be shared and become a proxy that may not represent who you are physically. And this space is fundamentally about understanding who’s behind the screen. So the more you digitize, you have this very perverse thing in which you have a less fidelity in terms of who’s behind the screen. So that was kind of like one of the key things. But in terms of how we solve this problem, a lot of things have evolved since that, right? Early on, we made a huge bet on making sure automation was a key part of our platform.

But we also embraced a lot of things around evolutions around, I think, around 2019, 2020, there was this big push around no-code kind of technologies. And we looked a lot towards that in terms of how can we apply the methodologies there to improve the ease of use and the capabilities of our platform. So we embraced a lot of ideas around kind of like the no-code workflow orchestration, retool us like dashboard building tools, the ideas that they had, but remixed it into what would it look like for an identity platform. Today, I think there’s a whole boom around like AI, generative AI and LLMs. And we’re looking a lot at that, a lot of the most innovative startups there to think about ideas there in terms of how can that be applied towards what would be this universal identity platform that can support any business. business out there.

In terms of the tactical aspects of the strategy, it’s constantly evolved. And I think that every single time it evolves, it takes a lot into account of what’s kind of like these major innovations in almost parallel industries, or even sometimes the complete tangential industries, and how can we apply ideas there towards our business to kind of continue forward this idea of what would a identity platform be? It’s so interesting that you say that way, because I, so beforehand, I was sitting down thinking about your product and how it’s evolved. Even when we first met, I think primarily around ID verification, right? And sort of automation. You’ve mentioned three pieces of technology.

I think all have played a really critical role or could play a role in your business. That first is you mentioned automation. And within ID verification alone, there was a lot of… Basically, the old solutions were literally it would send the photo of an ID to a sort of big warehouse of people. And they’d literally look at it and say, do these have all the markings of an accurate US ID or whatever country or whatever? And it would take minutes to get back or whatever. So automation was a big piece of that early part of the puzzle. And it sort of maps to this framework that I have for all of the things that sort of encompass the identity space, right?

So it’s sort of, if I think about one bucket, it’s like, you have to gather the data, whether it is photos of IDs or social security numbers or whatever. And you guys applied a lot of automation to that. Then you go to the second piece where it’s like, all right, we’ve gathered the data. Now we need to process the data, whether that is the process of gathering the data itself, or we need to like get it somewhere, go through, it might be workflows, It might be third-party checks where you have to ping the social security department or whatever. And you talked a lot about no code.

And it feels like your workflows have taken on a lot of those drag and drop functionality. Now, the third piece of it, I think, is really interesting and not something I’ve been as close to sort of since moving away from your board and stuff like that. So maybe this has played a real role over the last year or so. But I think the third piece is sort of reviewing the data, whether it’s like a simple report where you say, hey, these are our rates or whatever, or something much more complex. you guys have this really cool graph visualization where you can take a look and understand your data.

And it feels like actually from an AI perspective, there is a function of like asking questions of your data, which is becoming much more common prompt engineering. And it feels like that could actually also have a real implication for your product. But that’s like the framework I use when I think about this whole space. And you’re able to offer your own solutions, plug in other solutions, what have you. Is that the framework you use to think about the product? Or would you do you think off base how would you frame the i think that’s incredibly accurate in terms of how we’ve kind of gone through this overall space right and i think it’s always this combination for us which i think you actually so ultimately kind of put together it’s this idea of on one hand like the core thesis of what we believe in for identity has always been the same and there’s like three fundamental pillars there that we think a lot about right the first of which is uh no one size fits all flexibility so important if you want to build a universal identity layer that works for any industry.

The second is that there’s going to be a lifecycle identity, right? In which like you’ll verify yourself across many different touch points. Let’s say when you first sign up versus when you recover your account, there’s many different points in the lifecycle. And the third of which is helping the management and orchestration of that data, making sure that the data itself, there’s more regulations, more compliance requirements about that data than ever. And there’s, you know, like increasing regulation. And for businesses, managing that data is harder than ever. We want to help how you review it, how you kind of move it around, how you visualize it, all of those different aspects, those are the three core pillars.

Those pillars haven’t changed since the earliest days of Persona. Those were always like kind of core bets that we have. What I think is really interesting and kind of what you’ve highlighted there is this idea of, at the same time, the world of technology is changing. People, I think, sometimes are overly dismissive of it being a hype cycle. And is it overhyped? Almost certainly. But at the same time, there’s a reason why it is hyped. And it’s because there’s a lot of innovation to be unlocked. And I think reapplying those major trends into the lens of what your fundamental thesis as it is as a business is always really important.

I think a lot about how strategy evolves in that sense. I always think Amazon had a really good encapsulation of this, which is like the idea for Amazon in the earliest days, like ultra consumer centrism and moving fast, right? Like that bet on the immutable. So I think the famous thing from Jeff Bezos one time was, there’s never been a customer in time who was like, I really wish my package arrives later. There’s always the immutables of like what you should bet on. So good. Yeah. First principle stuff. But at the same time, Amazon has also been a company who has consistently been at the forefront of new technical innovation, right?

AWS was way ahead of its time in many ways of like when it was first conceived and thought of. It’s a great showing of a company in which they both always have had a good idea of what the foundations, the underpinnings, the first principles that drive the business, while at the same time embracing the new ideas within a technology. And I think that’s something that we think a lot about, which is what are our four first principles that we believe apply to identity space? While at the same time, you know, what are trends that we think are 10-year trends? As a lot of investors oftentimes say, you want to be on those 10-year trends.

Well, at the same time, not losing sight of, there are some really cool things happening right now. We should consider it. Like, think about how it applies to our business. I think it’s foolish to rebrand your company as such. I think these days we see a bunch of companies where it’s like, oh, we’re not like this. We’re like an AI company. It’s like, all right, let’s take a step back on this. Yeah. You don’t have the persona.ai domain. That’s it. You didn’t switch over to that. All that way. But at the same time, there’s a lot of ideas there in which you probably can improve your business and strategy with.

I think that’s a really good way to think about it. If you start to unpack your product, right, where you’re finding the opportunities in the surface area of your product to leverage different technology and what have you. But there is sort of this, like, there’s the strategy and it’s sort of this lagging reality where there’s like, today, we are solving these things for our customers right now. Like they need these things right now. We’re selling it to them. They’re measuring us on that. And then there is the sort of longer term strategy of how our product evolves. And you mentioned this a little bit, the persona product has sort of come a long way where again, in the early days, ID verification, taking pictures of photos, doing it very quickly.

That was probably the gateway drug, right? Where it was like, that was what people got into persona. Is that still true today where there is sort of the gateway on ID verification and then it moves in? Has another product in your sort of suite emerged as like, oh, more and more people are starting to find us because of the workflow tool or whatever it might be, or how has it changed? So three years ago, we made a bet that We didn’t want to be a single gateway type of product in which we have only a single way into the platform. In all honesty, I think at that time, that was a rather contentious decision for us, like both internally and even externally.

I think like it was one of these in which we had a lot of uncertainty, actually, from I think many, many different parties. In retrospect, that decision has been pretty darn good. And for a lot of reasons, right? At that time, it wasn’t just because like, oh, we’d love to have a platform business in which, you know, there’s all sorts of ways in which you can enter into business. the theory at that time was that first off like flexibility is such a core thing but we have like our core kind of flagship of this idea of like focusing essentially on visual forms of verifications of biometric based or id based documentary based like a lot of visual forms that that just won’t be the end-all be-all and we always believe that that won’t be the end-all be-all we already were experimenting with database based like kind of verifications integrations with external parties we understood already that there was like a whole like kind of like ecosystem so we put a pretty big and not to mention like some of the newer products we were developing at that time, we thought a lot about like, how could those end up becoming like somewhat independent?

Because, you know, why not? Like, if we’re really trying to build a platform and ecosystem, everything should be a building block. And at that time, one thing we thought a lot about was like how the magic [?] of AWS isn’t that it’s just like allows you to operate your infrastructure on the cloud is that you can get started on AWS from almost any one of the products that they have. And over time, gradually migrate your way there. Because I don’t think that the future of most people how they adopt persona will be like, I’m going to take on all the persona platform, It’ll probably be through bits and pieces.

And those bits and pieces have to be the ones that are like the biggest pain point for those businesses. So that was the perspective about three years ago. And today, I think like that’s definitely become really true for us in which like we have a lot of folks now who even use our SDKs. Actually, this was just from a couple of days ago in which like they’re doing pharmaceutical kind of like data collection because collecting kind of personal information, especially in relation to health records is incredibly sensitive, incredibly kind of regulated. So they’re utilizing our SDK to help create dynamic flows in terms of fulfilling prescriptions.

That doesn’t require any sort of verification whatsoever. It’s just collection of sensitive personal information. We have folks who are using Graph as their core entry point in the personal platform. Workflows has become something in which people utilize it just as the core decision-making platform. And they already have a lot of other vendors that they rely on today. I’d like to believe that one day they’ll utilize us. But those vendors, I think, are working for them, are fantastic for them. And so we want to make sure that they can continue to have the great experience while at the same time leveraging the workflow technology we have internally.

And then, of course, one of the most recent things for us is leveraging a lot of internal tools for stuff like KYB. And where we specialize for KYB is in the business lookup records. What we realized at KYB, a large half the complexity of it is the orchestration of getting all that data, making sure all the beneficial owners are who they say they are, making sure that you can create a quick flow to collect all that data and fan out and go through all the operational tasks and all that’s built on these core building blocks of Persona. And I think that overall perspective has, you know, especially in today’s ecosystem around tech, I think it’s added a lot of resiliency to the business as well, because it’s made it such that we aren’t just pinned to the trend.

We can continue to kind of attach ourselves to these macro trends that we believe, these 10-year trends that we believe in, the flexibility, the management of the data. These are ideas that we can really pin the overall business to, which helps insulate us from, I would say, the ups and downs. Is biometric good or bad? I don’t know. And I don’t want us to just be a biometric business. What we are is an identity company. Biometrics is one form of identity, and it’s a very powerful form. It’s a very novel form. Something that you also kind of hinted there when you talk about like, hey, like our expertise is sort of in this area of being able to manage all this data and these different processes.

And so it sounds like, you know, I know there’s a lot of sort of third party tools out there, whether it is like, hey, somebody else is going to go out and be a specialist in scraping business filings and getting the information there that you can ping against or whatever. Same thing with different types of IDs or whatever it might be. So when you think about your current customers, what would you say, like, if it’s a percentage or a rough ballpark of all of them using Persona in addition to third party tools, whether indirectly, like they’ve have relationships with third parties that they’re using in tandem with you or potentially through Persona, right?

They’re sort of pinging something that you guys have built an integration with. How interconnected is the Persona platform with all these other third parties? So the answer is both. And these days, especially as we work with larger and larger customers and as we move towards the enterprise space, two of the five largest companies in the world now operate a significant part of their business through Persona. For those type of businesses, they have a ton of existing relationships and a lot of existing technologies already within their platform. We have not replaced all of that, not even close. I would say we’re just scratching the surface. We’re tackling the small kind of challenges they face. and even for them they both have existing relationships and also like they start adopting new kind of you know like vendors through persona in our marketplace in which they can integrate with you know other kind of vendors on the market as well so both occur actually right now for that i think it’s a combination of all of those right in which like sometimes if you’re for some smaller businesses who are just starting off they find a lot of success by starting with a clean slate and having everything kind of housed on a singular platform at the same time when they find something that doesn’t work or we’re missing functionality or the functionality isn’t exactly what they’re looking for, they integrate with other vendors through our platform.

And I think a lot of earlier stage companies find that to be a really kind of like great approach. And I think it’s really important for us as a business to like have the humility to say like, our solution may not be the best for yours right now. Maybe one day it’ll get there, but today that’s not the case. So there are other vendors out there who I think do a fantastic job. There are other players out there who I think have done a really great job and we want to make sure that you always are building the best identity experience for your business. For larger businesses in which they have a lot, then in those sort of situations, either they already have existing integrations with external players, and the migration towards like having us at that time centerpiece, the control center for all these different parts, it’s a gradual migration, happens over time.

And generally they’ll adopt certain solutions from Persona while at the same time, you know, gradually kind of doing more of the orchestration through it over time. At least I like to believe that. It also seems like when I think about this space, and this goes back to a previous point you made where you sort of went from tackling what felt like a very commoditized space, where it’s actually like there’s a lot of varying degrees of quality of the different products that are out there and all these different reasons why you can exist in a space where lots of other people are trying to offer a similar thing.

That’s sort of phase one. Now you start to think about the other piece you talked about, which is consolidation. And it’s certainly, as far as I can tell, the identity space, maybe there’s some consolidated players or pieces that people have brought in house or whatever, but it’s certainly not consolidated. It’s a big network of lots of different things that specialize in different areas and spike in things. And even I look at I remember a year or two ago when Clear, who I’m sure a lot of people may not know this, but they’re familiar with Clear as just this thing that makes it a little easier to get through the airport security line.

It’s actually like they also are sort of pinning themselves on identity. Like that’s just their entry point is, hey, we need you to fill out your information and scan your passport and do all these things. And then, boom, we’re an identity company. And also, yeah, it’s easier to get through the airport line, whatever. right? Like everybody has these different entry points, but at the end of the day, people are trying to build this unified identity company. And I think about other spaces like companies like GitLab, where they really had this idea of, hey, we want to be an end-to-end solution. We want to make it flexible to your point where you can use other tools at various levels of CICD or whatever you want to do.

But at the end of the day, our dream would be that we have all of those things. We are this consolidated end-to-end platform. So when you think about the competitive landscape in identity is that the vision for you guys is basically to build or buy end to end everything that you would need in the stack in the long term or is there always this need for checks and balances of multiple players doing multiple things i actually uh on the side i love that you bring up gitlab because they’re one of my favorite companies so i’ve never met sid i won’t pretend as if i know at all who this guy is i’ve you know creepily watched some of his like podcasts and interviews online all the key parts of the gitlab handbook i mean there’s so much in it, but I love how everything’s just complete open source in this company.

I learn so much from him. I had the chance to invest in [Sid] when I was at Coatue, and you guys would really get along. You really should meet, because it’s very much… You just can see the world very similarly, which is really cool to hear how much respect you have for him. He’s a remarkable leader, truly. And I think GitLab’s a remarkable company. With regards to that, I think they’re spot on in terms of they’re also one of the companies in which I think a lot about from a strategic perspective of I appreciate how they have executed it. I love that they think about the whole end-to-end like CI, CD, developer experience lifecycle from end-to-end, while at the same time having a perspective.

They’d love if everyone’s operating their whole pipeline through GitLab, but at the same time have the humility to understand that each one of these are just parts of it. And everyone has preferences and trying their best to make sure that every one of these has competition. I think that it’s important that on every aspect of identity, there’s continued push towards innovation on it. And the way in which to achieve that has to be through some degree of competitive pressure. I think otherwise space is languished. There becomes a lack of innovation. There becomes a lack of kind of like progress in that overall space. And for a space as critical as identity, I think it’s important that we’re always pushing boundaries on it and not creating an ecosystem in which like effectively, there’s no way that people can compete anymore.

I think that’s a terrible kind of outcome for this overall space. That’s why I think for AWS, like there’s a great aspect of, despite AWS owning so much of the overall cloud infrastructure space, you have businesses like Cloudflare that just, they execute from a CDN level, infinitely better. I don’t want to disparage Amazon anyway, but the reality is I think Cloudflare is an infinitely better platform right now, at least for what they do in comparison to some of the offerings right now for a lot of the cloud platforms. I think that the long-term identity is that there will be fewer players. I think that there are so many players today and I think that might have been an offshoot of the last 10 years of the great growth of tech.

And I think that we’re going to probably see a smaller number moving forward. And some of that already is happening today. At the same time, I hope this space doesn’t become one in which it’s impossible to innovate in. I hope that there’s always like for every aspect of the lifecycle identity, whether that’s the orchestration aspect, the data management, the verifications and all the new forms of verification that are releasing, whether that be MDLs, new digital wallets, all these forms, that there continues to be a push towards like embracing those new technologies as opposed to like extinguishing new ideas on those fronts because you have some sort of lock bolt on this overall space.

And that’s why I think like the GitLab kind of perspective is really great, which is I hope that for every aspect of that lifecycle of identity that Persona has built and has competed on and has our competitive offerings on, that those can stand on their own legs. And actually to tie it back to the previous point you made, that was a lot of the theory of why we thought so much about this. One of our core parts of our strategy is decoupling the platform, making sure every part of it’s its own independent entry point into the Persona kind of ecosystem. So the push at that time of why we thought decoupling was so important wasn’t just because like, oh, you know, maybe like, I think that we don’t want like a single just gateway persona.

The idea at that time was that every one of the products that we create should be independently a great product. I think it pushes us from a product perspective, pushes from an internal perspective. And I think the worst happens when everything rests on other things and you just have subpar products that people are forced to use because they bought, they’re already pinned on like, you know, your core product. I think it just builds a bad platform. And long term, it creates a very fragile and brittle platform in which once your entry point is no longer in vogue, the rest of the business kind of effectively becomes fragile as well.

Anyways, towards the long point of this, I think that there’s going to be consolidation. But at the same time, I hope that we never build business in which you’re just locked into it. I think that’s not the right way to think about how to build a great platform. And I think GitLab has done that better than anyone. It seems like the distinction that you’re drawing is the distinction between consolidation and centralization, right? If you get to a point where everything is really centralized into one, maybe two vendors, there’s a reason that we legally try and avoid monopolies because it makes it harder for everybody to make money and do well and get a good product and be well served and all these things.

And maybe another analogy that we can draw timely now won’t be when we release this episode in a little while, but Meta just launched Llama 2, right? And it’s basically the first real hit in the open source side of building a foundation model in AI. And what’s been really interesting to watch, and Contrary Research put out a deep dive recently about this sort of centralization of AI versus openness and what role open source plays in AI. But it’s crazy how even a company like Google that has historically been built in AI, at least, as a very open company, right? You look at the Transformer paper that basically made GPT possible, all open and enabling everybody else.

But as soon as ChatGPT launched, everybody was just like, oh, lock it down. Like, we got to take control of everything. And basically, Meta is the only company that, at least the only large company that has really taken a sort of open-minded approach, an open source approach in AI. And it seems like the fear there is that the technology and AI gets so centralized and so controlled that it makes it much harder to control for things like bias because you have a few companies that control the entire conversation. It’s even scarier, it feels like. And I mean, AI, you know, who knows the sort of we can get philosophical about the existential threats or crises, whatever.

But like, even in identity, it feels much more palatable, because people have seen what happens with things like identity theft, and what have you, if you have this sort of centralized model where all of the systems are being controlled by one company, that starts to feel really intimidating, right? They say or have a lot of control over your ability to interact with the internet writ large, because they control your identity and all these various points. And so it feels like when you think about those kinds of potentially like existential threats, does it feel like civilization is a big deal? Or are there other sort of maybe it’s more of a cyber attack evolution that is more intimidating you?

Or kind of what are the things that scare you about this space or that you’re sort of actively fighting against? Identity is a scary space. I’ve joked actually in the earliest days of Persona, one joke I used to say to Charles, like, it’s scary if we fail. But I’m no stranger to failure, Charles. I’m not that scared of it. What I’m much more scared of, if it works out, like, what if we build something where we hate? a fascinating thing. But I think like these days in the world of tech, you know, we’ve had unbridled optimism for 10 years. And there are certain things that I think like have been released now within the world of tech in which I think many people may look back on and be like, maybe this was not the best thing that was built.

And I don’t think it might have been better for society. We don’t know. Dirty’s out right now. But like, there’s definitely suspicions now of like certain things that some of the largest companies now are concerned if this was good or bad. I think for identity, like obviously, there’s a tremendous number of threats from over-centralization and all that, right? I think the cyber attack side actually isn’t to me like the biggest threat towards it. I can speak more about why I believe that the case, you know, a very common critique is like, well, if Persona becomes one of the core infrastructural businesses, doesn’t that put a huge target on your back?

And the quip I always say to that is, well, I think, look, we’re on a lot of these cloud platforms. So we’re a strict subset of how incredibly, you know, hacking a cloud platform is going to be infinitely more lucrative, right? Hacking Visa is going to be incredibly lucrative in that sense. I actually think that allows you to protect those platforms far better. Having actually this hyper decentralized, where at the same time, the data is the same on every one of these platforms just means that anywhere, you can break in any one of these systems. It’s effectively like, he’s the fault of anywhere. I don’t think that’s right.

But I do think actually what you call out there is this idea of like, what if there’s an over-centralized network of your identity is controlled by just a single private entity. Your existence online and wherever your digital footprint is owned in that sort of way. And I actually strongly am against that idea. One way in which I think we architect personas is explicitly against that vision. This idea, like, we think so much about GitLab and AWS and businesses like that because our perspective is persona doesn’t control your data. We should not be the ones, there shouldn’t be this third party that you go through to be yourself online.

And over time, you know, if they do become successful, there’s this like third party entity that effectively controls your identity and everyone. Our explicit perspective is that we want to offer the infrastructural layers such that the status quo continues to be that you, anytime you want to share your identity online, It is on the business that you choose, chosen to work with to make that decision. For them, you know, they are the controllers of your data. They continue to be the one. There’s almost this decentralization of who owns your data and who’s like managing it, who you’re establishing or trust a relationship with. And what we instead offer is the infrastructural parts for them to operate better and for your data to be managed securely.

In the same way that I think a lot of platforms have increased security in many ways because they’ve made it such that not everyone’s building out their own infrastructure. We want to be that enablement layer. But our goal is explicitly not one day to become the world in which we effectively own your digital footprint and everything’s operating through a third party. And towards some of the other businesses that you’ve brought up earlier, I think that’s a very explicit kind of difference in mentality. And our belief is that AWS type mentality towards the future of identity is better than, say, a single sign-on type of form of identity.

Take it from the sort of existential future and some of those risks that you’ve laid out, bring it in now to the very real present. I think most sort of end users are really only familiar with the identity plays of the largest companies, just based on things like sign in with Google or sign in with Amazon or Apple Pay or whatever. Sort of, you just see that surface where you just think like, oh, that’s convenient. Like I can click a button and log in with my Gmail everywhere I go or whatever. And then you start to think about the risk that that creates of like, like, okay, well, if your Gmail becomes compromised, that creates risks systemically throughout your entire digital identity.

How have you seen those companies? Are they legitimate sort of threats to you on the identity side? Are they are there companies, maybe one of them is making a bigger play than others? Or are they sort of just like, they just want to control those buttons as much as they can, but they’re not really making a play of it? Like, how are those guys playing in this space? More so than ever, I think there’s a fragmentation in the space of identity. Identity, I think, has been a very, like, kind of a, it’s a word that means a lot, is all I’ll say, right? Like, there’s a lot of ways in which people interpret identity.

And I think that the biggest fragmentation we’re seeing in identity today is an increasing kind of realization that the idea of a physical identity and the idea of a digital identity are fundamentally different from each other. And trying to piecemeal those together as, like, a single unified concept is not realistic. The first one that actually emerged was the idea of digital identity for online. The way I wish I always like to frame digital identity is that if I want someone else to be able to be myself online, I should have the option of freedom to do so. If I want someone else to log into my Reddit account, I should have the freedom to do so.

And another way I oftentimes like to frame it is that Reddit doesn’t care if I have five usernames. Reddit shouldn’t care. I can establish anytime I want a new identity online, I can do so. So when I think about like players like single sign on like your Google logins or Facebook logins. For Google login, the reality for your Gmail is that you can have as many Gmail accounts as you want. In fact, if I want you to have access to my Gmail, I should have that freedom to do so there’s no reason not to it’s an email address in that kind of perspective. And I think for that world of digital identity, it’s incredibly useful.

There’s a lot of services online, along the most influential services online in which a truly kind of like self identified identity is what they fundamentally care about. The one I used to give the analogy to was like, Netflix doesn’t care if you share your identity, right? They don’t care if you share your account, you know, why not? But I use Netflix as an example, because they have also pivoted towards this new emerging world of identity, in which I think it’s largely spurred on by the last 10 years, in which if you look 20 years back, let’s say 2000, 2010, we’re in an era in which it’s all about digitally native type businesses, in which, you know, you have social networks, in which every interaction occurs fully online, you have blogs, in which every interaction occurs fully online, all the value that’s provided to the information exchange occurs completely digitally.

The last 10 years, I think, has been a very interesting evolution of what if we instead think of not just like the value that the internet can provide purely digitally, but rather think about the internet from the perspective of how we can actually streamline things that occur in the physical world. So whether that be delivery, whether that be getting from point A to point B, whether that be finance, every single one of these things, whether that be getting a degree or education, all of these ideas that actually occur in the physical world. What about from that perspective? And it’s interesting because digital identity doesn’t apply nearly as well towards that physical world.

Because for example, if you are going to go get a degree, at the end of the day, it has to be you who took all those courses. I can’t choose to go and share my account with someone else and later get this degree because that’s impacting my physical world. If I want to order alcohol, you know, I am well over 21 now, but had I been under, I have to make sure that, you know, like the person who gets the delivery, the person who is, it isn’t just someone else’s identity that’s doing so. If I’m making a huge trade online or opening a bank account, it has to be my bank account.

I can’t just have someone else open it on my behalf. I can’t open up 10 of these and say like, they can be anyone online. There’s a physical impact on all of these. And identity from that perspective, this idea of this physical identity, I think is a very nascent idea of what does it mean to create an infrastructure. And I think that today, we consistently try to use the paradigms of how we think about digital identity applied towards it. And my perspective is that actually, it’s really kind of orthogonal. And actually, how we have to think about physical identity has to be more so from first principles.

I don’t think of actually like a lot of these single sign-on technologies as kind of like being more of a threat towards the world physical identity, because I think they miss a couple of the very key differences of why physical identity has historically been so challenging. ideas around if everything’s digital, then that means account sharing is very easy. And pure account sharing suddenly breaks the law of the constructs that have to be true for a physical identity. Ideas that ordering alcohol versus getting a million dollar loan from a risk perspective are wildly different from each other. And if you’re using Facebook sign in for both of those, that doesn’t make any sense.

The regulatory just aspects of, I always love to share fun facts about like crazy regulations, but technically in Massachusetts, if you want to get or buy alcohol, you have to have a Massachusetts ID. No one respects this law. I’m not saying anyone respects it, but this is a real law. If you go online, if you want to order alcohol in Massachusetts, you have to have a Massachusetts ID. Stuff like this, there’s all these regulatory things that complicate the physical identity world that trying to apply a pure digital approach towards it just doesn’t really work. So for us, we think a lot more about what does that new layer look like and how do we have to build it and free ourselves from the shackle of like, well, well, what if we just made it like a username, logit, password type of thing, or maybe even WebAuthn.

I think there’s ways in which WebAuthn can help. But at the same time, it isn’t just going to be like that. You have to know who the person behind the screen is. That’s so interesting. I don’t think that I have made that distinction well enough in the past, right? That sort of combination of the physical and the digital world and how they interplay with each other, which is really interesting. And it’s a good segue just in terms of like into my sort of final question, where we think about the world right now that we’re in, the sort of environment that we’re in. We talked, I think before we started recording, we were chatting a little bit about the sort of world we’re in for tech right now.

We’re sort of facing these macroeconomic pressures and all these different things and how people are making buying decisions and stuff. And we’ve kind of gone through now the journey from, you were talking about in the early days when you were raising your seed round, all the way now you’ve raised millions of dollars of funding. You’re employing a ton of people. You’ve grown into a really sizable business, a very different world. And we’ve talked about the future of what Persona could be and the environment that plays into. If you were to speak to the folks who are maybe earlier in their journey than you are, They’re still in that discovery phase.

They’re still chasing a problem, but they’re staring down the barrel of a very different sort of environment than you were when you were raising your seed round. What advice do you have for those folks as they strive to solve problems that they’ve maybe faced themselves similar to you, but they’re looking at all these different problems that we’re having and getting access to funding or acquiring customers or whatever. What advice would you give to those folks? More so the aspect of like self-discovery. Throughout this overall journey, the psychological and like the mental side of it all of like understanding yourself. And like in many ways, especially as founders, as the company grows, your beliefs, your perspective, your framework of how you view the world of fairness, of good, of, you know, like what’s right and wrong, that spreads.

Everyone else starts like, the culture that like implicitly is created from your core belief system is powerful. For myself, I actually do a lot of self-reflection. I mean, over these past five years of just like, what does all this mean? And like, what matters to me? And the advice I really have to give here is to not underestimate that part of like, how much understanding yourself applies to that overall journey of the business. And that the leap of faith towards going off and being a founder, it has to be more about the journey than it will ever be about the destination. I can share the story about how, oh, in the early days, you know, a lot of investors thought that what we were doing, you know, might be commoditized and there may not be an opportunity there.

But an alternative viewing of that is, I got really damn lucky that they were wrong. Like if the consensus is wrong, like that means you were really lucky, right? despite you know sometimes being uh going against the grain working out uh nine out ten times usually uh there’s the wisdom of the crowd so in this sort of situation we got really darn lucky that like maybe you know we were a little ahead of the curve of nine out ten you’d be wrong and i think uh the reason why that’s important is that you focus too much on the destination everything stops making sense it doesn’t make sense to go off and try to do it start it doesn’t make sense to you know go off and like try to build something from scratch there’s nothing about it makes sense from a pure like rational level you start thinking about opportunity cost of well why not go back you know like oh am i wasting away my career and all these things and instead i think actually what to make that leap of faith it’s a lot more about like embracing the aspect of the journey and embracing the self-discovery that comes along with that journey and i will say throughout this overall journey at the very least i’ve understood what i believe in my overall values what i think about more so than ever i understand that about myself better than ever.

And I think that is invaluable. In all respects, I do, I have really enjoyed this journey. I have really kind of found a lot of, I have a lot of gratitude for this overall kind of people I’ve met, including you, like all these folks I never would have crossed paths with in any kind of respect in my life. All the folks I have the pleasure to work with, all that I think is just invaluable. And that aspect of it, I think has made it all kind of, and I think very little about like where persona will wind up these days. On one hand, I think about the strategy and the future of identity so much.

On the other hand, I kind of just am like, it depends on where it takes us, right? But like, I don’t regret where we’ve gone to and I’m excited to see what’s to come. That’s all I can say. It’s like, don’t focus too much on the destination. Like if you’re going to make it, embrace the journey and embrace what you’ll learn from that journey as well. I love that. I think that’s such good advice, especially, I mean, that is the advice when things get hard, right? Is that it is appreciating the sort of struggle in the moment of it and appreciating what it means. So I love that advice.

Rick, thank you so much for having this conversation. It was so fun to get to dive into all these different areas. Definitely areas that I was not prepared, you know, questions and stuff that it’s so interesting to think about. So I appreciate you kind of taking me on the journey with you. Likewise, Kyle. Thank you so much for having me again. It’s been great discussing all of these. And I love how deep we were able to get on some of these topics and get past kind of, you know, I would say like the surface level of what identity means. I think it’s been really cool exploring these ideas with you.

Awesome, man. Thanks so much.

[00:45:10] Thanks for listening to Contrary Research Radio. If you want to hear or read more from us, you can subscribe on your favorite podcast player or visit us at research.contrary.com.

Connections

• Research Radio — Full Archive: the show.
• Persona and Rick Song: the company and its CEO.
• The Hits Business: the essay Kyle’s notes tie to the conversation.