Getting Creative at Scale

The brief period of extreme hype over NFTs mostly centered around static, collectible JPEGs, but there are actually far more use cases that can unlock dynamic functionality and provide long-term value.

The unique, programmable assets serve as carriers of original media, providing a home and distribution channel for content across the blockchain. This is possible thanks to the work of companies like Pinata, which has developed the infrastructure to support a wide range of NFT projects.

The Omaha-based startup, which was formed in 2018 and has been partnered with Greylock since 2021, began with the goal to provide a decentralized storage solution to make IPFS stable and reliable enough to store off-chain data at scale. Rather than thinking of NFTs as a permanent, collectible asset, Pinata co-founders Kyle Tut and Matt Ober believed their true functionality was as tools to store and exchange data that could be consumed. At the time, the company’s perspective on the utility of NFTs wasn’t common, and amid the crypto winter, initial traction was slow.

But seeing the potential for NFTs to serve as vehicles for creators to manage, share, and monetize dynamic content, Pinata persisted. Today, the company works with tens of thousands of users as well as leading marketplaces and platforms like OpenSea.

Tut joined me on Greymatter to discuss the company’s journey, how it works with customers today, and their vision for how NFTs can unlock a world of opportunities for creators across music, gaming, videos, and even websites. You can listen to the conversation at the link below or wherever you get your podcasts.

EPISODE TRANSCRIPT

Mike Duboe:
Hi, everyone. Today we are joined by Kyle Tut, CEO of Pinata, which is the home for NFT media. I’ll let Kyle explain what that is in the conversation.

We have been working with Pinata since the Seed and then the Series A. We’ve absolutely loved working with this team and are excited to get into some of the background of Pinata, broader thoughts around the NFT market and decentralized storage in general.

So Kyle, thanks for joining.

Kyle Tut:
Yeah, absolutely. Happy to be here.

MD:
Yeah. So a good place to start, Kyle, would be giving folks an overview of your background and what initially led you down the NFT rabbit hole.

KT:
Yeah, absolutely. So like many in the crypto space, I got into this because I had bought some Bitcoin back in the day and then ultimately ended up getting interested in Ethereum.

So in 2015, I bought Bitcoin, but I’m not really like a crypto finance trader guy. I’ve always wanted to build my own company. And so I was playing around with it, but I was really just in it to kind of figure out what the technology is. And so, Ethereum was really, really interesting to me with smart contracts in 2016. At the time, I was working at my first job out of college, which was at a motorsports startup where we were working with professional race teams around audio communications. So completely outside the realm of blockchain and crypto, but I was spending my nights and weekends learning about crypto and what it could do. And that led me to gaining enough confidence in April of 2017 to quit my job and jump into blockchain and crypto full-time.

At the time, I didn’t have a specific place I was going. I just knew I wanted to eventually build a company using this technology, and I was just going to go on a journey and figure out what that was going to be. And eventually, that ended up being Pinata.

MD:
Awesome. All right. So let’s get into Pinata. It started at ETHBerlin Hackathon. Tell us the backstory there and kind of what was the spark that actually led you down this path with Pinata.

KT:
Yeah, absolutely.

So as mentioned, I quit my job, and I’m actually based out of Omaha, Nebraska. And so, as you can imagine, there’s not a bunch of resources here for me to learn blockchain and crypto. So the first thing I ended up doing was jumping on a plane and flying to New York to compete at the Consensus Hackathon in 2017. And the reason I did that is I knew, as a non-technical founder, I needed to understand the technology at a much deeper level and have a really strong foundation with the technology. So I went to this hackathon, started meeting blockchain engineers, and obviously was sitting in rooms with them building. And I kept doing that all summer, just flying to different blockchain hackathons and kind of building out a network.

But I was still living in Omaha, so I wanted to find friends that I could hang out with here, so I ended up starting the blockchain meetup group in town that actually grew to like 500 people. The first person to reach out to me was actually my co-founder and CTO, Matt. He was a software engineer at a consulting company. And I was like, “Hey, you should come with me to these hackathons. This crypto thing is real. There’s a lot of excitement around it.” And the first hackathon we went to together was actually ETHWaterloo. And ETHWaterloo is where the CryptoKitties NFT application ended up launching.

So Matt and I have always had NFTs kind of in our story, and they’ve always been a big part of our journey. But up to that point, we were just going to hackathons. And we had started a consulting company where we’re building blockchain applications for various people, but hadn’t started Pinata yet.

But through that experience is how we were figuring out that storing data on-chain was exponentially expensive. So I think at the time in 2018, to store one gigabyte of data on Ethereum, it costs something like four and a half million dollars. And so what everybody was doing was using a protocol called IPFS (or the InterPlanetary File System) to actually store and manage their data off-chain. The problem back then was it wasn’t very fast, and it wasn’t very stable. And so Matt and I had the simple idea of let’s just make IPFS as fast and stable as possible.

And so we came up with that idea, ended up launching the idea and Pinata ultimately at the ETHBerlin Hackathon in the fall of 2018. We ended up winning that hackathon, and then we thought we were going to take over the world, raise a bunch of money, and everybody was going to use us.

Unfortunately, there were two factors going on where that wasn’t true. And we really ended up struggling for close to two years, because, number one, NFTs weren’t a big thing yet. The NFT community back in 2018 was all of six people at the time. And the other thing was it was a crypto winter, so we were struggling with people moving out of the space. There wasn’t a lot of hype and so we just weren’t able to get a bunch of users. And we effectively had to just keep going, keep calling people, keep trying to find use cases for us until, in 2020, we started to see the NFT market start to take off, and then our success followed.

MD:
Yeah. I think when we and Nate and Offline Ventures found you, it was March of 2021, when some of these NFT projects were happening.

KT:
February 28th. I remember.

MD:
Credit to Nate on that one. But we were starting to see these NFT projects and marketplaces start to pop, and so we scoured through docs to look at what infrastructure was being used, and Pinata kept coming up. It felt like there was almost some ubiquity on NFT projects at the time [specifically] of Pinata within NFT projects at the time.

But as you mentioned, you had braved through basically a few years of there being really not a whole lot of usage. I’m curious in your mind and in the broader team, what gave you that conviction over that time? Because it feels like a pretty challenging space to keep building within, with adoption so limited.

KT:
Yeah. So our conviction was around the idea that we knew storing data on-chain at scale was just never going to be something that was going to work. Blockchains are not good at storing data at scale. And so, we just really needed some use-case in the blockchain space to happen where they needed to store more data off-chain, and that obviously ended up being NFTs. But 2018, 2019, it was mostly DeFi applications, ERC20 tokens that aren’t storing a bunch of data off-chain. And so we really just needed the market to come to us.

I always talk about it as if we were right. But when you’re right, and you’re early, it’s the same thing as being wrong. And we were wrong for close to two years. But if we weren’t there early, if we weren’t going to those hackathons…So 2018, 2019, 2020, we kept going to hackathons. We kept building our brand in the dev community by going to those hackathons. And then I spent every day just calling as many projects as I possibly could, talking to them about why would they use Pinata. Why would they use IPFS or not, and what projects were they working on? And then, ultimately, through all of that, we were also blogging, and our technical blog was teaching people how to build in this space, how to build NFTs. And that’s ultimately where we started to see traction.

So we were very committed to IPFS. I think a lot of people questioned that back then, but it turned out well for us. And it was kind of in the early days, we would see one person using us every day, and then a week later, it would turn into two and then four, and it would go on from there. And that kept us active and kept chasing what we were going after.

MD:
Yeah. Before moving onto some of your perspectives on the future of NFTs and building through crypto winters, I want to spend a moment on IPFS. It’s a concept I’m not sure how many listeners will be deeply familiar with – I think those building in web3 understand it – but talk a little bit more about why you made the choice to actually align yourself so closely with IPFS.

KT:
Yeah, absolutely. So we kind of have a unique view on IPFS, I think compared to most people. If they’re familiar with it, I think most people think of IPFS as this peer-to-peer protocol, and it’s a distributed/decentralized protocol. What we’re most interested in with IPFS is that it’s a content-addressable system. So what that means is when you upload content to IPFS, it generates what’s called a CID or a content identifier. And those identifiers are unique to the file itself. And so if you have a file and it changes just a little bit, that CID ends up changing. And when you combine that with an NFT or a blockchain, you get this really nice append-only or immutable record of that file and what’s happened over time.

And so for us, what that ultimately means is, from an accounting perspective – this is kind of boring – but from an accounting perspective, this is actually really, really powerful at scale. And it’s really, really powerful when you’re transacting between two different parties on an open-pseudo anonymous blockchain like Ethereum, or pick any of them. And it makes you be able to trust the data itself instead of who’s holding it. So you don’t have to trust that the person that you just grabbed that file from was doing something malicious or they changed the data in some way. You can just trust that CID is what it’s supposed to be. And that’s kind of the core thing that we get excited about.

Now, as it’s related to things like decentralization, IPFS, at the end of the day, can be as decentralized or centralized as you want to make it, which we think is actually an advantage. And the other kind of component where in this conversation that people bring up is around permanence and keeping data around for a long time. And IPFS itself doesn’t necessarily keep data forever. You have to make sure that you pin it to an IPFS node to keep it up. And that’s ultimately what people use us for. They pin the content on our IPFS nodes, and we ensure that they stay up and can be served and distributed appropriately. But as it relates to permanence around data and some of the other decentralized storage protocols: we ran into this question a lot every time we were talking with teams is they’re like, “Hey, we need this stuff to be permanently associated with NFTs.” And we always thought that was kind of a weird perspective.

We actually think it’s interesting to have NFTs that last for a second or milliseconds. And we never actually approached NFTs in the same way that a lot of the market did, which was thinking of them as art that needs to last forever or as collectibles. We just thought NFTs were a tool, and you need to be able to make sure that the data is there, and then when the NFTs go away, the data is no longer there. So permanence has never been a big factor in how we approach data and, ultimately, NFTs.

MD:
Yeah, you got into one of the topics I was hoping to address, which is around just permanence is a topic that’s coming up more right now. And to some extent, Pinata is a layer on top of whether it be IPFS or are we even eventually, you could be agnostic to this.

Let’s double click on this just a little bit further. For projects that are considering decentralized storage solutions, where will (and won’t) permanence matter? If there are certain applications where you actually think permanent is the right solution, I’d be curious for some examples of what you think those could be.

KT:
Yeah, absolutely. So this gets a little bit deeper into whether or not NFTs are actually assets or not. And so I love this topic, and internally at Pinata, we developed this framework where we put NFTs on a spectrum from consumables to assets. And consumables ultimately go to zero, and their value ends up going to zero. And there’s a finite amount of usage that you can get out of a consumable.

The way I like to describe it is kind of thinking of the spectrum as a coffee shop. So with a coffee shop, there’s the building it’s sitting in. That’s an asset. There is the LLC that the coffee shop is run by. That’s an asset. But then there’s the cup of coffee itself, and that’s a consumable. And you can buy that cup of coffee, and you own it once you pay five bucks for it. But once you drink that coffee, ultimately, the value goes to zero.

And so when you start framing it that way, that not all NFTs are assets – and then not all NFTs have to go up in price – ultimately, that means that not all NFTs have to last forever. And I would actually make the argument that the majority of NFTs don’t have to, or are consumables and have a finite life.

The other thing I think that you really have to kind of peel back is, in the early days of NFTs, there was a narrative going around that if an NFT lasts forever, it’s more valuable. And we, again, always thought that was silly in the sense that just because a rock lasts forever doesn’t mean it’s necessarily more valuable. And so there are just some threads (maybe driven by Twitter rhetoric) that have driven this idea that NFTs need to last forever.

The other interesting thing that we started to see is people started burning NFTs. So they would just start basically deleting them or sending them to a burn account, and then they would disappear. And we started seeing that very early on outside of the art and collectible NFT market, but more of the permissioning market is kind of what I would call it.

And so we’ve always kind of approached this as, at the end of the day, NFTs are just tools, and they’re really good at exchanging data, moving data being used as permissions, but that doesn’t ultimately mean that they have to last forever. And we think that, ultimately, there’s a lot of good use cases where they won’t.

I ended up writing a blog back in 2020 called Who Is Responsible for NFT Data? And I posed the question, is DaVinci responsible for the Mona Lisa today? And obviously, the answer to that is no. DaVinci is no longer with us, and the Louvre Museum is ultimately the one care of the NFT. And the point of that is that if something’s important enough to keep around for a long time, people are going to ensure that it stays around. And with IPFS and that CID component I was talking about, it’s actually very, very easy to transfer responsibility of an NFT to the next person. So when you buy an NFT from somebody else with IPFS and CIDs, ultimately, you can ensure that it stays live and is around for as long as you need it to be.

MD:
Great. So I think a lot of people rightfully look at Pinata as a pinning solution on top of IPFS or broadly related to storage. If you look at Pinata’s vision early-on versus where the product’s going right now, what are some examples that illustrate where you think the opportunity is? And what does that bodes for NFTs (more broadly) going forward?

KT:
Yeah, absolutely. So when we first started, we were just, as you mentioned, just an IPFS pinning service, which is really no different than just being data storage at the end of the day, but we were doing it on top of IPFS. Where NFTs took us and the needs of the market essentially is we started to realize that we had to get much more sophisticated around media and ensure that we could distribute content at scale.

So effectively, what we started to see is that these NFT projects were their own little media hubs or media brands, if you will. And they needed web 2.0 speed and web 2.0 scale associated with whether it’s images or videos or whatever the content is ultimately that these NFTs were, they need to be able to distribute that to millions of people. And so we had to start building functionality and scale so that these NFT projects, typically they start as a team of one or maybe three people, have the ability to go from nothing to millions of people viewing their content overnight.

And it kind of brings up this interesting question of, if you can’t sell a YouTube link as an NFT, then where are you ultimately going to distribute your content at? Or if you can’t use Dropbox to distribute your content, where are you going to do it? And that answer ultimately ends up being Pinata for all of these projects.

So that’s kind of on the media side and the scale side. Where we then started to see the market going is that, ultimately, private content and IP control became extremely important.

So in the early days, it was like all of this data was publicly associated with NFTs, but what we’ve seen is it’s now shifted to a significant portion of the market using NFTs for this concept called token gating. Token gating is a simple idea of where you only serve or distribute the content based on whether or not you own the NFT. And so what happens is you basically give somebody content, it checks whether or not they own the token or the NFT in their wallet. If they do, they get access to the content. If they don’t, then they’re not able to view the content.

And so we started to build out a feature called Submarine that allows NFT projects and marketplaces to easily token gate their content and serve content based on NFTs. Where that eventually kind of iterated too is this product called submarine.me, where we make it really, really easy for non-technical people to actually come in, token gate whatever type of content they want. They can token gate it across multiple chains and lock it down to whether it’s at the NFT. If it’s 10,000 NFTs, they can do it for all 10,000, or they can actually lock it down to the individual NFT itself.

And we think ultimately what that’s doing, in the grand scheme of things, is it allows creators and these NFT projects to build their own business models based on how they’re locking down this content with NFTs. And it’s really using these NFTs almost as a permissioning mechanism. So we’re really excited to see what kind of business models can come out of that.

MD:
What are some of your favorite examples so far of projects that are using maybe some of the use cases with more advanced functionality, if you will?

KT:
Yeah, absolutely. So the one other thing we saw is that people started actually uploading full applications to us. And so they were using NFTs, token gated NFTs, to serve full applications like you’d have on your phone or even on your desktop. And so we’re actually seeing people use this functionality to token gate games is a big one.

But we’re also seeing people token gate music where they’re actually serving not just the file itself, but they’re actually token gating, basically, a full experience around the song. So it has a video, and it kind of has a brand wrapped around this music player that’s a full application that is ultimately being served through being attached through to an NFT.

MD:
Yeah. If you think about the broader NFT market in a three to five-year time horizon, what do you think looks most different about it than today?

KT:
So when Matt and I went to ETHWaterloo and saw CryptoKitties launch, the first thing I did is I came back to Omaha, and I was like, “All right. How do I apply this to something that I could describe to somebody in Omaha?” And so the first thing I did is I wrote a blog in December of 2017 where I applied NFTs to farming. So I think it was like a use case for ERC-721 tokens crypto farming, or something to that effect. And what I did is I created an NFT that represented an acre of land, and you were just using combine data – the big tractors going through corn fields. They have a bunch of data flowing off them, and I was taking the data from those combines and attaching them to the NFT themselves.

I initially thought that the use case for NFTs was going to be for transacting data between parties and not media or gaming or anything like that. I was trying to apply it to IoT use cases, enterprise use cases, those types of things. And so at Pinata, the way we’ve always been thinking about it is what we just went through in the kind of mania of NFTs is obviously a use case.

But we think ultimately that NFTs and IPFS can be used for all data and can be used to transact data. It can be used to permission data. And we think ultimately that all data is going to be attached to NFTs and ultimately uploaded to IPFS. And so where we think it’s going in the future is, I kind of already talked about this, but full applications uploaded to and attached to NFTs.

As I mentioned, I do think IoT data is going to be transacted by being attached to NFTs. I think algorithms and anything you can think of essentially that is a file or a folder can ultimately be attached to an NFT, and it can be transacted. And that’s where we get really excited. And as it’s related to kind of the NFT market itself and the way that we approached it, the unique thing about Pinata is we’ve never actually minted NFTs. We’ve been kind of hands-off in that regard, and we did that on purpose.

And the reason for that is, ultimately, we didn’t have a good grasp on number one, which blockchain or blockchains was going to win. So we wanted to be kind of horizontally indexed across the whole market. And then we also wanted to ensure that NFT innovation at the token level could happen without us influencing it.

We are seeing NFT projects with a bunch of great kind of innovations around the token itself, and we think there’s still a bunch of room for that, and we can just plug into that and not kind of interfere with that. So yeah, we ultimately think NFTs have a long lifespan and are going to touch everything, but there was a little bit of silliness in the market over the last year.

MD:
Yeah. What advice might you have for other founders who are actually building through this winter who maybe have not before? I know you’re kind of a bit of a grizzled veteran here, all considered. So how would you address them?

KT:
Yeah. So in these downturns, I actually think from a building perspective, it’s easier because there’s much less noise, right. And so the thing that I always talked about with my product team during the mania was like, “Don’t follow behaviors that are happening because NFTs are really expensive, or they’re being traded, or there’s a lot of speculation around them.

And so when you’re in the hype bubble in the mania, you can accidentally build the product in the wrong direction because you’re following a behavior that is not going to last over time. And so that’s what I talk about with my team is like, “Let’s make sure that when we’re building a new product or feature, especially during that hype cycle, was to ensure that we felt that that behavior was going to last regardless of whether or not that NFT was worth $1 or a million dollars.”

As you turn into the downturn, a lot of that noise goes away, and it’s actually much clearer, I think, to be able to build, and you can have conversations. Your voice carries farther. You can actually pick up the phone and talk with people. It’s not so busy. I think right now is actually the greatest time to be entering this space. And it’s where, ultimately, Pinata built itself back in 2018 and 2019 was in those downturn.

And ultimately, what that looked like is we were calling people, we were going to hackathons, and we were blogging about our experience, and that’s where we were able to build our brand, become trusted. Right now is a time where people will trust you more because you’re in it, not for the hype. You’re in it for better reasons. And so yeah, I think it’s a great time to enter.

MD:
Yeah. I mean, this brings up a related topic actually. And listeners might not know, but Kyle’s based out of Omaha, and you’ve been building the team kind of in, I guess, what one might call a non-core market, which I think can be perhaps focusing, but also comes with its own kind of challenges. What have you learned in that journey and building a startup out of Omaha, and what advice do you have to founders building outside of what one might call core markets?

KT:
Yeah, absolutely. So I have a person in Omaha who everybody obviously knows as Warren Buffet. He has a really good insight that I found out myself. And then I read in a book that he actually found out about it in 1956. And the insight is this: Warren Buffet used to actually be in Wall Street, and he was building or creating his career in Wall Street. And ultimately, what he ended up moving back to Omaha for was this concept that in Wall Street, he was just listening to all the rumors, and he needed an ability to go somewhere where he could think clearly without social validation or without external kind of hype influencing his decision-making. And so he ultimately, in 1956, ended up moving back to Omaha because it gave him that space and kind of isolated him from the mania and from the hype.

And so I’ve leveraged that a lot being in Omaha because you are isolated. Obviously, we have the internet, and we have Twitter, and I can read everything everybody else is reading on Twitter as well. But I’m able to just kind of focus on the core idea, and I have the space and room to really think about these ideas at a deep level.

Now that part is great. The part that’s tough (and I figured out how to solve for) was that we don’t have the best information. There is amazing information when you fly into San Francisco or Silicon Valley, and I always talked about it as updating my cash on what is the hottest topic. What are the new trends? Who’s building what, and why are they building it? Those things are really, really important, and I always try to make it back so that I can update my cash. And then I’ll fly back to Omaha, think it through, think about is that behavior going to happen for a long time or is it kind of a short-term behavior that is happening.

So I think it can be a huge advantage if you use it. You also have to build towards the strength of whatever city you’re in. Not all cities are going to be the same. With Omaha and Nebraska in general, it’s a very supportive community. They might not necessarily know what I’m doing. I couldn’t quite describe it. But they do completely support me, and I use that to my advantage 100%. Whereas if I was in a bigger city or something like that, maybe I wouldn’t have had all the opportunities that I’ve had here.

MD:
Yeah. These are phenomenal insights.

Following along on this thread of unique aspects of building a company in your space during these times with these dynamics, if you look at what a modern web3 technologist might need to be good at, that’s different from web2 more broadly, I’d be curious what advice you would add for others.

And just to comment, I think Pinata, from the outside, seems to have done an exceptional job of having a mix of talent from different backgrounds and also obsessing over this core problem. You mentioned Submarine coming out of an internal geek-out session, and you have the connectivity with a bunch of creators to be able to test that concept out and see it through.So it feels like there is a mix there in terms of background and just kind of perspectives. What advice might you give to a founder starting in web3 right now who may or may not be from the background as they think about composition of team?

KT:
Yeah. So not only are we based out of Omaha, so my co-founder and I are in Omaha. Matt is here as well. But we are a fully distributed team and we leaned completely into that. Part of that is because a lot of our success happened during COVID. But the other reason I did that is because it’s a very strong way to build diversity of thought in your team.

We’re a team of 45 today. We’re in 20 different states, and eight or nine different countries is where our talent base is coming from. And ultimately, what that gives you is a bunch of data collection people who have a different perspective on the world and are thinking in a different way and are able to approach problems in unique ways that other people aren’t. And we try to lean into that as much as possible.

And when I think of web3 and decentralization, I think that the core tenant of it is how do you develop different thought processes and lean into that? And I think if I was going to say anything, it’s to lean into the uniqueness of web3. It is a different way of thinking compared to web2, and use that to your advantage. That, ultimately, is where all the fun is. I always talk about it as, with web3, you can’t Google the answers. You have to come up with them yourself, and that’s the most fun part.

The other thing at Pinata we talk about a lot is we’re not going to beat Google at their own game. Pick any web2 company, and we’re not going to beat them at their own game. We have to play our own game. And so we talk about this concept of counterculture and how do we build a game that we know how to play that other people have to play with us versus playing theirs. And that’s ultimately where things like margin come from, and that’s where innovation is going to be.

MD:
Yeah. Well, Kyle, it’s been an awesome conversation. Before we wrap, are there any shout outs you want to give to anyone who’s curious about Pinata or any folks out there who might be interested in the space more broadly?

KT:
Yeah, absolutely. So you can find Pinata at pinata.cloud, and then I’m on Twitter at @kyletut. And in the space, I think, ultimately, you just have to be curious right now and make sure that you go out and learn as much as you can. A great way to do that would be to check out our technical blog on Medium and start building.

MD:
Sweet. Thanks, Kyle. It’s been a great chat and honor to work with you, you as always. Thanks for coming on.

KT:
Absolutely. Thanks, Mike.