Nathan Barry, founder of ConvertKit, joins Sam and Shaan to walk through his “Billion Dollar Creator” framework — the idea that the most profitable thing any creator can do is launch their own product rather than sell attention to brands. They cover the ladders of wealth creation, Conor McGregor’s business empire, why communities are hard to scale, the WealthX data business, a hot girl side hustle in UGC ads, and the coming wave of tele-operated heavy machinery.
Speakers: Sam Parr (host), Shaan Puri (host), Nathan Barry (guest, founder of ConvertKit)
Opening: Nathan’s Airbnb House and ConvertKit Revenue [00:00:00]
Sam: You got to choose the business model that’s right for you. But answer the question: can this audience I have make me a billionaire?
Shaan: I have one house that’s going to pull off like $15,000 a month and it’s just a single house. You’ll beat it by quite a bit.
Nathan: No way. Really?
Shaan: What’s the house worth?
Nathan: I bought it for $900K.
Shaan: $700 a night? Where is it?
Nathan: It’s in Boise, right next to the Boise State University stadium.
Sam: Oh, so you get like football tourists? Who’s renting this thing for $700 a night in Boise?
Nathan: We’re live by the way — Nathan, we’re live.
Nathan: Great. So you get all the graduation traffic, everyone coming in for the university, and then just anyone coming to downtown. It’s a five-bedroom brand new construction.
Sam: It actually looks like the kitchen is similar to the place you just linked me to. Why haven’t you bought more? A million-dollar house making $15K a month — why don’t you buy 50 of them?
Nathan: I just bought this one. It’s not making that yet, but the bookings are coming in. I did $8K in January, and the summer bookings are coming in at $800, $900, even $1,000 a night. I just bought a hot tub today, so that’ll help.
Shaan: Nice. So let’s intro Nathan. I’ve known Nathan since a Gumroad meetup — Shaan, I went to a Gumroad meetup, I’m not joking. Maybe 2012, 2013. Nathan was doing Gumroad, he had this book called Authority, a cool blog, sold all types of stuff online.
Then he launched this thing called ConvertKit and I was like, Nathan, the other thing’s working, just do that. Anyway, now it’s a business that Sam’s in, I think both of us invested — wait, actually I didn’t end up in it, there’s some mechanics problem with us being an LLC and the pass-through stuff gets complicated.
Sam: But I’m in the cool club.
Shaan: And basically his business ConvertKit does — what is it now? About $29 million in recurring revenue?
Nathan: Yeah, about that.
Shaan: And the revenue’s all completely live. If you Google “ConvertKit revenue” — it’s like a Baremetrics public dashboard.
Nathan: Yeah, it’s ConvertKit.Baremetrics.com. We’ve had it public from the beginning.
Sam: Do you regret that your business is actually kicking ass and you’re still totally transparent? You’re giving more value than you’re getting back at this point.
Nathan: I think it’s great if it’s for a mission reason and a terrible idea if it’s purely for marketing. I actually had two friends sit me down at one point and say, Nathan, this is an intervention — take down the public metrics page. And they were right — they’d talked to competitors who said wow, this is so helpful. But really, it’s a mission thing. The whole mission for the company is to help creators earn a living. If we can put this public blueprint out there so someone at $50K MRR can look back and see what our churn was at the same point — that’s the point.
Shaan: Okay, now I believe you’re not using it for marketing. If you don’t even know how much traffic the page gets, that tells me it’s real.
The Billion Dollar Creator Framework [00:12:00]
Shaan: So Nathan, you’ve got this blog post I want to walk through. Let’s do the Billion Dollar Creator one.
Nathan: So the article asks: what is the most profitable place to direct attention? Everything we do — podcasting, TikTok, being a movie star — you have attention and brands want it. They’ll sponsor you. The question is: what’s the most profitable thing to do with that attention long-term?
The answer is: create your own product and drive attention to something where you actually build equity, rather than selling ads.
Take Jessica Alba. She’s made a lot of money from movies, but the bulk of her wealth is from starting The Honest Company — being the spokesperson for her own business. Or Ryan Reynolds — he was doing ads for other people, probably getting paid a million bucks here, $10 million there. At some point he goes, forget that, I’m going to buy my own companies. Mint Mobile and Aviation Gin. He doesn’t get cash, he gets equity.
That’s actually my hypothesis with ConvertKit. I have attention on the internet through my blog and newsletter. How do I want to monetize it? Sponsorships, ebooks, membership — no. I want to monetize it through ConvertKit, a SaaS company. That’s my version of the Billion Dollar Creator.
Sam: Rule number one is you have to build more than a personal brand. You give the example of Jessica Alba, and also Mark Sisson from Primal Kitchen.
Nathan: Right. Mark’s blog is called Mark’s Daily Apple — the leading paleo health blog. A blog like that, doing this in 2006 to 2010, you could make a million bucks a year. But that’s it — it’s all about him and his name. What he did was start Primal Kitchen, using the audience to kick off these paleo-friendly ketchup and mayonnaise products. His email list had maybe 100,000 people — not huge by today’s standards, but trusted. He sells it to Kraft for $200 million. And he still owns the audience — he can do the next thing from it.
Kylie Cosmetics is probably the biggest example most people know. Instead of taking $250K per Instagram post or $2 million per endorsement — whatever that number got to — Kylie goes, forget that, I’m going to launch my own skincare line. Now it’s a billion-dollar brand.
Kim Kardashian has Skims. Kanye has Yeezy. They all turned to say: whoever was most willing to pay me to advertise their product is now my competitor, and I’m going to launch my own brand.
Shaan: There’s a guy in the NBA who gets made fun of for this — LaVar Ball. He has three sons all trying to make it to the NBA. Nike and Adidas both offered the oldest one contracts, and instead LaVar created Big Baller Brand. The shoes kind of sucked, and the execution was poor. But actually, the move was right. The youngest son — the one who was kind of the wild card — turned out to be the star player. If they’d built it properly around him, it probably would have worked.
Nathan: The trend between all of these people — that’s rule number two: sell products, not attention. And Conor McGregor is the extreme example.
Shaan: So Conor — in the UFC, fighter pay is brutal. The percentage of revenue going to fighters is like 15 to 20%, compared to 50% in the NBA and NFL. So fighters go out there, get their face beaten in, make $20K to $80K, maybe do it two or three times a year.
What Conor did was instead of selling the attention, he created a brand around every part of his lifestyle. People like his suits at press conferences — so he launches a suit brand. He’s Irish, so he launches an Irish whiskey. He just sold that for reportedly around $100 million. He’s super fit — McGregor FAST, the workout program. Recovery spray. He’s thinking about a sports betting exchange — wait, should I take DraftKings’s money or should I launch McGregor Kings? He just opened a bar. Used that bar as the backdrop to film him creating a stout — now he’s selling a stout brand.
The guy is just selling every piece of his lifestyle as an independent brand. Fighting is going to be the lowest part of his income stream. My bet is Conor McGregor becomes a billionaire from this.
Nathan: Rule number three: drive higher customer value through recurring or repeat purchases. You’d rather have a product people buy many times versus a one-off. The most valuable brands sell products someone buys again and again. Primal Kitchen — you go through ketchup every month. That consumable is worth more than something that lasts forever.
Nathan: Rule number four: choose a better business model. Vani Hari, the Food Babe, had an information products business — cookbooks, meal plans. Then she teamed up with Derek Halpern and made a health supplements company. Completely different model. Andrew Wilkinson is another great example — he used his cash-flow-positive agency Meta Lab as a vehicle to go buy high-quality recurring revenue SaaS businesses. He was trading medium-quality revenue for high-multiple revenue.
Shaan: Three people I think left money on the table. Ramit Sethi — huge audience in personal finance, advocating for low-fee index fund investing. He could have built Wealthfront, which just got acquired for $1.4 billion. Instead of selling courses, he could have built the software product.
Tim Ferriss — huge trusted audience, said he tried every supplement and here are my issues with them. He could have done what Joe Rogan did with Onnit. I think it’s a $500 million business given the size and trust of his audience.
Nathan: And Marie Forleo builds everything around her name instead of brands that can be dissociated from her. The product is her.
Sam: But Sean, you’re doing the exact same thing. You’ve got this big audience and you’re building this other thing that people are begging to know about.
Shaan: Totally. That’s why I love this article. I did it with the fund. Podcast ad revenue — first year I made maybe $70K. But the value of that audience felt bigger than the ad revenue. So I said, what if I raise a fund from people who’ve been listening? No pitch meetings, no presentations. Just: if you trust me and want to invest alongside me, you can.
Now that fund is an $8 million a year investment vehicle. Two percent management fee is not much, but 20% carry means I’m getting $1.6 million of startup equity per year for free. And if those startups 5x in seven to ten years, I’m getting something like $5 million a year of value out of something created from the podcast.
The new thing I’m building — the Milk Road — isn’t called Shaan’s Crypto Newsletter. It has its own brand name, staff, and can be something big. That’s version of following the framework.
The Ladders of Wealth [00:30:00]
Shaan: Let’s go to the Ladders of Wealth because this is the other nice framework you have.
Nathan: So you have this image of a series of ladders. It’s not one ladder you go straight up — it’s that if you choose this method of wealth creation, your ladder can be this high. If you choose this other method, it can go one rung higher or four rungs higher.
The smallest ladder: selling time for money. That’s having a job.
Next: your own service business. Agency, law firm, consultant — you charge by the project or hour.
Next: productized services. Instead of charging $100 an hour to rewrite copy, you say: for $1,000 flat, I will rewrite your landing page. A few important things happen there. One, the purchase is made without you talking to the customer. Two, it’s pre-packaged — they know exactly what they’re buying. Three, time and money are decoupled — if you get better or hire someone, you don’t have to do it yourself.
Jack Butcher, who’s our friend, has a course called Build Once Sell Twice. Such a good name. You build the thing once, then sell it as many times as you want. The upside is incredible. The downside nobody talks about is how hard it is to build the first one.
Next up: products — digital or physical. Then SaaS and recurring revenue. And at the top: marketplaces and social networks, which require the most skills to pull off.
Sam: What’s interesting about this is that it kind of explains why you — as long as I’ve known you, since we were both like 23 — you’ve always been incredibly patient. You’ve also done a bunch of book launches and other things that have sucked.
Nathan: Yeah, he had this great Twitter thread — not “sucked,” but something like nine things I’ve created that flopped.
Nathan: That’s the whole point of this article. Business and wealth creation is the combination of a thousand little skills. If you try to do it all at once, you’ll fail. The ladder forces you to learn sequentially.
I think about it with the strip malls versus skyscrapers analogy. You have a piece of land and you want to make money quickly. So you build a strip mall — a little office building, add a Subway, add a Radio Shack. Each one makes a little cash but you’re expanding horizontally, using up your land. The blogger who has an ebook, a membership site, a podcast, a course — that’s strip malls.
My approach has always been the skyscraper model: do one thing and keep pouring everything into it. Just keep going taller.
Shaan: Though you haven’t always done that, right? You’ve done book launches, paid newsletters, real estate.
Nathan: True. And I think those are okay as creative outlets that help you focus on the main thing. My friend Sean Blanc says: if you work with your mind, you should rest with your hands. I like that idea. Doing an Airbnb house or a one-off paid newsletter can be a contrast that makes you better at the main thing.
Sam: What opportunities or problems have you discovered that you can’t pursue right now — things you’d love to see someone else build?
Nathan: The honest answer is I’d go start a Gumroad competitor. And in a way I already did — ConvertKit Commerce. The thing I always said when asked what I’d do if not this: I’d start a Gumroad competitor. Because we have the email side, and if we add commerce, now creators are paying us while also paying us. It changes the churn dynamics completely.
Gumroad by itself is not a great business model at 4% of a $10 PDF. But pair it with email marketing and you have a creator platform — that’s a fantastic business model.
ConvertKit Operations and Patient Building [00:45:00]
Sam: As long as I’ve known you, you’ve been patient and long-term focused. How?
Nathan: Part of it is personality — patient and relentless probably describe me. But also, I think about the skyscraper model. I’m just going to do one thing and keep making it taller.
Are you still hands-on at ConvertKit?
Nathan: Way too hands-on. The thing people don’t tell you in startups is that the growth you see is a whole bunch of stacked S-curves. You figure out things that work, it takes off, then it levels out and you have to figure out the next stage. I’m completely into recruiting new executives, working with the team to design the next version of the product. We’ve had some executives leave over the last year, so I’m very in the weeds.
Sam: I don’t know if you publicize this, but I’m on this email list where Nathan sends out his entire net worth. You can see how much cash he has, which stocks he owns, his angel investments — all of his finances.
Nathan: It’s all the content I wanted other people to publish. Sam talks about Fat FIRE and that fantastic subreddit — I just want more of those conversations. I write a lot about how to get from zero to $100K online, but once you get to $200K or more, people don’t talk about what you do with your money. So I made a private newsletter, charge $100 just to filter out anyone who’s going to say I’m bragging. There are only about 200 people on it.
Shaan: The way I’d describe your blog or email is: it’s kind of like face tattoos and cornrows. I think it’s super cool when other people have them, but I don’t want it myself.
Friends With Benefits and Social Tokens [00:52:00]
Sam: I want to talk about Friends With Benefits — the first time I’ve seen something in this space that actually interests me.
Shaan: Okay, so what is this? Friends With Benefits is a community, basically a membership club created by this guy in LA who was a co-founder of Brud — they make Lil Miquela, that Instagram influencer that’s a digital avatar. He creates Friends With Benefits — basically a digital Soho House. Invite-only, and you buy the FWB token to enter. You buy it, you get access to the club, which is basically a Discord group. No physical place to go, no perks or discounts. Just access to the Discord.
People were buying in for like 75 tokens at the beginning, which was maybe $100. Now that token has appreciated like crazy. The membership is roughly $7K now, and when I talked about this last episode and we waited a week, the token went up 46%.
Just to put numbers around it: Equinox, the high-end gym — the membership is maybe $2,400 a year. This Discord is now double that. And this is a $90 million market cap community.
Sam: Obviously I’m a little skeptical. It’s like, okay, cool, you’re just going to run this like an art project? Either it’s a fad — exclusive things get less cool when you try to grow them — or are you doing this because you want to be a cool guy or because it’s actually a solid business?
Shaan: But here’s what’s interesting to me: these are called social tokens. The concept is any creator, any group of creators, could create one of these. Instead of renting your spot in the club through a monthly fee, you own your spot. There’s a limited number of spots. That’s an asset you could flip to someone else.
Everyone is aligned. Instead of a $9.99 per month expense, it’s like I paid $900 and I own this thing. If my bet was right that this podcast is going to get bigger — well, it did. If there were only 1,000 spots in our private group where we’re sharing insights, making introductions, doing deals — that could be pretty valuable. And those 1,000 seats appreciate as we get bigger.
Nathan: I actually do think this model is really interesting. People are incentivized to stay and make the community valuable because their token goes up. That’s a great incentive.
Shaan: Let me put something in perspective. Soho House went public recently — $3 billion market cap. 100,000 members, $2,000 to $3,000 to join, but they’ve got $870 million of debt, they furloughed 90% of their workforce during COVID. Churned only 10% of members, which shows the loyalty.
My point: if you were going to do a Soho House model using this token mechanic, you could build something big. The token incentivizes members to make the club great. I would actually 100% do this with a physical space model — use the token money to buy buildings.
Sam: The only reason you wouldn’t do it is if you don’t want more work. Because as soon as someone becomes a member of your club, your personality means you will not sleep until they feel like they’re getting 10x value.
Shaan: That’s exactly right. The other interesting thing: communities of communities are great businesses. Discord and Reddit are both like $10 billion companies. So ColabLand — the platform that runs these token-gated communities — if I was going to put money anywhere, I’d put it in ColabLand, not Friends With Benefits.
WealthX Teardown: Data Wants to Be Free [01:03:00]
Shaan: Can I tell you about one more interesting thing related to rich people?
Sam: Yes.
Shaan: Have you heard of WealthX?
Sam: No.
Shaan: WealthX is a database company. They use publicly available records — property records, tax returns, donation records — to find people who are ultra high net worth, defined as $30 million in assets. They create this massive database. There are about 700,000 to 800,000 people in the world who are considered ultra high net worth.
WealthX recently got acquired by a publicly traded company. I was able to look at their numbers. They were doing about $12 million in revenue with $800K in profit, and they sold for $20 million in cash. They sell subscriptions for $15,000 a year — mostly to financial advisors and wealth managers.
I think this business is shockingly easy to copy. How they get data: who owns a private jet, who’s donating to nonprofits (that’s public), property purchases, certain hobbies, being on lists like Tiger 21. Not hard to build the data pipeline.
The founder of CB Insights told me their whole model is: there’s a bunch of dirty, muddy data out there, and we go find it, rinse it off, make it easy to consume. That’s exactly what WealthX did.
I’m pretty certain this business is easy to replicate today. Ways they get data: jet ownership, nonprofit donations, property purchases, club memberships. Then you build a pretty interface on top. With import.io, Airtable, Zapier — you could build a robust MVP without a huge team.
WealthX probably didn’t make a ton of money because they hired too many people. When they launched in 2010, a lot of that third-party tech wasn’t available. They had a bloated staff doing manual work.
CB Insights did this way bigger — probably worth close to a billion dollars. Same principle: data is there, it’s muddy, we dig it up, wash it off, package it. You couldn’t have gone and got it yourself — or you wouldn’t want to. So let us do that for you.
The Hot Girl Side Hustle: UGC Advertising [01:13:00]
Shaan: Let me do a more fun one. I call it my hot girl side hustle.
I’m trying to identify simple things someone in a certain demographic can do to generate $10K to $50K a month with no big upfront costs and no hard skills required.
The dominant way to advertise now is not super high production value content. It’s what we call selfie content — or UGC, user-generated content. I just call it selfie because nobody knows what UGC means.
Brands aren’t just doing this on social. If you look at a TV commercial for Chime, it’s the same thing. Digital-first companies, when they finally go to TV, their whole training has been: this thing that looks like a normal person talking — that’s what works. People skip past the polished stuff.
Sam: I know this. When we first started running ads — my team ran eight figures in ads. We’d test four types of creative: a hot guy, a pretty girl next door, a smoking hot model, and a plain girl. The pretty girl next door consistently outperformed — significantly — among both men and women.
Shaan: There’s a woman on Twitter, Social Savannah — Savannah Sanchez — and what she does is: send me the product, I’ll take a bunch of selfie-style videos, I’ll be the actor in your ads, and I know what makes ads work because I understand the creative side.
She used to work at an agency called Common Thread Collective. I imagine what happened: they were like, hey, Savannah, you’re good-looking and charismatic — let’s just use you. And then she learned how the business worked and spun out her own agency.
This essentially unbundles the ad agency. Agencies do creative, strategy, media buying, reporting, multi-platform management. The smart thing is to just do the most important part — the creative — and skip the undifferentiated services. If you’re good at it, you become the Sandwich Video of DTC brands.
Sam: I think she’s probably making $50K to $100K a month. Maybe more. Brands need this content. I don’t want to arrange photo shoots — I just want to send the product to one person, and she films 30 videos in her bedroom, and I have a month’s worth of ad creative to test on Facebook.
Nathan: This is basically the Shaan Puri model applied to advertising. It’s completely unbundling the ad agency around the most leveraged piece.
Blue-Collar Tech: Tele-Operated Heavy Machinery [01:25:00]
Shaan: All right, I’ll do a more high-tech one. This is actually more blue-collar.
There are a lot of heavy machinery drivers on construction sites — forklifts, cranes, things that dig up dirt. Really expensive labor, highly skilled, also dangerous.
Fast-forward 10 to 20 years: I think all of this gets replaced with tele-operated machinery. Not self-driving — tele-operated. A driver sits in a “driver city” — could be anywhere in the world — with three monitors: front windshield, left window, right window. They drive the thing remotely. The latency has been engineered down so low that I’ve sat in a car being driven by someone at a computer somewhere else, and they drove me around a parking lot and parked it perfectly.
But here’s the more interesting model. You only need a human driver at an intersection, or for a left turn, or when parking. The highway stuff — the computer can handle that. So what they do is: the computer drives 80% of the time, and only when it hits a defined situation — like a left turn — does it pop up on the driver’s screen. The driver takes over, does the left turn, says “back to autopilot,” and the machine drives off again.
That’s called human-in-the-loop technology. Mostly AI, but a human takes over for the hard parts. One driver can work on 10 construction sites simultaneously. They make 3x more. The cost of the operation goes down.
There’s a company called Phantom Auto doing this for vehicles. For construction, there are a few companies I don’t know the names of.
Sam: This is actually related to a company called Winner Winner — run by our friend Joe Spicer. They have this massive warehouse in Las Vegas full of claw machines, and you can play them remotely from your phone. When you win a prize, they mail it to you.
Shaan: Exactly the same concept. And Winner Winner’s numbers were actually pretty decent. Same principle as tele-operated construction — the machine can do 80%, the human handles the last 20% that requires judgment.
Tele-operation is going to be more popular than people realize: remote construction, remote cleaning, remote arcade games. Win-win-win — humans work remotely, make more, and the cost of operations goes down.
Pavel Durov and Telegram [01:32:00]
Shaan: Do you follow the founder of Telegram, Pavel Durov?
Nathan: No.
Shaan: He started VK — basically the Russian Facebook. Facebook never beat VK despite beating every other country’s social network. Then the Russian government basically took over VK, he refused to cooperate, and he had to flee.
He creates Telegram — encrypted messaging, basically WhatsApp but not owned by Facebook. He self-funded it for a long time, put like $50 million of his own money in. 36 years old, worth $20 billion.
His Instagram is hilarious. You will never see anyone else in his feed. No friends, no girlfriend, no family. Just him, usually with no shirt on. He’s very ripped. Pictures of him like a model — sitting on a sand dune, riding a horse into the sunset, in a speedo under a waterfall.
Nathan: He’s like a James Bond villain who’s genuinely idealistic.
Shaan: He runs updates like: Telegram remains 100% free, 100% encrypted, 100% independent. He has a strong idealistic streak around privacy and freedom.
They did attempt an ICO — I think they raised $3 to $4 billion — and then the SEC said you can’t do that without following securities regulations, so they just pulled the plug and gave money back. But for a couple of years they held billions of dollars in float. Crazy story.
Nathan: That company is doing that as like a skeleton crew?
Shaan: Look at their jobs page — five open positions. Content moderator, translator, junior accountant, assistant to the CEO. Maybe 30 engineers in total. One of the most valuable companies per employee in the world.
ConvertKit Competitors, Mailchimp Drama, and Closing [01:40:00]
Shaan: I’m a ConvertKit user and I complain to Nathan all the time. I pay like $400 a month and I keep asking for a discount.
Nathan: You complained, I didn’t give you one. Tim McGraw and Arnold Schwarzenegger both have ConvertKit newsletters and pay full price and never ask me about it. It’s always the internet influencers who say I should get it for free.
Shaan: Arnold Schwarzenegger’s on ConvertKit?
Nathan: Yes. His list is probably bigger than yours actually.
Sam: I was wondering — are you Mormon?
Shaan: Dude, I’ve been thinking this forever. You live in Idaho, you had kids at 21, you work hard, you’re honest, you don’t seem to party.
Nathan: Sam’s been saying this for a while. I mostly hear it from you.
Sam: You have hardcore Mormon energy.
Nathan: The caffeine’s a no for me, so I get it.
Shaan: So on the Mailchimp thing — their CEO, Ben Chestnut, had to step down. Did you see what happened?
Sam: He basically sent an internal email criticizing the practice of employees introducing themselves with preferred pronouns. Said he thought this approach did more harm than good. And then he stepped down coincidentally right after.
Shaan: Whether you agree with him or disagree — having to step down from your own company for having a point of view on this seems like a pretty severe penalty.
Sam: His intentions, reading the email, don’t seem negative. His intention seems to be about inclusion and productivity.
Shaan: There’s a part of the email that got dicier: he talked about a tiny number of transgender employees making a whole bunch of other employees adopt a new communication paradigm. That’s where it got more charged.
Sam: Either way, rip Ben. I like Ben. I don’t know him, but I’ve emailed him a ton of times and he’s never replied, so we kind of know each other.
Shaan: Well, karma.
Nathan: He gave me the wrong hotel. [Laughs] I emailed him to meet up at a conference, he said meet me in the hotel lobby — gave me the wrong hotel. So I missed the meeting.
Shaan: Incredible. Nathan, thanks for coming on. We’ll hang out in Austin when you get that new property.
Nathan: Challenge accepted. Grudge initiated.
Earlier in the episode, Sam told the story of how he emailed 20 New York hotels asking for a discounted rate in exchange for mention to his podcast and newsletter audience — and got several $1,000-a-night hotels down to $350 to $400 a night. The tactic: mention the podcast audience and newsletter, explain you need three months, and say your budget is $350 a night.
Sam also described his law of reciprocity. When he used to buy cars, he’d bring two cans of Coke and offer one to the salesperson — and get discounts through this.