Acquired XX
[0] That's woodblocks, right?
[1] Do your best impression.
[2] That's good one.
[3] Welcome to Season 4, Episode 9 of Acquired, the podcast about technology acquisitions, IPOs, and direct public listings.
[4] I'm Ben Gilbert.
[5] I'm David Rosenpull.
[6] And we are your hosts.
[7] Today we are covering the spiritual sister of our previous episode on Zoom, Slack, and the largest enterprise IPO in history.
[8] And, of course, it wasn't even actually an IPO, but rather a direct listing.
[9] Barry McCarthy, he rides again, the acquired superhero.
[10] Today we'll talk about what all that means, the crazy history of Slack's evolution from even before it was founded, where its roots really are, and break down their business and their market cap today and what it could do in the future.
[11] Okay, listeners, now is a great time to thank one of our big partners here at Acquired, ServiceNow.
[12] Yes, Service Now is the AI platform for business transformation, helping automate processes, improve service delivery, and increase efficiency.
[13] 85 % of the Fortune 500 runs on them, and they have quickly joined the Microsoft's at the NVIDias as one of the most important enterprise technology vendors in the world.
[14] And, just like them, Service Now has AI baked in everywhere in their platform.
[15] They're also a major partner of both Microsoft and Nvidia.
[16] I was at Nvidia's GTC earlier this year and Jensen brought up ServiceNow and their partnership many times throughout the keynote.
[17] So why is Service Now so important to both Nvidia and Microsoft companies we've explored deeply in the last year on the show?
[18] Well, AI in the real world is only as good as the bedrock platform it's built into.
[19] So whether you're looking for AI to supercharge developers and IT, empower and streamline customer service or enable HR to deliver better employee experiences.
[20] Service Now is the platform that can make it possible.
[21] Interestingly, employees can not only get answers to their questions, but they're offered actions that they can take immediately.
[22] For example, smarter self -service for changing 401K contributions directly through AI -powered chat, or developers building apps faster with AI -powered code generation, or service agents that can use AI to notify you of a product that needs replacement before people even chat with you.
[23] With ServiceNow's platform, your business can put AI to work today.
[24] It's pretty incredible that ServiceNow built AI directly into their platform.
[25] So all the integration work to prepare for it that otherwise would have taken you years is already done.
[26] So if you want to learn more about the ServiceNow platform and how it can turbocharge the time to deploy AI for your business, go over to ServiceNow .com slash acquired.
[27] And when you get in touch, just tell them Ben and David sent you.
[28] now?
[29] So, we start pretty far away from Slack's gleaming new San Francisco downtown San Francisco headquarters, pretty far away from Wall Street.
[30] In fact, very, very far away from Wall Street.
[31] We start in 1973 in Lund, British Columbia in Canada.
[32] Lund is, I'm not sure exactly how far maybe an hour or two north of Vancouver in British Columbia.
[33] and it was where Highway 101 ended in the mid -70s, north of Vancouver.
[34] And we start there in a small log cabin located on a hippie commune.
[35] And this cabin does not have electricity.
[36] It does not have running water.
[37] But in 1973, there is a boy born in this cabin to parents David and Norma Butterfield, who were, as you can imagine from this season.
[38] I am setting here and describing pretty hardcore hippies at the time in 1973.
[39] And the boy's name that Norma gives birth to this year is Dharma Jeremy Butterfield.
[40] Dharma is an incredible name for someone born on a hippie commune.
[41] Incredible, incredible.
[42] So perfect.
[43] Let's rewind just a little bit, though.
[44] How and why did we get here?
[45] So we start with David Butterfield, the father of Dharma.
[46] So David was an American originally.
[47] he was from Connecticut, and he was a college student at Harvard during 1968, during the Summer of Love.
[48] And after the Summer of Love, he decides he's going to drop out of school, drop out of Harvard.
[49] You know, he's been disillusioned by the RFK and Martin Luther King assassinations, all the, you know, both great and terrible things that happened that year in 1968.
[50] He drops out of Harvard, and he moves to France.
[51] But shortly after he gets to France, he receives a telegram one day from the United States government informing him that he has been drafted for the Vietnam War.
[52] Does this explain the Canada thing?
[53] This does eventually explain the Canada thing.
[54] He comes back to the U .S. and he does enlist and he goes to basic training and he learned, you know, while he's in training that he had wanted originally to be a medic.
[55] He learns, though, that he's going to become a sergeant and he's going to get assigned to active combat duty and he's going to get shipped off to Vietnam, very probably he knows to his death.
[56] He decides one day before he gets shipped off, I don't know exactly where he was stationed for training, somewhere in the north of the U .S., and he decides to just drive to Canada and not come back one day.
[57] He drives across the border.
[58] He ends up in Montreal, and while he's in Montreal, he meets Norma, who I believe was his student at McGill at the time in Montreal, and they've found.
[59] fall in love, they get married, and they decide that they're going to move out west to British Columbia.
[60] So they just start driving.
[61] And they drive across Canada, across the country.
[62] They get to Vancouver, and then they turn north.
[63] And then the highway ends in Lund.
[64] And it turns out there is actually a hippie commune already in Lund.
[65] So they joined the commune.
[66] It's kind of an amazing story.
[67] And we're going to continue it.
[68] But David sadly passed away in 2017.
[69] And in his obituary, I just love this.
[70] It says, if you asked David who he was, he often said, who I am is I love my wife.
[71] That's just so nice.
[72] It's so great.
[73] What a, what a cool person.
[74] You know, David, this is exactly how I thought the episode of the largest enterprise software IPO in history would start.
[75] Exactly.
[76] Exactly.
[77] Back to Dharma.
[78] He grows up initially, still no running water or electricity.
[79] He's on this commune.
[80] When he gets to be about five, though, David and Norma are kind of starting to question the whole, you know, their sort of life choices and anti -capitalist philosophy.
[81] You know, it's now the late 70s.
[82] The Vietnam War is long over.
[83] And they say, you know, maybe, maybe it's time to sort of rejoin normal society.
[84] We have a five -year -old now.
[85] So they move down to beautiful Victoria, the capital of British Columbia, beautiful, beautiful town.
[86] If listeners, if you haven't been highly recommended.
[87] Especially, uh, you Seattleites.
[88] It's way too easy to take the Clipper.
[89] Don't miss it.
[90] Yeah, definitely don't miss it.
[91] So they moved down to Victoria and they rejoin mainstream society there.
[92] David ends up, so he had built the log cabin, uh, himself.
[93] He ends up getting into real estate development and he becomes a prominent sustainable real estate developer.
[94] Um, really cool in Victoria.
[95] Stort though, so he's now growing up, he gets his first computer at age seven in Victoria and becomes immediately enamored of it when, or Stort, I should say, Dharma.
[96] When he is, when he is 12, he decides to change his name to Stort because he's probably getting teased at school at this point in time and he wants to be a little more normal.
[97] And Stort is like the most normal name that he can think of.
[98] So he legally changes his own name at 12 to Stort Butterfield.
[99] He ends up staying local for college.
[100] He goes to college at the University of Victoria.
[101] He studies philosophy.
[102] And as you might imagine, given the lineage of his parents, he's quite smart and does very well in school, and he thinks he's going to become a philosophy professor.
[103] That's his goal.
[104] He goes off to Cambridge in the UK, and he gets a master's in philosophy.
[105] And I assume it was part of a PhD program that he was intending to complete.
[106] For listeners who have not heard Stuart speak before, it is very apparent when you hear of speak that he has a background in philosophy.
[107] He sort of can't help but diving into the existential and the metaphysical in any topic.
[108] In any topic, And I think there's a great episode with Kara Swisher on Recode Decode, where he's sort of talking about why gaming is interesting for the world, why communication is interesting for the world, and ties it to such a sort of strong base in philosophy where I feel like every time I've watched or heard an interview with this guy, you have these aha moments that you otherwise would never, ever have.
[109] Yeah, he's cut from a different cloth for sure than your normal tech CEO.
[110] We mentioned, though, he got his first computer at age seven.
[111] he'd always also been kind of fascinated by computers and this was kind of the early days of the internet and he also as legend has it became really interested in the internet because he was a huge of course fish fan amazing the bandfish and so fish had a newsgroup at wreck dot music dot fish and apparently young dharma slash store it was all over this news group david what is a news group Well, you might ask, what is IRC?
[112] News groups, of course, were the, gosh, I don't know, what do you call it, like, predecessor to Reddit, I guess, on the spiritual predecessor.
[113] Yeah, is it sort of the pre -BBS or bulletin -port system, way of finding community on the Internet?
[114] Yeah.
[115] So, Stuart, even though he's pursuing his academic career and philosophy, he is really into the Internet, has gotten into coding on the side, making websites all through college.
[116] And so while he's, after he finishes the master's portion of his program at Cambridge, it's now 1998, and the web 1 .0 bubble is fully inflated.
[117] And he says, you know what, I need to, I got to throw in with this.
[118] I got to throw it.
[119] Like, now is the time.
[120] I can't miss this train.
[121] So he moves back to Vancouver in British Columbia.
[122] And he joins a startup called communicate .com.
[123] Dude, the domain names were so easy to come by that.
[124] Well, and domain names was the name of the game here.
[125] So the idea of communicate .com is they were going to build the everything store on the internet, the e -commerce platform, and they were going to do it by owning domain names for all of things that you might buy or things you might be interested in.
[126] And that was going to be the front end.
[127] So you would go to whatever .com and you would buy stuff powered by communicate .com.
[128] Only the things that could be...
[129] Which is so funny, because this was the year that Amazon IPOed.
[130] So Amazon was already definitely a thing, but was...
[131] But only for books.
[132] Only for books at the time.
[133] As you might expect, given that business plan I just described, and being in Vancouver, it was a total disaster.
[134] And so Stewart shows up, and he's quickly like, man, I thought I was going to get rich here in the, you know, sell out from academia, get into the dot -com.
[135] he decides to head for the exits pretty quickly.
[136] And that was very prescient because shortly after he does, of course, the company blows up.
[137] But he leaves and he meets somebody really important there.
[138] He meets his coworker, Jason Classen, who he's had next to at Communicate.
[139] And Jason, of course, had a side hustle at the time because everybody had a side hustle.
[140] And Jason's side hustle was he was building another website called Gradfinder .com.
[141] And this was like a competitor to class.
[142] Classmates .com, Seattle Company.
[143] It was sort of proto, early, early proto Facebook of find people that you went to high school with or you went to college with and stay in touch with them online.
[144] Actually a pretty good idea.
[145] So Stewart convinces Jason, hey, let's get out of here from this crazy company.
[146] Let's go build your website, Gradfinder .com, as a real company.
[147] So they leave.
[148] And very quickly, I think within a year, they end up selling Gradfinder .com to another dot com for a small amount of money.
[149] Stewart's first exit.
[150] Stewart's first exit.
[151] Everybody celebrates.
[152] They decide they're going to take the year off.
[153] They didn't make life -changing money, but enough money to kind of take the year off and enjoy life.
[154] During that year, of course, the bubble bursts.
[155] And so they all come back to, you know, nuclear waste land in the internet sector.
[156] During that time, though, Stewart starts a blog.
[157] He's kind of always on the leading edge of what's going on on the internet.
[158] And he meets somebody in the blogosphere.
[159] He meets a woman named Katerina Fake, who also famously had and still has a very popular blog during the time.
[160] And they kind of strike up a relationship through blogging.
[161] And one thing leads to another, and they end up getting married.
[162] And Katerina moves out to Vancouver.
[163] Wow.
[164] Yeah, pretty amazing.
[165] Another relationship that Stuart fosters online, the store launches this thing initially as during the year off kind of as just a fun little competition he wants to launch called the 5K competition i think it was a play on like a 5k you know running race yeah yeah and the challenge the 5k competition was a challenge for people around the world to build websites that were five kilobytes or less whoa difficult then literally impossible literally impossible i think a lot of these websites were like black and white like very very simple but um but it was like how cool of a website could you make this was the challenge in less than five kilobytes of memory creativity and what is it constraints inspire creativity indeed and i don't know if this this probably wasn't stuart's idea at the time but if you you know could devise some ruse to um you know to identify really talented programmers out there what better wig thing could you come up with than something like this and so a guy named Eric Costello, who I believe was in St. Louis at the time, enters the 5K challenge, the 5K competition, and Stort's like, wow, what you've created is amazing.
[166] Like, we should work together on something.
[167] Eric ends up also moving out to Vancouver.
[168] And this kind of rag tag team of Stort and Jason, his co -founder from gradfinder .com, Katarina, his wife, and Eric, this whole, you know, web community, they decide, you know, we should do something together.
[169] The whole internet gets together in one place.
[170] It's almost like a commune, one might say.
[171] An internet commune.
[172] So this is now 2002.
[173] It is nuclear wasteland in tech and the internet from a business perspective.
[174] But the four of these folks, they're really inspired by these relationships they've built online, the kind of power that they're seeing still exists in the internet.
[175] And they decide, you know, we can use web browser.
[176] technology to create like something pretty cool.
[177] And wouldn't it be cool if we made a game that people could socialize and interact with each other and build these types of relationships just online in a browser through like a more rich experience than these, you know, news groups and bulletin boards and things that people are using?
[178] And of course, that would be very cool.
[179] But listeners, just think back to yourselves about websites you were using in 2002 and web technology that was available in 2002.
[180] And then even think today how difficult it would be to implement a massive multiplayer online game using the web today.
[181] David, I suspect that was quite a trying undertaking.
[182] And I think they knew this because they decided, you know, like most game companies, they were going to have a name for the company and then they were going to have a name for the game.
[183] The name for the company was Ludacorp.
[184] I think they knew the scale of the undertaking that they were embarking on here.
[185] They decided to name the game, Game Neverending.
[186] Which would be prescient.
[187] Also a prescient.
[188] So the four of them, they work on it for a while.
[189] They ship an early version and people out there on the internet, you know, again, these folks are kind of already sort of internet celebrities to the extent that that meant something back then.
[190] People start using it.
[191] And there's a community of people that really like this early version.
[192] But the team, you know, it's just the four of them.
[193] they can't get enough money and resources to like actually build out the real full vision of what they want to try and achieve.
[194] And so they figure out that, you know, they're trying to raise money.
[195] They can't raise money.
[196] In 2002, it's like if you're raising venture capital at all, it's on terrible terms.
[197] It's only for basically sure things or big enterprise software plays.
[198] No one's getting funded on a consumer dream to be able to do something like this.
[199] Not to mention here are these like, you know, equivalent of modern hippies hanging out in Vancouver who all met on the internet and two of them are married.
[200] I could just imagine venture capitalist reaction at this point in time.
[201] So they realize they got to do something.
[202] They're not going to be able to raise venture capital, but they got to get some more money to make this happen.
[203] So Stuart and Katerina, as the story goes and Stewart swears that this is true, they go to New York for a tech conference and this is all swirling in their heads.
[204] And when they get there, Stuart gets food poisoning.
[205] and he gets really, really sick.
[206] And he's up all night in their hotel room in New York, you know, can't sleep, very sick.
[207] And he starts, like, getting kind of basically fever dreams and hallucinating.
[208] I'm pretty sure this isn't the first time Stuart has hallucinated.
[209] But in this particular time when he's hallucinating, this idea comes to him.
[210] And the idea is, hey, we're using some pretty cool technology in the browser that we're trying to make this game with.
[211] but we could use this same technology to build some other cool like tools and services for consumers more broadly.
[212] If we did that and we built something, we could probably sell it pretty quickly.
[213] Remember, he and Jason have sold Gradfinder .com already.
[214] They're used to these kind of quick flip ideas.
[215] And so it starts like, man, if we make a side project, we could, maybe we could flip it pretty quickly for like a million dollars or so.
[216] We could use those proceeds to keep funding the game and then we can build out the game.
[217] So all in service.
[218] of the game of course.
[219] All in service of the game.
[220] And so as he says, he writes out all the ideas of things they could do on scraps of paper at 5 a .m. in the hotel in New York.
[221] And he realizes that photo sharing is, and photo uploading is something they could do.
[222] And I believe it was that they had built a photo uploader for the game that you could upload your profile picture.
[223] I believe that was the core of the technology.
[224] Yeah.
[225] And I think if I'm remembering right, there was even sort of an ancillary thing where you could upload alternative photos other than the profile picture.
[226] So if people looked at your profile in the game, there were photos in there that you could use for other things too.
[227] Ah, okay, cool.
[228] So like, man, we could just repurpose this technology and let people share photos, you know, more broadly.
[229] Digital cameras are a thing.
[230] I mean, I remember.
[231] It should theoretically be an explosion of digital photography.
[232] Totally.
[233] I was, I was heading off to college at this point in time.
[234] And all my, you know, classmates that were showing up as freshmen at, uh, at college, everybody had their point and click digital camera.
[235] You get your cool picks.
[236] You got your power shy.
[237] You get your Sony Digital Elf.
[238] I mean, this is like, this is the era.
[239] Yeah.
[240] Oh, man, the Sony Digital Elf.
[241] Oh, man. I hadn't thought about that in years.
[242] Okay.
[243] So they, the four of them, they decide, well, we're going to have a vote on this.
[244] Are we going to keep plugging away at the game or should we pause the game, go work on this photo sharing idea?
[245] They do the vote and they decide not to do it.
[246] Stewart is outvoted.
[247] And, um, look, Stuart, you.
[248] You seem like you got to get a head on your shoulders and you got a nose for, you know.
[249] But you literally came up with this during a fever dream.
[250] Yeah.
[251] Stuart, though, in this would presage his skill and power as an executive and a CEO and a broker of influence, he starts going to work on Eric.
[252] He sees Eric is the weak link here.
[253] He thinks he can get him to change his vote.
[254] And he does.
[255] He gets Eric to change his vote that pushes it over the edge and what the service that would become known as Flickr is born.
[256] We're now in December of 2003.
[257] And keep in mind, too, the proposal that we should do a photo sharing site, it's still seven or eight years from being a meme.
[258] Like, the notion of, oh, yeah, we're just going to pivot into photo sharing.
[259] Like, not only were they the first ones to do that, they were a whole technology wave before the set of people that, you know, you're all laughing about in your heads on, oh, you had another photo sharing app.
[260] Like, we are so far in the early.
[261] innings here.
[262] Photo sharing is not on anybody's mind in technology.
[263] They realize, though, that if they're going to really commercialize this service, they need some extra engineering resources just before them, again, isn't going to be enough.
[264] Well, it turns out, so where do they turn?
[265] There is this guy in the UK just outside of London named Cal Henderson, who is a hardcore fan of game never ending in the early pro to day.
[266] Oh, that's how Cal and Stewart met.
[267] That's how Cal. This is how Cal and Stewart meet.
[268] And Cal was such a hardcore fan.
[269] You know, he's active on the bulletin boards about Game Neverending.
[270] He wants to, like, really get in there and help.
[271] He hacks their email service.
[272] Hey, guys.
[273] Just wanted to say hi.
[274] I'm super talented.
[275] Yeah.
[276] Look at me. Look at me. And so they were like, I imagine before this idea to start working on Flickr, they were like, man, what are we going to do with this guy?
[277] And Story, again, in, you know, just showing these flashes of brilliance as an executive and marshaler of resources, he's like, you know what, Cal really cares about the game.
[278] We're doing this new thing to get the resources to build the game.
[279] Maybe Cal would help us.
[280] So he gets in touch with Cal and he convinces him to not work on the game, but come help them build this service.
[281] And it turns out, of course, as you would imagine, by Cal hacking into their email servers, Cal is like an incredible developer, just incredible, like a truly world, world, world class developer.
[282] So do they move him from the UK to Vancouver?
[283] Eventually they do, but at first, remember, they don't have any money.
[284] Stuart's like, the way I'll pay you is, um, you Cal, you know, like all nerds at this time, he has a, he has an Amazon wish list.
[285] How about as you start working for us, I'll just buy you stuff off your Amazon wish list.
[286] Raised on a commune and somehow never.
[287] never left yeah amazing amazing so once cal is on board things just fly very quickly so it was December 2003 when they decided to build Flickr by February 2004 it's ready and so like i mean we're talking 60 days or less uh what would that be like less like 45 days yeah i mean you have some existing infrastructure to build it on because you know game never ending had servers and you know the ability to bring different people together on the internet.
[288] So like there was infrastructure there, but it's not like there was cloud services.
[289] No. And no development frameworks, no web 2 .0 technologies.
[290] This is the invention of web 2 .0 technologies here.
[291] When we know from the name, I mean, every web 2 .0 company after them copied the no E and flicker.
[292] Yeah, indeed.
[293] So it's now February and starts like, man, we're ready to go.
[294] Like, let's ship this thing.
[295] And the rest of the team's like, we're shipping the thing and starts like, yeah, I'm going to go talk at this conference, O 'Reilly's putting on their E -Tech conference in San Diego.
[296] Great name.
[297] Yeah.
[298] Rather than talk about the game, I'm just going to launch Flickr.
[299] Like, we're going to do it.
[300] I'll have a stage.
[301] It'll be perfect.
[302] So he does.
[303] He announces and demos.
[304] The team works through the works through the night, the night before the presentation.
[305] They launch Flickr on stage at the E -Tech conference.
[306] And remember, all the bloggers that, you know, the blogging is sort of like this is where they all come together.
[307] They're part of this community.
[308] All the bloggers are there.
[309] People are just blown away.
[310] They had never seen, nobody had ever seen anything like this.
[311] A web service native in a browser where you can actually do something useful.
[312] Like it's an actual application built on the web.
[313] And if you think about the web to this point, it had been mostly sort of unidirectional content.
[314] Like they were flat websites.
[315] You would go.
[316] You could potentially see some pros.
[317] There was images that were uploaded.
[318] You know, we had the, advent of e -commerce with Amazon having, you know, successfully navigated through the dot -com bubble.
[319] So you've got the first place where you can actually start to buy things.
[320] But there's still not the notion that we have today that websites are interactive.
[321] Yeah.
[322] That I can upload things, download things, interact with other people.
[323] Yeah, that's not on the actual site itself.
[324] You needed to download something like Napster in order to do that.
[325] So the press is, is blown away there.
[326] And some mainstream press, but even more importantly, all these other bloggers that are there.
[327] And so the blogosphere ends up being the perfect early adopters for Flickr because all of these people have blogs.
[328] Most of them run on blogger.
[329] So at this point, Google had already bought Blogger at Williams's company, but Blogger didn't have native photos in it.
[330] And so what does Google and blogger do?
[331] They see this and they're like, oh, great, use Flickr for photos if you want to host photos on your blog and it just starts to spiral and this is this is like the google of yester yester year where rather than saying oh we're going to clone that into our product it's oh cool that that's great then we don't have to build it we'll just put a button to make your thing work sure you can have some users man i mean this really is the it's the these are the young idealistic don't be evil days of the internet yep or i would say the post dot com crash people had seen the evil where we're going to be idealistic once again um and so so flaker really starts to take off uh this They quickly get a lot of M &A interests.
[332] So the Stewart's plan is coming true.
[333] Google, Yahoo, Ask Jeeves.
[334] Remember Ask Jeeves?
[335] The big three, man. Of course.
[336] They all want to buy the company.
[337] Yahoo becomes the most serious and moves the fastest.
[338] And by the end of 2004, the deal is done.
[339] And Flickr is acquired by Yahoo within, you know, the...
[340] I don't think I ever realized how short the independent life of Flickr was.
[341] It was a brief flicker in time.
[342] Oh, boy.
[343] Okay.
[344] I'm sorry.
[345] I'm sorry.
[346] Oh, David.
[347] By, so by the end of 2004, the deal is done, Yahoo acquires Flickr for, quote, between $22 to $25 million.
[348] It's a pretty narrow range.
[349] Like, I feel comfortable not knowing the exact number.
[350] Everyone high fives and celebrates.
[351] But there's just one catch, though.
[352] Yahoo is paying them all this money, but they want the team in addition to Flickr.
[353] and they put a three -year earn -out.
[354] So there's no going back to game, never -ending right away.
[355] Yahoo moves them all down to San Francisco.
[356] The plan is that they are going to continue to build and grow Flickr within Yahoo, with all of the resources and reach that Yahoo as a major Internet company can give them.
[357] Now, David, you know as a person who does a podcast on the Internet about technology acquisitions.
[358] That's how it tends to go, right?
[359] You have this amazing resource and engineering pool and brand of big company.
[360] You have this wildly disruptive product that gets bought.
[361] Surely it's a match made in heaven.
[362] Well, actually today, depending on who the acquirer is or how much they listen to acquire, there might be a good chance that that could be the case.
[363] But back in 2004 and with Yahoo at the time, yeah, no. So the team gets there and it is not a match made in heaven.
[364] Oil and water.
[365] Yeah, let's just leave it at that.
[366] Stuart talks about this a lot.
[367] He says a couple things.
[368] One, just the decision to sell and do that, you know, obviously that was the plan from the beginning with pivoting into Flickr.
[369] But even more than that, you know, everybody he knew, all the people advised him.
[370] And they were like, yeah, sell the company.
[371] Like, of course, fan.
[372] You're like a couple months in.
[373] You got this traction.
[374] It's not like you've raised a lot of money.
[375] There's not a ton of employees.
[376] You're going to get rich individually from it.
[377] Like, this is it.
[378] This is the dream.
[379] Like, yeah.
[380] 22 to $25 million hit that bid.
[381] And it wasn't even a question.
[382] But now, of course, nobody would ever do that.
[383] Like, never, ever, ever would somebody do that, you know, Instagram selling for a billion dollars to Facebook.
[384] Like, people wouldn't even do that.
[385] You know, if you have a tiger by the tail like this, you know, you're going to be flooded with offers to invest from venture capitalists, everybody telling you not to sell, go big, you know, build a big independent company.
[386] None of that existed back then.
[387] So it was like, I don't think the thought even crossed their minds once not to sell to then you know he talks about the experience at yahoo and um i think it actually you know as we've i think seen so far in this history stuart has a lot of raw potential certainly creatively but also on the business and and leadership side yeah and while i think his time at yahoo was oil and water like he says you he learned a lot there too like he saw how a real big company is run uh for better or for worse you know how and he worked for Brad Garlinghouse, who was a really talented executive there at Yahoo. I mean, these were the days like Jeff Wiener was there.
[388] There were a lot of great folks at Yahoo. How long did Stuart stay?
[389] Do you know?
[390] So he stayed, so the whole team was a three -year earn -out.
[391] They stayed for the three.
[392] The whole team stayed for the three -year earn -out.
[393] But right at the end of it, everyone else leaves.
[394] Caterina leaves.
[395] That was a big deal.
[396] And so we're now in 2008.
[397] And folks might remember, these are the days when Yahoo is a mess.
[398] These are the Carol Bart's days and everything.
[399] Oh, my gosh.
[400] And Kara Swisher is all over the company.
[401] This is like the heyday of Kara Swisher.
[402] Take down after takedown.
[403] Just blog post after blog post on all things.
[404] Just painting a picture of what's going on internally and it's not pretty.
[405] And one thing to remember about why Katerina leaving was such a big deal was because even though Flickr was a photo sharing website, everyone who was on it felt like they were a member of the community and like it was a lifestyle they had opted into and and catarina and just it's a semi -sense to her but i catarina was a largely the public face of this is the community that i belong to and these are the types of people that you know i love to leave comments on their pictures i love it when they leave comments on mine you know i think there was a social graph there was an amount of following and friends that you could have on there and so it was like you joined a hippie commune and your leader left.
[406] I wasn't.
[407] It was more than a hippie commune.
[408] I mean, Stort and Katerina were on the cover of Newsweek.
[409] Yeah, I shouldn't, I shouldn't try and belittle that.
[410] Yeah.
[411] No, but I mean, it was the same ethos, but it was, it was, this was like big.
[412] This was, this was really mainstream, especially among younger demographics.
[413] All of this with everything going on at Yahoo, like, they were really upset.
[414] And so they actually convinced Stor, so Jason, Eric, and Katerina all leave as soon as the are not done.
[415] They convinced Storner, to stay a little longer.
[416] So he does.
[417] He stays a few extra months.
[418] Ultimately, though, he can't take it either.
[419] And so in the summer of 2008, he leaves.
[420] And he writes, Brad, his boss makes him write a, he's like, dude, you can't just leave.
[421] You got to write like a resignation man. He writes something like nice, like the company's in all this trouble that I can talk about your time here.
[422] And Stuart writes this hilarious resignation memo that's still on the internet where it's just complete like absurdism.
[423] He talks about how, yeah.
[424] Yahoo is like a metals, you know, production company and tin is, the tin, the metal is in Stort's blood and he joined to like mine tin and now they're doing it.
[425] It's absurd.
[426] I'll look it up after the show.
[427] Yeah, it's funny.
[428] We'll link to it in the show notes.
[429] Anyway, that is the end of Flicker and the end of Stort's brief internment at Yahoo. Unfortunately, though, and sadly, it's also the end of Stort and Katerina's marriage right after Stort leaves.
[430] The marriage, the and Katerina ends up moving back to New York.
[431] She then co -founds the company Hunch with Chris Dixon in New York.
[432] It's up getting acquired by eBay.
[433] And Katerina also becomes the chairwoman of Etsy.
[434] So really talented and great career she's had since.
[435] Stort, though, he's kind of like, whoa.
[436] I grew up on a commune.
[437] And here I am.
[438] Sure those were his first words on Yahoo. I'm done.
[439] He needs to just take a break.
[440] He leaves San Francisco and he moves back to Vancouver.
[441] Take some time off through the rest of 2008.
[442] But by the beginning of 2009, you know, the other folks, Cal and...
[443] He's got timing, by the way.
[444] Like, both of these companies are started in the rubble of financial crashes.
[445] Totally, totally.
[446] Yeah, it's beginning of 2009.
[447] The world has fallen apart.
[448] Eric and Jason and Cal, they've all left Yahoo too.
[449] And remember, you know, they made certainly a good amount of money on the Flickr acquisition, but not enough that you can survive forever, especially after the financial crisis.
[450] And they're like, Kay Stewart, like, we're ready to get the band back together.
[451] What's next, man?
[452] You're our fearless leader.
[453] And so they come back up to Vancouver and they say, you know, game never ending.
[454] The time wasn't right.
[455] It didn't happen the first time.
[456] But it's now 2009.
[457] So many more people are on.
[458] online.
[459] Gaming is so much more a thing.
[460] World of Warcraft is huge.
[461] Massively multiplayer online gaming is so much more a thing.
[462] Maybe now's the time to take another shot at this.
[463] So they do.
[464] They start, they start the company.
[465] Once again, it is a different company name from the game name.
[466] They call the company Tiny Speck, where the four of them start.
[467] And they call the game Glitch.
[468] I don't know if it was glitch in the matrix or exactly where the name came from.
[469] But they see World of Warcraft.
[470] They see how popular that is, they say, once again, much like with Flickr and the original vision for game never ending, we can do a lot of what World Warcraft is doing.
[471] We can do it on the web, natively, in the browser.
[472] And then people don't have to go buy a disc, because people are still doing that at that point in time, or download the game.
[473] They can just play within the browser.
[474] But again, like, we're idealist.
[475] The thing that kind of sucks about World Warcraft is people are killing each other.
[476] It's all about violence.
[477] Like, what if we made this game glitch about nonviolence and let people, you know, create and collaborate together.
[478] Hey, sounds great to me. Yeah, sounds, sounds great.
[479] And Stuart and TM, again, they're pretty famous at this point.
[480] The investor community is also like, yeah, that sounds great.
[481] Here's a lot of money.
[482] Yeah, we saw what you did last time.
[483] The kind of what you're doing is irrelevant, but it sounds like a sufficiently large swing.
[484] We're in.
[485] We're in.
[486] So right now the game, I will say I would kill to be in on Stewart's next gaming company.
[487] Absolutely.
[488] As with anybody.
[489] I don't think there's going to be a next gaming company.
[490] I'm calling that now.
[491] So they raise one and a half million from angels right out of the gate, including Super Angels at this point, Mark Andreessen and Ben Horowitz, who are just in the process of starting their farm.
[492] And Excel also participates in this angel round.
[493] And it is hot.
[494] Tons of people, engineers want to come work for them.
[495] People want to give them more money.
[496] In 2010, they raise a series A led by Excel, $5 million, in August of 2010, Kara Swisher flies up to Vancouver.
[497] We're going to link to this in the show notes.
[498] This is amazing.
[499] To interview store, this is during her flip cam era where she was taking a flip cam around to tech companies and just like impromptu video interviewing people.
[500] This was, yeah, I remember this because that was also the summer that I interned at Cisco when we bought them and everyone was wildly confused on why that happened.
[501] That's a good episode.
[502] It's all because of Kara.
[503] So she flies up to interview store.
[504] and get an early look at glitch and man it's crazy you can understand why nobody played this thing 2011 it's still so hot the company raises another almost 11 million dollars from the firm andreson horowitz so this is now 17 18 million in the bank for a thing that hasn't shipped and is a little suspect nonviolent MMO yeah all right interestingly though they you know the team and story they they're such celebrities they have reached really really across America, across the world for engineers and designers who want to come work for them.
[505] And so they're like, yeah, we're in Vancouver, but we'll set up a remote team.
[506] So they have employees, they have employees in New York, they have employees in San Francisco and Vancouver, all over the place.
[507] They end up staffing up to about 40 engineers and designers working on the game.
[508] And it becomes pretty hard to collaborate.
[509] These are pre -Zoom days.
[510] It's pre -anything days.
[511] they need, they need...
[512] Who knows how else you could communicate online with the remote team?
[513] They realize they need some tools to help them, you know, get worked at and collaborate remotely.
[514] So they turn to, given the, you know, hippie roots, hippie internet roots of Storten team, they turn to the natural solution to this problem.
[515] Internet relay chat.
[516] Woo!
[517] IRC, baby.
[518] It's 1988 all over again.
[519] And for folks who don't know IRC, it's a early, early internet protocol that is used to allow people to connect to a common server and using a terminal -based interface basically have chat rooms where you can directly message people, you can talk in a room about a designated topic, and you can stand this up for the open source project of your choice to be able to communicate about anything that you and other software developers are working on.
[520] You lost me at connect to a server.
[521] And that actually was the point.
[522] IRC was super cool.
[523] And still is.
[524] Still is.
[525] It still is super cool and widely used and has this really cool concept of channels that are persistent where you can have discussions around topics that are independent of users.
[526] But it's not exactly accessible.
[527] So the team at TinySpec, they start using it.
[528] And they're like, you know, we could build some really lightweight stuff around IRC to make it more accessible as we onboard new team members.
[529] for all these new engineers and designers that were hiring.
[530] They may or may not have used IRC before.
[531] Certainly the onboarding process is difficult.
[532] So they build a bunch of tools.
[533] They internally call the system line feed.
[534] And it really, you know, helps them be productive.
[535] So productive that finally, two plus years after starting the company, sensing a theme here, they finally launch glitch at the end of 2011.
[536] Big, big, you know, press, you know, Kara has done all these interviews.
[537] the tech community knows all about what they're working on.
[538] They've raised all this money from these great venture capital firms.
[539] And it's a flop.
[540] Oh, yeah.
[541] I mean, nobody uses the thing.
[542] Hundreds of people flood in to try it.
[543] Literally hundreds of people flood in to use the thing.
[544] So very quickly, they launched in September of 2011.
[545] In November, they dialed back into beta.
[546] Never a good sign.
[547] I think Stewart branded that they unlaunched the product.
[548] Exactly.
[549] And so they keep tinkering and releasing and, pulling it back into beta and re -releasing and trying to make this thing work and it just doesn't beta never -ending beta never -ending indeed and so finally a year later one night in october 2012 they've been pounding on this for a year steward has another sleepless night i don't think he had food poisoning this time he's he's got all these people these 40 people that depend on him and he's like man this is just not going to work like it's I I know we're supposed to keep trying we still have money in the bank but but I've lost faith it's it's done and he talks about it he said it was like the feeling that you get when you know you you kind of you know you have to fire someone and you just like you don't want to you want to you want to try and make it work but but that thought that feeling it's just constantly like in your stomach of your mind he said that's how he felt about the game and the company.
[550] So he wakes up the next day and he writes an email to the board and he says, quote, in typical store fashion, I do not feel that we are pouring gas on a fire here.
[551] More like pouring good whiskey on a drugstore heating pad.
[552] It is unlikely to burst into flames.
[553] And the board's like, yeah, I mean, you're right.
[554] Like we were thinking the same thing, dude.
[555] And so store it that.
[556] And after that, he calls an all -hands meeting from Vancouver, gets the San Francisco and New York teams and all the remote people on the phone.
[557] And he just burst into tears.
[558] And he starts crying.
[559] And he says, guys, it's over.
[560] You know, like, no, and nobody on the team saw this coming.
[561] They still had plenty of money in the bank.
[562] They thought they were working at this hot company.
[563] And, uh, and here's story is saying, I'm calling it.
[564] I've lost faith.
[565] We're done.
[566] And so they lay off all but the four co -founders and, and they keep four and employees.
[567] But before they do anything, Stewart's like, he's just devastated by this.
[568] They do everything they can and to get jobs for everybody else.
[569] I was going to say importantly, and we'll see this come back to be a huge asset for the company later, the way that they did these layouts was incredibly humane.
[570] And the way that they sort of took care of everyone, the sentiment leaving was I would 100 % go work for Stewart again, even though this thing completely, you know, fireballed.
[571] They take a few months, and literally the mission is get everybody a job.
[572] So they set up a website called Hire a Genius, and they create profiles for all of the former employees on the website.
[573] It's pretty cool.
[574] It's pretty cool what they do.
[575] They also upload all of the code and all of the art and design from Glitch to a server and put it in the public domain.
[576] So they're like, we've built all this.
[577] We want it to exist.
[578] I bet that's been like, what's the visual equivalent of sampling?
[579] I bet it's been used in other games.
[580] Yeah, I bet.
[581] I mean, it's up there.
[582] Anyone can go grab it, use it, do whatever with it.
[583] It's still there.
[584] And it's funny on the sidebar of the page that it's hosting it, you know, it says it's about Tiny Speck and about Glitch.
[585] And then there's like, and Tiny Speck is working on a new project.
[586] It's called Slack.
[587] Check it out.
[588] That's pretty funny.
[589] They haven't updated the page in a while.
[590] Yeah, we've thought of a new name since Line Feed.
[591] Yeah.
[592] Before you take it back to the story, so you mentioned that, email that went out to the board.
[593] So John O 'Farrell forwarded an email to Mark Andresen and the whole Andresen Horowitz team.
[594] And John O 'Farrell was the board member for the partner at Andresen Horowitz and their board member.
[595] And Mark Anderson presently replies to, I think it's the whole Andresen Horowitz partnership.
[596] It is what it is.
[597] Dot, dot, dot.
[598] If history repeats itself and they come up with the next flicker, all will still end well, Mark.
[599] Well, spoiler alert.
[600] They don't come up with the next flicker.
[601] No. No. They come up with something about a thousand times bigger.
[602] Sounds about right.
[603] Yeah.
[604] So far.
[605] Could be even bigger.
[606] So in my notes, I often break these sections out into chapters.
[607] This chapter is called the Phoenix Rises.
[608] So they still have $5 million left in the bank.
[609] And the reason that they kept four of the employees and the four co -founders stayed on together is they wanted to do something.
[610] They felt like they've been through this experience together.
[611] They had this incredible core team.
[612] And Stuart has this quote.
[613] He says, when you go through a trauma together with a group of people, you get bound together and you just want to keep working.
[614] And like, it's so true.
[615] I can totally relate to this.
[616] And what's super cool is they end up eventually being able to hire back a lot of the people that they laid off, which is great.
[617] So what are they going to do?
[618] well this is remember this is the end of 2012 something else had happened in 2012 that not many people outside the tech world took note of but lots of people in the kind of engineering world were aware of there was a company called atlasian covered on a previous acquired episode atlasian in 2012 in march acquired a little san francisco company called hip chat hip chat had become all the rage amongst developer teams.
[619] Oh, yeah.
[620] All the cool dev teams were using it.
[621] You wanted to be at a company where you could develop and talk to the rest of your team in HipChat.
[622] Yeah, I mean, it was like this will foreshadow what becomes a tiny spec here, but almost like a perk company's engineering teams were like, yeah, come work for us.
[623] Like, we're a Hipchat shop.
[624] Like, it was like, oh yeah, like these guys get it.
[625] And you're cool, they're using modern communication tools.
[626] Indeed.
[627] And so Hipchat had been launched by a bunch of young guys and they launched in 2010 within four months they had a thousand companies that were signing up mostly dev teams using hip chat and sounds like they must have used a very self -serve way of uh sounds like it well and then when they get acquired by alassian and of course the dna of alacian is developer focused tools self -serve no sales uh famously we talked about in the alacian episode there are no sales people that work at atlasian it's all just organic adoption by mostly development teams that really turbocharged hip chat and of course as we also talked about on the Atlassian episode, the major investors, not venture capital backers, but they'd done secondary deal in Atlassian, were Excel.
[628] And of course, who is the largest shareholder in Slack?
[629] Excel.
[630] Interesting.
[631] Now, we don't know for sure that all of this was on Stewart and the team's mind as they were casting about for what to do next.
[632] But it's important to note that this was happening in the background.
[633] Like, the moment was arriving for modern chat -based communication software.
[634] God, we, I mean, pulling forward a tech theme here, this seems to come up, I don't know, every other episode that there was a moment in time where the world was ready for a thing.
[635] And so multiple parties arrived at it simultaneously.
[636] I mean, you think about ride sharing, you think about social networking, think about the Instagram acquisition and all the other Instagram like companies that started right around that same time.
[637] And famously, Andreessen Horowitz, which had invested in the seed round of bourbon, which became Instagram.
[638] They did not invest in the Series A because, they were they were in pick please they were in pick please and they backed pick please man and the interesting thing about this is sometimes it's driven by a technology like mobile's taking off cameras are finally good enough it seems like mobile photo sharing is going to be a thing other times the the line is less clear on the technology that it's more sort of a social moment has arrived where people are comfortable with where technology is so you can use technology that's been available for several years to make something happen where there's a cultural understanding that that people are ready.
[639] And of course, I'm sure Slack and HipChat had to use, like there had to be some sufficient advancement in Ajax or I don't know if people were still calling it Ajax at the time, but the ability to have web pages that refreshed without refreshing the whole page.
[640] But it certainly wasn't like the iPhone shipped.
[641] And then it made possible this, you know, multi -party text communication thing.
[642] Well, I think what we've seen in our last couple episodes, here on Acquired, and we're going to see here, the really, truly huge great companies get built at the Venn diagram intersection of a technology shift, a cultural shift, and a business model shift.
[643] And HIPChat hit the technology shift and the cultural shift.
[644] They didn't hit the business model shift quite right.
[645] And so we're going to talk about what Slack does here now.
[646] So tiny spec, they realize, man, we've got this tool, line feed that we built around IRC.
[647] the interface is cleaner.
[648] It's more modern, sort of like HipChat.
[649] More importantly, it's really easy to join.
[650] Like, if you wanted to join an IRC channel, you know, you had to, like, connect to the server.
[651] Like, no normal human is going to do that.
[652] And so there are all these benefits of line feed.
[653] And they're like, kind of within a week after they got the team jobs, they were like, yeah, line feed.
[654] We're going to do that.
[655] Let's make that the product.
[656] Most pivots take much longer to arrive.
[657] Well, it's funny.
[658] When Story talks about this, he's like, it felt like an eternity that we were trying to figure out what to do.
[659] And he's like, then I went back and looked at the, history in line feed and it was a week by the end of the week we're like yeah we're doing this wow um so they go to the investors they go to excel and entries no horowitz and they're like we're going to build out you know enterprise uh what not enterprise but team chat communication and you know now in in history i think the word enterprise wasn't uttered at us to its mouth for a while yeah and um you know the the the story now is that the investors were like oh yeah how great we believe in you like it's flicker all over we'll do it again i think the reality was a little more nuance than that.
[660] Remember, Excel is largest outside shareholder in at Lassian at this point.
[661] So I think they were like more like yeah, okay, fine.
[662] But like you guys are doing productivity software?
[663] Like, okay.
[664] But fortunately, they say all right, you know, do it.
[665] We'll let our we'll let our money ride.
[666] Keep the $5 million and see what you can do.
[667] They start working on it and they have to decide what to call it.
[668] The team is like line feed.
[669] That's that's not accessible.
[670] enough.
[671] That's like the IRC of names.
[672] Exactly.
[673] Exactly.
[674] Stuart comes up with the name one night and he's like, I have it.
[675] We're going to have an acronym for the company, Slack.
[676] I love the word Slack.
[677] And the acronym is searchable log of all conversation and knowledge, S -L -A -C -K.
[678] Which fortunately they never used in any marketing material and was just known by the name Slack.
[679] And so we'll link to this in the show notes.
[680] He posted a few years ago on Twitter the actual like screenshot of the conversation with the team on line feed about this.
[681] And he's like telling everybody's idea.
[682] And I think it's Eric is like, man, dude, that's not good.
[683] How similar?
[684] Actually, I haven't clicked this yet.
[685] I'm now clicking the link.
[686] How similar does this actually look to the Slack UI?
[687] Not similar.
[688] Oh, not at all.
[689] Not at all.
[690] It's just IRC.
[691] Medallab had not been involved yet.
[692] Oh, yeah, because this is literally just text.
[693] Yeah, the line feed internally, like, it was, they'd made joining easier.
[694] And they'd done a lot of, like, this is just IRC.
[695] improvements, but there's no, there's no nice UI on top of it yet.
[696] So it really was like having a pretty big vision to see that you could turn this into what would become slack.
[697] So the team hates it.
[698] Stuart says, oh, to sell him, he says, ah, it's just a code name.
[699] Like, oh, yeah, we'll use that for now.
[700] Yeah, we'll just use it for now.
[701] And then Eric says, ah, cool.
[702] Yeah.
[703] Again, the story, you know, he's, he's, um, very creative.
[704] He's like the Captain Picard.
[705] He is, you know, he's got like real management and leadership chops.
[706] He knows how to broker influence.
[707] So he's like, yeah, yeah, just a codename.
[708] Well, then he goes off and he's like, all right, well, I'm going to see what the domain name is if this is available for Slack .com.
[709] Turns out, he, the Slack .com is registered to an electrical engineer in Wisconsin who is using the domain to host pictures of his cats.
[710] I mean, that's going to be a hard thing to pry off a guy.
[711] If by hard thing, you mean very, very, easy for like a decent amount of money.
[712] I mean, I wouldn't sell, but okay.
[713] Well, yeah, of course.
[714] I would, I would be, no, of course I would sell.
[715] They get the domain name, Slack becomes the name.
[716] They build out the product within a couple short months.
[717] Once again, you know, echoes back to the Flickr days.
[718] Yep.
[719] They bring on, really importantly, Medallab.
[720] And aren't, do I have this right?
[721] Metalab is a Victoria -based company, right?
[722] They're based in Victoria, yeah.
[723] So I finally just put that together, why they would be the...
[724] Yep.
[725] And Metelab is a UI design firm.
[726] really, really truly world -class based in Victoria.
[727] They had done some cool productivity apps of their own.
[728] I think was it flow?
[729] They had a GTD or like a to -do list tool at one point.
[730] Ah, he's competing with you and sees the day.
[731] Yeah, theirs was very nice.
[732] I'm sure yours was nice too.
[733] Maybe not quite as nice.
[734] Thank you.
[735] But yes.
[736] So they bring on Metal Lab to redo the UI, and the UI is so different from anything anyone's ever seen.
[737] These guys do tiny now, right?
[738] Is that that's the same group?
[739] It's a basically small -cap private equity shop that owns really, really design -focused, beautiful, cash -flowing technology products.
[740] So, like, Dribble.
[741] They bought Dribble.
[742] Ah, cool.
[743] Cool.
[744] Cool.
[745] Everyone, I assume, knows the Slack UI, because hopefully you're in our acquired Slack.
[746] If you're not, come to go to Acquired .com and sign up to join.
[747] It is accessible to the normal person to a degree that, you know, even HipChat can't come close to.
[748] You join and Slackbot welcomes you.
[749] And just by interacting with Slack.
[750] you get a feel for how you're supposed to use the product.
[751] Yep.
[752] So mid -2013, again, after just a couple months, they're ready.
[753] They're ready to take it out into the world.
[754] And they go out to some of their friends and other companies, try and convince them to try it.
[755] And it turns out they had kind of a hard time because people are like, what is this?
[756] Why should I use it?
[757] I use Hipchat.
[758] So I have my own story about that.
[759] I mean, when this launched and I remember seeing on like hacker news and this was like a darling of Silicon Valley, check out this new cool.
[760] It's the new way to work.
[761] I mean, people were building it as like, this is the new way that teams work together.
[762] I'm like, okay, well, I got to give a shot.
[763] And I'm working at Microsoft at the time.
[764] And I go, I sign out for it.
[765] I make like whatever my obscure team name at Microsoft .com was.
[766] Like, I'm sending it around and people are, A, reticent to join, B. I'm on link.
[767] I use yammer.
[768] Yeah.
[769] It's like, what is the, so it's, it's link.
[770] Like, it's replacing chat or is it replacing email?
[771] And I'm like, oh, it's sort of in the middle.
[772] It has all these integrations.
[773] But like, it's actually being the first person trying to sell it to your team, if your team wasn't already using some type of sort of ever persistent chat, it's actually kind of a hard thing to get people to understand why it would be useful.
[774] Yeah.
[775] So through the summer, they work with a couple, you know, friends, companies to land this messaging, and they figure out, they get really good at explaining why, because most people have never used something like this before, what it is, why you should use it, why it helps, why it's better than email, and more importantly, just making the onboarding process, super, super easy and fast.
[776] So the first decent -sized company they get to use it is Ardeo.
[777] Ardeo is about 120 -person organization.
[778] They sign up.
[779] First, it's the engineering team that starts using it.
[780] Then it starts to spread.
[781] And pretty quickly, the whole company's using it, and they're addicted.
[782] And from these RDO and a couple other companies, what happens is the companies start evangelizing.
[783] And all the employees of the companies start telling their friends at other companies, hey, we've switched to this thing called Slack.
[784] It's amazing.
[785] It's changing the way we work, and that starts to build a ton of buzz.
[786] And so...
[787] Which I'll say is interesting, it's that popping out to Playbook for a second, it requires a word of mouth, not usage, because there is an intra -company network effect with Slack where you're inviting all your colleagues, but there's not a cross -company network effect.
[788] Whereas you take the flip side of that, like Zoom, there's a cross -company network effect, because to communicate with someone at that other company, you need to click the Zoom link to join and then you're a user.
[789] With Slack, it has to be that I'm getting so much utility out of it at my company and it's grown within my company that now I go and tell somebody at a different company about it.
[790] And importantly, we'll talk about this in a minute.
[791] Slack and Stuart and the team embrace this fact and they make, they realize that Twitter is their friend here and people tweeting about how much they love Slack.
[792] And then the Slack team amplifying those tweets is a big part about generating buzz and building up demand.
[793] So they officially launch Slack does in a preview release to the public in August 2013, again, less than a year after making the pivot, on the first day that they launch, 8 ,000 companies sign up for the wait list.
[794] I mean, this is incredible.
[795] This is how much buzz had already been building, how much of a celebrity started team were.
[796] Yeah, exactly.
[797] And then within two weeks, there's 15 ,000 companies that are signed.
[798] up, or teams that are signed up on the wait list to use it.
[799] They pretty quickly launch paid, and I alluded to business model in a minute, a minute ago with the paid plan, and we're going to talk about the freemium model, within a couple weeks of launching paid, they're at a million dollars in ARR.
[800] And like most SaaS companies, you know, takes a year, two years to get to that level.
[801] They're there in weeks.
[802] By February 2014, they have 10 ,000 new users signing up a day.
[803] These are not companies, but users within the companies, signing up a day.
[804] They have 135 ,000 paying users, and it's even at large enterprises.
[805] So at this point...
[806] That's what, like 10 bucks a seat?
[807] No, five bucks a seat.
[808] It's seven or eight bucks a month percy, or $80 a year or $8 a month, I think.
[809] And what's cool is they're not trying to target the enterprise yet, but they have all these huge organizations.
[810] They're teams that have organically adopted, like you did at Microsoft.
[811] They said at this point in time, February 2014, there are nine different teams at Adobe that are all paying users of Slack.
[812] Whoa.
[813] It's funny, you could, there was a little security exploit for a while where you could figure out what the names of teams at a company were because Slack had built this really ingenious user -friendly feature that actually exposed information companies may not have wanted to at the very beginning.
[814] I remember trying to sign up with an email address because the way that they made sure you were at the company is by looking at your domain name.
[815] So I'm like, okay, like, Ben at Apple .com.
[816] Like, let's see if they'll let me sign up as an Apple employee and join the Apple Slack.
[817] And, of course, it doesn't.
[818] But what it does say is, hey, your company already has eight teams using Slack.
[819] Here's the name of each of those slacks.
[820] And I was like, oh, my God, like, I'm learning the names of.
[821] iPhone 10, code project at Apple Slack.
[822] What is P42X?
[823] Project Purple?
[824] No, I'm making something up.
[825] But, yeah, it just goes to show you, though, like how well thought through it was for, okay, someone at a company is going to try and join because they're just poking around and figuring out, does my company have an account or not?
[826] Or how does that even work?
[827] Because this was a different way than software was sold and adopted before.
[828] Let's make it easy for people to find the team that they're looking for within their company once we know they have that email address.
[829] Yeah.
[830] So on the back of this, April 2014, there raised $43 million in a series C led by Mamun at Moon at Moon.
[831] at social capital.
[832] He's now at Kleiner Perkins, which is an amazing venture investment.
[833] So let's take a step back here.
[834] I talked a minute ago about the Venn diagram of clearly technology shift, cultural shift, these are happening.
[835] That's the background for Slack and hipchap before it.
[836] But the third piece, the business model, what Slack did that was so brilliant and is just the parallels to Zoom are perfect, is they nailed how to structure.
[837] the freemium model for this product and service.
[838] So HIPChat, when HIPChat launched and everything else, certainly all the Microsoft products and whatnot in this chat -based collaboration space, the freemium gate was number of users.
[839] So HIPChat was free for up to five users.
[840] Once you had more than five users, you had to pay.
[841] And of course, that makes sense.
[842] If you're taking an enterprise approach to things, like, yeah, you're charging on a per se basis.
[843] Like you should, you know, let people try it a little bit.
[844] You run a tried and true playbook there.
[845] but gate the real value.
[846] Stewart and Team, you know, they came from the consumer world.
[847] They came from the gaming world more than that.
[848] And when you're trying to get people to use, to play games online, and a lot of them are freemium, you want to encourage usage, right?
[849] You want to encourage usage, and you want to show as much of the value of the product to as many people as possible without having the gating item of paying.
[850] And if you think about the freemium gaming world, too, they sort of notoriously only monetize 1 % and the vast majority of people who are using it are using it at basically its full functionality, but never paying.
[851] Yes.
[852] So what do they do?
[853] They set the freemium gate not on number of users.
[854] Slack is free, as evidenced by the acquired Slack.
[855] We have thousands of people in there.
[856] Not paying a dime.
[857] Not paying a dime, and we probably never will.
[858] And it's full Believe me, it's good marketing for Slack.
[859] But it's full featured.
[860] It's the full Slack product.
[861] And the pay gate is for the number of messages in the recent archive that you can search and access.
[862] So a relatively smaller feature.
[863] But if you're a workplace...
[864] Oh, yeah.
[865] I can't imagine not having that at the office where it's like, we definitely talked about that in Slack like eight months ago.
[866] What was it?
[867] And certainly there's that use case.
[868] There's also the new team members are joining.
[869] They're not up to speed on everything going on.
[870] They join all these channels.
[871] And then they're like, oh, okay, I want to find out what's going on on my team or what's going on on this project.
[872] If you can't go back and look at the archives and search the archives, that's impossible to do.
[873] So it's a really important feature for work teams to be able to do this.
[874] And so Slack uses that as the freemium gate.
[875] And that's huge.
[876] And I think there's some other more sort of admin -y -type features that requires pay as well.
[877] And a number of integrations that you can do, you're capped at 10 integrations, third -party integrations on the free tier.
[878] So other important stuff.
[879] But the main thing is that they allow people and teams to come on for free and get real work value out of the product, the full product, without having to pay.
[880] And that is just super, super critical.
[881] and nobody was thinking about enterprise monetization in this way at the time.
[882] And honestly, it's, I fully believe it is because of the background of Stewart and the team that they brought this approach to the market.
[883] Like if they had nobody from, even at Lassian, but certainly not Microsoft or, you know, anywhere else was going to think about freemium tier construction this way.
[884] This is how we should do a productivity application.
[885] And this is another good thing that we've seen across a lot of episodes.
[886] for a lot of this disruptive innovation, you need to smash together the learnings and best practices from one corner of the world into a different corner of the world that is not used to thinking about things that way.
[887] And to be able to, you know, say, what would it look like if a team of idealistic game designers, you know, built productivity?
[888] Yeah, like, of course it's going to look different than a shop like Atlassian that's only ever built enterprise collaboration software.
[889] And of course, yes, in the business model, which I think you're right.
[890] I hadn't thought about it before, but in this classic question of why did Slack win when HIPChat was already a thing, I mean, there's UI reasons why, you know, Slack appealed to non -engineering teams.
[891] And Atlassian sort of always said, oh, anyone can use HIP chat.
[892] But like, if you really, that was more lip service than it was actual product work.
[893] But Slack was truly from day one designed to be something that anybody in the organization could use and not just engineering teams.
[894] Well, and I think it's also for other teams within organizations that weren't used to this category of software using anything like this, the way they structured, the freemium model made it accessible for those organizations to try to use it.
[895] Even if Slack were designed to the pixel the same way, but you had to pay for more than five people to use it, there's no way that the New York Times Newsroom would have tried it.
[896] You know, it just, by removing that barrier and they were really thoughtful about this and Stuart from the beginning was talking about like they were creating a category when you're creating a category you need to take this approach there's a lot of education you need to do your product has to do education your marketing has to do education yeah by the way the New York Times Newsroom now uses a home rolled slackbot that they created to manage the approval workflow for the push notifications that go out.
[897] At one point, I tweeted something like, what a killer job to be the person that approves the New York Times push notifications.
[898] And a New York Times employee sent me a thing that's a cool medium post on here's how that actually works in Slack and how people go back and forth and edit and make suggestions and then submit it for approval into the queue.
[899] That was so cool.
[900] I think we talked about it on the Elass in the episode.
[901] But that was the moment for me and I think a lot of people in tech when we're like, oh, wow, the New York Times newsroom is using this like this is this is way more than hip chat this is something that is like has the potential to be as big as as slack has become so on the back of this at the end of 2014 they raise a 120 million dollars at a over a billion dollar valuation led by gv and incliner perkins and it's just off to the races since well well since day one when they launched out of at a preview The beginning of 2015, they're adding a million dollars in paid ARR every month, and growth just keeps accelerating from there.
[902] Super interestingly, in July 2018, we didn't do an episode on this, but we could have done a follow -up.
[903] Slack ends up buying HipChat.
[904] So within Atlassian, Atlassian had rebranded HipChat to Stride.
[905] I think, wasn't that a different product?
[906] Yeah, they rebuilt the product, and then they sunsetted Hip Chat.
[907] They still had Hip Chat.
[908] But they were transitioning everyone to stride.
[909] Also stride.
[910] Like, can you come up with a name that's more different than a couple letters from Slack?
[911] I mean, when that happened to people, you know, Slack must have just been high -fiving one another.
[912] Not like, we've won, it's over.
[913] And indeed, probably one of the smartest things that Lassian has done is they recognized that quickly, too.
[914] So they sold essentially those users to Slack.
[915] In like one of the best deals in history ever negotiated, Stewart managed to get all the users.
[916] What was the, like, equity arrangement that happened?
[917] I don't remember exactly how much equity, but, um, but yeah, Lassian got a, got a small amount of equity in Slack for it.
[918] Yeah.
[919] I mean, that, that, that was a highly triumphant moment for, uh, steward as a, and you have to imagine, probably Excel helped there, given that they were major shareholders in both companies.
[920] And then late 2018, following that, rumors start circling that Slack is considering a DPO for their public offering.
[921] Much like Spotify or, you know, just before them, 30 years before Ben and Jerry's.
[922] Indeed.
[923] Barry McCarthy, Garty.
[924] We referred to Barry at the top of the episode, Acquired superhero, former Netflix, CFO, instrumental in Netflix's history and then CFO, and then CFO, Spotify, and engineered their direct listing.
[925] Yep.
[926] Inspired by Ben and Jerry's.
[927] And I think not at all, but it is a fun piece of trivia.
[928] Yeah.
[929] Super fun piece.
[930] trivia.
[931] Maybe, maybe Barry had a fever dream one night from eating too much ben and juries.
[932] Yeah.
[933] Let's real quick go through what the differences are in an IPO and a direct listing are because I think it's worth, David, you and I have talked about this a few times.
[934] Listeners, we are extremely excited about the future of direct listings.
[935] And I think, at least I personally feel like it is a more modern way to go public than a sort of IPO process that just has tons and tons of baggage.
[936] to it, and not every company can do it because, well, actually, let's talk about why not every company can do it.
[937] Well, the IPO process was architected for a different era when venture capital was a cottage industry, and now venture capital and private market financings are an enormous global, highly professional, highly professional worldwide industry where that money managers of all sizes participate in, and so companies like Slack can raise, you know, billions of dollars in the private markets and grow to have a global brand grow to have a global brand in a way that just wasn't possible before so you needed when you were going public you needed to do an initial public offering where you were selling stock doing a financing event to do that because you needed that was your the IPO was your way to raise real growth capital that's no longer necessarily the case so for companies like slack honestly like Zoom although Zoom did a traditional IPO but companies that are profitable that don't need to raise profitable or close to profitable, that don't need to raise huge amounts of money.
[938] We should be clear.
[939] Slack is not profitable, but does have $800 million of cash on the balance sheet and has great unit economics.
[940] Yeah, capable of becoming profitable, making a choice to invest significantly in marketing.
[941] So they are not profitable at the moment.
[942] They don't need to raise that money.
[943] A DPO has a lot of advantages.
[944] Yep.
[945] In a DPO, the only, the only, the only, shares that start trading are all the existing shares.
[946] So there have to be selling shareholders rather than, you know, there's now 10 % of new liquidity created by us creating a bunch more shares in the company.
[947] So there's no dilution to the company.
[948] There's no, no dilution.
[949] Of course, because of that, they don't actually raise any cash.
[950] It's just investors and money and employees taking money off the table.
[951] But there's, you have to imagine.
[952] There's, and there's one other big benefit for the ecosystem as a whole and for employees of the company is there's no lockup when you do the yes so as I was going to say that the sort of butt to this whole thing is gosh it seems like there might be a lot of volatility involved in this so there's no lockup period employees can sell right away great benefit to employees maybe a little bit dangerous to the signaling for a company so how do you manage this well one thing is and as much as we'd like to say in this era of the direct listings the investment bankers are no longer making a big payday they're out of the process.
[953] The investment bankers are never out of the process.
[954] Yes, they move pure - It's like a law of thermodynamics.
[955] Yes, they move from a role that is underwriting, which is saying, I commit to buying 50 ,000 shares in this range that you have quoted me at, you know, approximately this price the night before the IPO before it starts trading.
[956] Maybe I'll get a pop.
[957] You know, it's this underwriting process that occurs where you commit to buying some amount of the shares that are being created.
[958] they now move into an advisory role where they say okay well why don't you pay us an amount of money it's still going to be tens of millions of dollars and we are going to on your behalf go and work with all existing shareholders to see who's going to sell and who's not and sort of at what price and we're going to come back to you one with a guidance of what we think your suggested initial trading price should be but again trades just start happening instantly and there's a sort of market making function that happens there where they're going to trade where supply and demand interest.
[959] intersect.
[960] And this is the real crucial part of sort of what they help you do there is say, look, we talk to these set of investors.
[961] Each one of them have committed to us, your advisory firm, to selling this number of their shares.
[962] So we can guarantee when you start trading that, you know, I don't know if this is true or not, but Excel will be selling 25 % of their shares.
[963] So that will be available.
[964] There's nice liquidity in the market for your new shareholders to have something to buy.
[965] It's organized.
[966] It's not just chaos.
[967] Yes.
[968] But it is, you know, much more sort of, you can think of it as like algorithmic rather than this previous process of, you know, lick your finger, put it in the air, see which way the wind is blowing and say, 24 shall be the amount for this share price.
[969] Well, to be fair, that IPO prices are set based on order books that are built during the road show process.
[970] But another important benefit for DPO's.
[971] Spoken like a true former investment banker.
[972] Of course, I have to defend.
[973] Another important benefit for the DPO's versus the IPO process.
[974] process is if you are a company that has already raised capital privately from many of the traditional public market mutual fund hedge fund type investors that would be large institutional buyers in an IPO of the new shares created, you run into a problem like what Uber faced when they were doing their IPO where all of those sets of buyers were already shareholders in Uber.
[975] They already had the exposure they wanted to the company.
[976] Their incremental appetite for new shares was relatively low.
[977] They wanted to be selling, not buying.
[978] And so when you do a DPO, you solve for that problem if you have it.
[979] Now, of course, Uber had to do an IPO because they needed the cash.
[980] Yeah, that's the other thing.
[981] So to do a DPO, one, you have to not need the cash like companies like Uber and Lyft do.
[982] Two, you need to already be a brand that doesn't need a roadshow process.
[983] Number one, so you don't have to go make all the bankers aware of your brand so that someone will commit to buying shares.
[984] Two, you don't need those banks to then go and get all of their retail investor clients excited about your company.
[985] So Slack, you know, Spotify made sense because they were a consumer company.
[986] One of the magical things about Slack is they've managed to turn this, I mean, it's enterprise software, and it has a consumer -like brand to it.
[987] And it has an awareness where you don't have to go and do a whole bunch of marketing around, you know, here's why you would want to buy this stock.
[988] It already had enough excitement around it.
[989] I mean, the story has, I believe, multiple times been on the cover of Newsweek.
[990] Why is Newsweek relevant?
[991] That's good.
[992] Well, it used to be relevant.
[993] It used to definitely be relevant.
[994] But it's a good sign that, you know, people know about this thing.
[995] So, well, in April of 2019 this year, Slack announces they are going to do a DPO.
[996] On April 26, they file the S -1.
[997] And then last week, Wednesday, June 19th, they set the reference price in the evening.
[998] The bankers do have $26 a share.
[999] on Thursday, June 20th, the company begins trading, closes up nearly 50 % to 3862 at the end of the first day of trading.
[1000] And again, this is not, the banks aren't supporting trading.
[1001] This is just free market trading happening here.
[1002] One of the ways in which we should be grading this later on in this episode is how volatile was it?
[1003] Like, did they manage to keep a relatively stable share price or, you know, was it just flying all over the place because the, the, um, there was no underwriting?
[1004] If by stable, you mean it went up very quickly.
[1005] So they close the day at $3862, which is a $19 .5 billion market cap.
[1006] Wow.
[1007] Yep.
[1008] Wow.
[1009] All right.
[1010] Should we talk about narratives?
[1011] That's awesome.
[1012] We should definitely do that.
[1013] We should also talk about sort of who owned the company at that point, just because I think it's kind of an interesting, it's interesting to dive into this a little bit.
[1014] Obviously, you've talked about Excel.
[1015] Excel owned close to a quarter of the company.
[1016] Incredible.
[1017] Which is highly atypical.
[1018] I mean, usually venture firms will own maybe 10, at IPO if you let a round and did north of 20 % at that round.
[1019] And so, you know, what's very clear here is Excel kept participating in a big, big way in all future rounds.
[1020] I believe they were in all the fundraising rounds that were happening along the way.
[1021] I believe they were actually not just not just doing pirata, but buying incremental ownership, I think, in a lot of those rounds.
[1022] That's a conviction bet right there that's paying off in a huge way for them.
[1023] Yeah.
[1024] Andrews and Horowitz owned 13 % social capital 10, soft bank with seven, which is around the same percentage that Stuart Butterfield himself owned so Stewart is now at least on paper a billionaire better than the outcome from Flickr I would say yes yes but it's also interesting to see all the other Silicon Valley CEOs who have been angels and I think this was reported by Forbes in a great piece recently but Jeff Weiner Patrick and John Collison CEO of Squarespace Biz Stone former CEO of Yammer David Sacks Jeremy Stoppelman of Yelp Yelp of Yelp Yelp So this had a kind of an amazing cadre of...
[1025] Yeah, I think it was from that angel round, the $1 .5 million.
[1026] Oh, is that what it was?
[1027] I believe so.
[1028] Wow.
[1029] And there's a lot of these guys I know also came in later on in sort of the low hundred millions and low billion dollar valuation rounds as well.
[1030] Just interesting to sort of know who is a big part of this, I guess, big liquidity event.
[1031] Yeah.
[1032] Well, it also speaks to, you know, we've made this point so many times on the show, especially because Stewart and Company and the...
[1033] co -founders and history here goes back to the early internet day or at least the well back to the web 1 .0 but but really back to the web 2 .0 days the world was just a smaller place back then like all these same people like they all that's like a little club you know they're all like the world was just not that big silicon valley was not that big a lot of them worked at yahoo together the yahoo mafia is strong all right so before narratives let just some facts to know about the company from their financial position when they went public.
[1034] So I mentioned they had $800 million in cash on the balance sheet.
[1035] They're still growing at 82 % a year.
[1036] So they're close to doubling year over year now as a public company.
[1037] They're generating in the fiscal year ending in January of this year.
[1038] They generated $400 million in revenue.
[1039] So previous year before that, just over $100 million in revenue.
[1040] So company's still doubling.
[1041] That's close to in line.
[1042] It's a little below.
[1043] triple double double.
[1044] Yep, a little in line with Zoom.
[1045] So Zoom's growing a little faster at close to 120%, whereas you've got Slack here around 80%.
[1046] But it's fun to kind of keep comparing to Zoom.
[1047] Zoom has a little less revenue.
[1048] So 330 million in revenue versus Slack's 400.
[1049] But they are cousin companies in a lot of ways.
[1050] Absolutely.
[1051] All right, listeners, our sponsor is one of our favorite companies, Vanta, and we have something very new from them to share.
[1052] Of course, you know Vanta enables companies to generate more revenue by getting their compliance certifications.
[1053] That's SOC2, ISO 2701.
[1054] But the thing that we want to share now is Vanta has grown to become the best security compliance platform as you hit hypergrowth and scale into a larger enterprise.
[1055] It's kind of wow.
[1056] When we first started working with Vanta and met Christina, my gosh, they had like a couple hundred customers, maybe.
[1057] Now they've got 5 ,000, some of the largest companies out there.
[1058] It's awesome.
[1059] Yeah.
[1060] And they're They offer a tremendous amount of customization now for more complex security needs.
[1061] So if you're a larger company and in the past, you showed Vanta to your compliance department, you might have heard something like, oh, well, we've already got a compliance process in place and we can't integrate this new thing.
[1062] But now, even if you already have a SOC2, Vanta makes maintaining your compliance even more efficient and robust.
[1063] They launched vendor risk management.
[1064] This allows your company to quickly understand the security posture of the vendors that you're choosing in a standardized way that cuts down on security review times.
[1065] This is great.
[1066] And then on the customization front, they now also enable custom frameworks built around your controls and policies.
[1067] Of course, that's in addition to the fact that with Vanta, you don't just become compliant once, you stay compliant with real -time data pulled from all of your systems, now all of your partner's systems, and you get a trust report page to prove it to your customers.
[1068] If you click the link in the show notes here or go to Vanta .com slash acquired, you can get a free trial.
[1069] And if you decide you love it, you will also get $1 ,000 off when you become a paying customer.
[1070] Make sure you go to vanta .com slash acquired.
[1071] So, moving into narratives.
[1072] Yeah.
[1073] Should we talk bowls first?
[1074] Yeah.
[1075] What, um, why do people love this company going into the IPO?
[1076] Well, I mean, there's a lot to love here.
[1077] There's a whole lot to love.
[1078] You know, the dynamics, similar to what you said of the comparison to Zoom, this is a product that Really, Zoom and Slack represent the new wave of go -to -market in the enterprise of products that are so good that they sell themselves, a freemium business model that is perfectly tailored to maximizing reach and acquisition and has the conversion markers set at just the right places to optimize.
[1079] there's extremely well managed.
[1080] The growth is incredible.
[1081] I mean, there's, like I said, a lot to like.
[1082] Yep.
[1083] And to get into some of the numbers a little bit there, one of the biggest things is crazy strong retention.
[1084] Yes.
[1085] So if you sort of model this out, it looks like only about 10 % of newly acquired paying customers churn in their first year, so over 90 % retention on paid customers.
[1086] And that if you look at a five -year window, about 80 % of customers that you pay to acquire are still around and are likely paying more.
[1087] I mean, that is.
[1088] Just crazy stickiness, great upselling.
[1089] That is truly world -class.
[1090] It's amazing.
[1091] The other thing on top of that, which is a secondary metric that falls out of that, is their cact -lTV ratio or the cost to acquire a customer versus the lifetime value of that customer.
[1092] And again, these are paying companies as the customer is about 13 times.
[1093] So while it costs them a lot to acquire a customer, and I'll talk about this in Bears, it costs about $8 ,000 to get a paying customer that more than pays for itself sort of in the long run of how long that that customer will be with you.
[1094] Another thing that we have to bring up here is Microsoft launched Teams.
[1095] Slack took out a full page ad in the New York Times when Microsoft launched Teams and said something along the lines, welcome to the party, you know, the water's warm, come join us, we've pioneered this thing.
[1096] Apple doing the same thing many years ago.
[1097] Yes, which was both lauded and widely criticized, but a baller move in some form.
[1098] Microsoft is very seriously competing here.
[1099] They've basically shifted their enterprise product strategy to say, hey, if you're using Office 365 and you're using OneDrive, you know, Microsoft Teams is the hub where it looks just like Slack, like you communicate with your people in here and then everything is sort of shared and mapped through here, and this is the way that you experience, you know, the rest of the Microsoft Enterprise Productivity Suite.
[1100] So pretty serious head -on competition.
[1101] The thing to like, if you're sort of a bull on Slack, is that, yeah, sure, Microsoft has teams, but that's more about Microsoft leveraging all their existing customer relationships and creating a Slack -like experience for its existing customers rather than the theoretically much broader set over the next 10, 20 years of people who aren't currently Microsoft customers and will be looking for a solution.
[1102] like this, Slack very likely wins those customers.
[1103] Yeah.
[1104] Well, and Stuart actually had a quote about this in the very beginning days of Slack, the product after the public launch.
[1105] He called it, I think, something like the 15 -year slow -motion unbundling of Microsoft as the Productivity Suite for the Enterprise.
[1106] And absolutely, that story is great if you buy into using all of Microsoft's services.
[1107] Yep.
[1108] Which work very well together.
[1109] Very, very well together.
[1110] But I think very few organizations, and especially new organizations, do that today.
[1111] Now, there may be a story when we get into Bears about, you know, the world, the famous quote that we talk about here all the time of the two ways to make money in business, bundling and unbundling.
[1112] We've been on a 15 -year unbundling journey in productivity that maybe it has now reached its apex with Slack, with Zoom, with our next episode that we're going to cover here on Acquired.
[1113] And maybe the time is right to begin the re -bundling, but for the moment, you know, a lot of people like the unbundled world.
[1114] Don't you bring a bear into here this bull section?
[1115] That's actually a very interesting quote for another time is, or a very interesting ledge to jump off of when will the re -bundling start occurring and what is the point of integration where we will start rebundling around.
[1116] One more thing in bowl before we move to bears is like Salesforce, their ecosystem, Slack's ecosystem and app platform, makes them very stick.
[1117] So once you really start to, you know, loop in a lot of these integrations, which Slack has done a very nice job of building from the very early days, you know, you really start to feel like we could move off this thing, but gosh, it kind of all works really well together.
[1118] And I think that's going to continue to be a story for them.
[1119] There's integrates.
[1120] But I think there's also even more important than that is just the, all the persistent archive of all of your team's work over the last.
[1121] Like, we would never leave Slack at Wave.
[1122] Like, we, well, we couldn't.
[1123] Even though we're just.
[1124] a three -person organization.
[1125] Yep.
[1126] Yeah, that's a great point.
[1127] All right, so Bears.
[1128] The company is doing well, but...
[1129] Not that well.
[1130] I think we thought before seeing the financials, or at least I thought, I was like, this could be the first profitable IPO of, well, the DPO, public offering of the year.
[1131] Yeah, no, it turns out actually they're losing quite a lot of money.
[1132] They are losing quite a lot of money.
[1133] And the phrase that, I can't remember where I saw it, but I think this sums it up very nicely, Zoom was actually the Slack we thought we had all along.
[1134] Now, this may be a little extreme.
[1135] I fully do believe that Slack is making the choice to invest heavily here.
[1136] Yes, here's why they're not profitable.
[1137] So Slack actually has a two to three year payback period since it's such a low price point.
[1138] And the thing that you kind of don't like about this is that it's backloaded.
[1139] So, you know, people are paying more in the later years of their customer lifetime.
[1140] So it will continue to incur losses building out its customer base here.
[1141] You land on an organization or a team, a few people join, maybe they don't pay for a while, then they start paying.
[1142] But then the team grows, more people join, more people join the company.
[1143] Then you start paying more per user.
[1144] Yep, absolutely.
[1145] Another thing to be bearish on here is that Slack's net dollar retention.
[1146] So this is the sort of beginning of period revenue plus your upgrades minus your down.
[1147] grades and your churn, all sort of divided by the beginning of period revenue.
[1148] So you're basically saying, how much does the next quarter or years revenue that comes from this set of customers, like what percentage of next quarter's revenue comes from the revenue that you currently have?
[1149] Yep.
[1150] It's still world class.
[1151] It's still one, 20, 30, 40 percent, something like that, but it has been declining.
[1152] And so there is some amount of, okay, churn, there's pay churn that's actually happening.
[1153] Yeah.
[1154] And it's either, is it people ditching it for Microsoft Teams?
[1155] Is it people saying, it's actually maybe disruptive to my workflow?
[1156] An email might be better.
[1157] So, you know, I don't want to be too bears here, still world class, but also declining.
[1158] And so you sort of want to see that bottom out soon.
[1159] Yep.
[1160] And then I think the last piece is competition, which you could have different views on.
[1161] I mean, I lean towards the, I don't think re -bundling is happening anytime soon, or at least not in a big way towards Microsoft.
[1162] But Teams is a viable competitor.
[1163] Very much so.
[1164] Very much viable competitor.
[1165] Ben Thompson's written a lot about this.
[1166] There are other competitors that are popping up.
[1167] Interestingly, as I was doing research, it seems a lot of people are starting to use Discord for work.
[1168] Whoa.
[1169] actually, yeah.
[1170] By the way, did you know also Facebook, Facebook work has like a ton of organizations?
[1171] I don't know how many are active, but like it's a very, very widely, like, activated product.
[1172] Wow.
[1173] Yeah.
[1174] That is shocking.
[1175] That is truly shocking.
[1176] Completely agree.
[1177] Please, if you are using Facebook for work.
[1178] Let us know.
[1179] Stop.
[1180] Acquired FM to email.
[1181] Come on.
[1182] All right.
[1183] So the net here is it's going to take a lot of cash and a lot of time to get profitable.
[1184] but as long as their efficiency of marketing spend and low churn continues investing in Slack, you can see why Slack is plowing these marketing dollars in and it's a great long -term use of capital.
[1185] And I thought one good quote to put a pin on this Bulls and Bear section is from my colleague Ben Rush at Pioneer Square Labs.
[1186] We were slacking about this last night.
[1187] And he told me, well, basically right now the way I look at it is it's a machine where a dollar in gets you close to $10 out over eight years, so you'd be a fool to not keep putting money in unless you believe that something fundamental is about to change to make them much, much less sticky.
[1188] And you're basically saying, look, I know it's going to take a long time to pay back, but boy, is it going to pay back in a big way.
[1189] That's why we're seeing the company right now that has such great indicators across the board in sort of their unit economics, their stickiness.
[1190] And yet it's still not a profitable company, just going to take some time.
[1191] Yep, yep.
[1192] Well, should we talk about what would it happen otherwise?
[1193] We could have had like the coolest game ever.
[1194] Oh, man. It actually was among the course that, like I used to work with a group that was like very, very hardcore gamers from, would basically try every single game and knew a rich history.
[1195] And there was a lot of buzz in the gaming community about what glitch was going to be and about the promise that it held.
[1196] It's so interesting Canning game be built?
[1197] I mean I guess Minecraft is the perfect example But Roblox Yeah, Roblox Like games where people don't kill each other Yes But it's like you look at Fortnite And like I kind of like If you take a step back from like killing each other Fortnite is a great example of this It's just a great gameplay mechanic Like and if you take the violence aspect out of it Like it's just hard to top that competition based You know if you think about it more like a sport than you do as, like, violence.
[1198] It's really hard to find any other mechanic that can top that.
[1199] Yep.
[1200] Yeah.
[1201] I think the more interesting, what would it happen otherwise is, what if this team hadn't brought this approach to this market?
[1202] Would somebody else have figured out how to do this?
[1203] I mean, yes, because you look at what Zoom figured out in the first 40 minutes of your meeting are free.
[1204] Yes, I absolutely think that someone else would have figured it out, but it may not have been done in such a, it may have been done later, it may not have been done in this exact way.
[1205] They may not have gated it on sort of number of messages.
[1206] I don't think Atlassian would have figured it out, though, because their DNA is just too...
[1207] Also, they had a product in market, that it would have been too disruptive to their existing product.
[1208] It would have been something.
[1209] Maybe we would all be using Discord for work.
[1210] Yeah.
[1211] Oh, that's actually, there's a chance that that's the most reasonable conclusion to draw here.
[1212] Yeah.
[1213] Wow, that's like a super hedged comment.
[1214] I should say, I'm in on that.
[1215] David, I'm in on your prediction there.
[1216] Well, you know, the gaming, DNA?
[1217] Yep.
[1218] Playbook.
[1219] Playbook.
[1220] So one thing I wanted to talk about here that we didn't really cover in the history and facts is one other discipline that the team brought from the gaming and consumer world to Slack is really a deep understanding and kind of analytics around the funnel of a user journey from, you know, acquisition to conversion to activation to retention.
[1221] It is no accident that Slack has world -class metrics around this.
[1222] And it's because this discipline is so important for gaming and online gaming.
[1223] Yeah, famously, the PMs at Zinga, who sort of pioneered this approach, were not the creative types.
[1224] They were the spreadsheet types.
[1225] They were the former bankers, the analytical types.
[1226] One thing that they realized really early on, again, you know, PMs at gaming companies look for these moments, switch or what is what is the signal when a user hits some threshold that they are then going to retain for a long period of time or what's that what's that magic or they're going to monetize or whatever what's that magic thing and then they optimize the game to funnel users to that magic moment slack realized pretty early on that that that magic moment is a team that has sent 2 ,000 messages back and forth which sounds like a lot but if you've used slack you know that like that happens real quick they figured out based on teams team size how long it takes for that to happen.
[1227] So if you're a relatively smaller team, it takes X amount of time.
[1228] If you're a relatively larger team, it takes like less than a day for that to happen.
[1229] They found that 93 % of teams that hit that 2 ,000 message threshold didn't turn, retained indefinitely.
[1230] And so they started architecting the product around doing everything to get new teams that were onboarded to that threshold as quickly as possible.
[1231] That makes sense.
[1232] Super, super interesting.
[1233] And again, this is like, think back to our Zoom episode in the world of enterprise software before this.
[1234] Nobody was thinking this way at Cisco at Microsoft at Salesforce.
[1235] This is completely new.
[1236] How can I take the CIO to a new dinner he's never experienced?
[1237] Exactly, exactly.
[1238] This is also a perfect example of a friend of the show, Alfred Lynn, at Sequoia.
[1239] He had a great thing that he said to us at Wave a while back, which is that what he really looks for in companies is, is do they understand, he called it input metrics as opposed to output metrics.
[1240] Like the output metrics are, what is your retention?
[1241] The input metrics is what is going to get you.
[1242] What do you understand about your product and your business is going to get people to retain?
[1243] And this is like a perfect example of that.
[1244] Great point.
[1245] Another thing that they did is, oh my God, they were their own first customer.
[1246] Like they, a lot of times I think startups, presume to know their customer and they, because it's frankly difficult to go out in the real world and have these sort of conversations, you often project the needs of the customer into a map that you already have in your head and just start building.
[1247] I mean, Slack was something where they already, with Line Feed, had a good amount of development under their belt of what makes it easy to onboard new team members.
[1248] What are the sort of bells and whistles that we would add onto IRC?
[1249] So they knew better than anyone what this thing could be and what would make it useful.
[1250] And so you don't have the luxury of being your own first customer because you're not building a gaming company that has zero enterprise value shutting that down and then pivoting but to the extent that you can get you know a hundred percent fidelity on what are the needs of your customer that's pretty cool yeah a few more here so like zoom david you're talking about they pursued this bottom -up distribution strategy you know low price freemium the go -to -market is similar the network effects that they have so you basically have this equation which is low price in the enterprise plus network effects equals unprecedented growth rates.
[1251] And it's interesting to see sort of how long Zoom and Slack have continued at these really high growth rates.
[1252] Obviously, like, their products, you know, aren't incredibly expensive.
[1253] So the tradeoff you're making there is that your average customer value, if you're weighing in all your free accounts, is actually quite low.
[1254] But your growth continue to stay high for a very, very long time.
[1255] It's just sort of interesting to sort of understand that equation a little bit, which gets me to their top of funnel is operating a free product.
[1256] And so I want to do a little analysis on this.
[1257] If you divide their total marketing spend last year by the number of new paying customers that they brought on last year, you end up finding a cost of customer acquisition around $8 ,000 per company.
[1258] So, okay, that's actually, you know, for a consumer company, that'd be crazy expensive.
[1259] For an enterprise company, it's not that bad.
[1260] But it's significant, $8 ,000 to bring on a paying customer.
[1261] However, when they break out their cost of sales, it's only about $600 per company.
[1262] And so what you see there, I mean, there's...
[1263] The acquired Slack community.
[1264] Yeah.
[1265] There's some marketing dollars in there from Billboard and stuff like that.
[1266] But basically, what you're seeing is that there's a massive expense hosting these free organizations that, if you think about the different may not ever convert to fade.
[1267] Why is their cost of sales only $600, but and yet they're somehow spending an average of $8 ,000.
[1268] Yeah.
[1269] It's the massive amount of free usage of Slack that's going on out there.
[1270] So while it's a brilliant and disruptive strategy, it's an expensive one to run a huge free product like that.
[1271] Yep.
[1272] Great point.
[1273] Great point.
[1274] The other thing that I wanted to talk about a little bit here is a couple of Zoom comparisons.
[1275] So we mentioned that Zoom is externally viral while Slack is internally viral.
[1276] that ends up showing up in the sales efficiency.
[1277] And so when you look at Slack's return on sales and marketing costs, 111%, great.
[1278] But Zoom's is 180%.
[1279] And it's sort of...
[1280] I was expecting 1 ,000 and 80%.
[1281] But yeah, it just screams to you like, whoa, when you have the capability to promote your thing organically to people at other companies, you know, it just sort of has that natural.
[1282] So Slack is, I would say, great, not a plus on sort of that metric.
[1283] The last thing that I'll sort of say on Zoom and Slack, and this is not an original thought, this is a collection of some other things that we definitely did some good research on around the web.
[1284] Zoom made a previous way of working better, but Slack invented a completely new use case at scale.
[1285] That's one of my key takeaways, as I compare these two companies, is I intrinsically understood how to use Zoom at first.
[1286] Great, yeah, I've used a lot of these.
[1287] They were mostly bad.
[1288] Oh, this one's good.
[1289] Slack was like, what is this thing?
[1290] and there's an education curve on that.
[1291] I had, but somehow Slack was less obvious to me when I came in what it was supposed to be.
[1292] Well, I guess even if you have used HipChat, I mean, if you were on a technical team using HipChat, Slack, once it's the whole organization, just the use case is different, right?
[1293] Yeah.
[1294] You have all these people who've never used it before.
[1295] So even if you have used something like this before, your experience changes because now the organization's experience has changed.
[1296] Yep, for sure.
[1297] All right.
[1298] Let's do a quick discussion here of a section that we inserted a few episodes back about value creation and value capture.
[1299] So value creation here.
[1300] Tons.
[1301] Oh my gosh.
[1302] All these companies are paying all this money to have this new way of working that's frictionless and real time and live and highly interactive.
[1303] But boy, at what cost?
[1304] And I think you see this complaint a lot flying around Twitter.
[1305] I certainly feel it a lot from coworkers who are saying, you've hijacked the way that I work.
[1306] and it's now an interrupt -driven workflow instead of me being allowed to go into flow on something and then pop my head up and I decide when I want to check email.
[1307] It's like the opposite of the four -hour work week.
[1308] It is the always on, always sort of stimulated, always interrupting you.
[1309] I think we are early innings in understanding the impact of this, very much like the studies that are starting to come out about open office space.
[1310] I still use open office space.
[1311] I still like it.
[1312] very much recognize the downsides of it.
[1313] This is open office space in technology.
[1314] And as someone can always tap you on the shoulder.
[1315] And of course, there's settings to change and stuff.
[1316] But there's a cultural understanding that you are reachable by Slack.
[1317] You know, when your dot is green or during working hours or whatever the thing is, I think we don't yet fully understand the negative impacts of that style of communication being the predominant expected one in the workplace.
[1318] Yeah.
[1319] Yeah.
[1320] Now, all that said, there's way less to criticize.
[1321] here about this company then like sharing economy companies like Uber and Lyft that, you know, we've talked about the potentially very destructive things in the world, like the underpayment of wages and things like that.
[1322] This company has none of those strong.
[1323] I think on the flip side, though, I've absolutely agree.
[1324] And I think it's a broader cultural issue.
[1325] We're just grappling with at this point in time, a moment in time of being addicted to our technology.
[1326] I think a big benefit I've definitely seen from Slack as you know as somebody who I just struggle with email like I'm such a like when writing email you literally never answer my emails I really like I'm terrible at email and it's and and I used to be much better at email because I like had to devote time to it and like I want to like make sure I've written everything correctly and whatnot yeah Slack has been just the most liberating thing for me like we have shared Slack channels with all of our portfolio companies at Wave.
[1327] And so all of the most important people in my work life internally and externally to me are on Slack.
[1328] And it's just like, it's like a weight has been lifted.
[1329] You know, I can communicate instantly easily.
[1330] I don't need to worry about, you know, structuring the perfect sentence or whatnot or crafting the right narrative.
[1331] Like I can just like communicate much more simply.
[1332] It's been great.
[1333] My email has suffered as a result of it as Ben knows.
[1334] But like, you know, we text, we chat on Slack, like, it's better.
[1335] So listeners, if you want to get in touch with me, hit me up in the acquired Slack.
[1336] I'll also say a thing that, David, you do extremely well.
[1337] And just like listeners know this about David, a lot of us are kind of on a hamster wheel with communication where it's like, ooh, got a new thing in, got to answer it out.
[1338] You've been very disciplined for years about saying, like, should I actually be just like responding to everything that comes in?
[1339] And let me be more intentional about what I let, hijack my life and what I don't.
[1340] And if I don't answer this thing, what's the worst thing that can happen?
[1341] And I think Slack is kind of, in some ways, the antithesis of that very intentional thinking.
[1342] But you've applied it to the fact that you're going to be very intentional about being on Slack and structuring Slack in a way that works for you.
[1343] Well, what's cool is that the just the cost of that is much lower on.
[1344] So it's much easier on Slack to be like, yep, I'll respond to almost anything.
[1345] Whereas on email, like that cost is so high.
[1346] Yeah.
[1347] Anyway, so there's it's just super interesting Slack is we talked about the cultural moment that was ready for it like it was so perfectly positioned for, because I think this trend is much broader than Slack but to capitalize on it.
[1348] Yeah, so Slack allows for frictionless always on seamless communication inside of an organization but everybody still has email for outside the organization and I'm sure that Slack has spent a lot of time thinking about this but you know what does this type of workflow look like?
[1349] Can we kill email for, you know, for all use cases instead of just half of them.
[1350] Shared channels in their current form, you couldn't, our specific use case as a venture capital firm is like great for it.
[1351] But it is amazing.
[1352] But if you're like a BD person, like how are you going to communicate over Slack with your, if your workflow and your external partners fit such that you could have shared channels, it's truly wonderful.
[1353] I wonder if, I wonder if Slack will ever launch like an inbox feature where it's like if somebody knows my Slack handle, they can just send a Slack message to my inbox that's universally accessible anywhere in the world.
[1354] Or maybe not.
[1355] And listeners will have to stay tuned for our next episode and season finale.
[1356] Dropping soon.
[1357] All right.
[1358] So grading?
[1359] Who, grading.
[1360] Yeah.
[1361] So listeners, as you know, with these recent IPOs or DPOs, we ask the question, five years from now, what will an A plus look like and what will an F look like?
[1362] And the big thing in my mind is Slack currently has a revenue multiple of 46x with their current trading market cap.
[1363] I haven't checked comps lately, but something tells me that's high.
[1364] If, I mean, it's their company with no earnings, so you can only go on a revenue multiple.
[1365] So what will they need to do to justify that?
[1366] And is it continue business as usual with high growth rates and a great cact -lTV ratio, albeit over a longish time span, and then just continuing to do that and hoping that the Tam on the number of organizations that are going to do the thing is really big?
[1367] Or do they need to be broader than what they currently are and sort of take the Dropbox approach of going, oh, crap, it turns out we were a $7 to $10 billion feature.
[1368] And now we need to figure out how do we become a bigger company if we want to be a more valuable institution?
[1369] What do you think on that?
[1370] Well, I think the A plus case is the answer is it's both.
[1371] We, I don't know if listeners you remember, the great, I think it was a Pizza Hut or a Domino's commercial back in the day with Dion Sanders and Jerry Jones.
[1372] I have no idea where this is going on.
[1373] Oh, man. These commercials were so great.
[1374] It was Jerry Jones, the owner of the Cowboys and Dion Sanders, when he played for the Cowboys.
[1375] And Dion signed the largest contract ever of $35 million a year.
[1376] And so this commercial, it was like they were, whatever pizza company they're advertising, you could get both a deal to get both two things.
[1377] And there was the negotiation between Jerry and Dionne and Jerry's like, so Dion, what do you think?
[1378] 15, 20 million?
[1379] Dion's like, both.
[1380] So anyway, long -winded way.
[1381] I think the A -plus is both.
[1382] Both the chat communication market is enormous, tons of headroom left to grow.
[1383] And Slack develops new monetized features, products, tool sets around that that they can own, maybe start to rebuild a bundle of the productivity suite in the enterprise and go from $80 per year per user to across a suite of products, you know, $200, $300 per user per year in the enterprise.
[1384] Yeah, it's interesting.
[1385] If you look at like a storage company like Dropbox, you know, the Dropbox's enterprise value effectively comes from People Share files on it and use it for that and pay for that.
[1386] That's a 10 -ish billion dollar opportunity.
[1387] If you look at group messaging, Slack, you know, that's a $20 -ish billion opportunity.
[1388] When you look at the aggregate of enterprise email and a full productivity suite with Microsoft Office, like, oh, my gosh, we're, you know, in a half trillion dollar opportunity.
[1389] And so it's interesting to sort of think about what the bundle looks like.
[1390] What do you need to own to be the most valuable piece of that?
[1391] And I think identity is definitely a piece of that.
[1392] Well, Slack own identity.
[1393] Microsoft sort of has active directory.
[1394] There are others that are sort of taking that for the non -Microsoft world.
[1395] Google, obviously, has taken a chunk of it.
[1396] It's hard to say sort of what the value of G Suite is, since it's not a thing that Google breaks out, but what is the value of their productivity business?
[1397] I suspect it's a lot larger than Slack's productivity business.
[1398] So a big question for me is at what point does Slack do the thing that Dropbox did with Dropbox paper?
[1399] And then recently with this kerfuffled thing that I need to understand more or whatever.
[1400] the new dropbox is.
[1401] But like when do they decide to, not to be too pinning on the air, but when do they decide to go build more of the productivity stack in a way that is the, not the periphery pieces of the sort of minnows of communication, but the broader heavy hitting, we're going right from the jugular.
[1402] The other thing that this screams to me and is one of the most important reasons to become a public company is not just build internally, but a choir.
[1403] Atlassian has done this.
[1404] extremely well within their niche.
[1405] Will Slack go by like Notion or, you know?
[1406] Whomever, yeah.
[1407] As a public company, it is much easier to do that than it is as a private company.
[1408] Yep.
[1409] Great point.
[1410] So A plus is they do both.
[1411] F is.
[1412] F is where you, we're all using Discord or Microsoft teams in the future.
[1413] I mean, frankly, I think the biggest thing that could could be really scary for them is Microsoft actually succeeds at acquiring a significant amount of the upstarts and non -Microsoft customers to work into the team's organization.
[1414] Well, I think what's also really, like, the, we maybe alluded to this a little bit on the Zoom episode, but I suspect we are going to be entering an age of relatively large acquisitions in the productivity world by both the traditional players, the Microsofts, the Salesforce, the Googles, and Slack and Zoom and these large public upstart companies.
[1415] Yep.
[1416] Do you have any that you want to go out on a limb and say?
[1417] No. I mean, nothing I feel good enough to speculate about other than like I...
[1418] Consolidation in the space feels like a good idea.
[1419] Yeah.
[1420] Yeah.
[1421] Cool.
[1422] Follow -ups.
[1423] Follow -ups, yes.
[1424] So we apologize.
[1425] We made an error during our Uber episode.
[1426] So apologies both for the error and that it's taken so long to correct it.
[1427] But thank you so much to all the folks who...
[1428] I think we had four or five people talk to us about this.
[1429] Who reached out about this.
[1430] We waited so long to correct it because we wanted to do it on an episode without a guest.
[1431] On the part of the delete Uber story, when at JFK Airport with the surge story, we actually had it reversed.
[1432] So the true story is that, or how it actually went down, is that Uber had taken a lot of heat previously for when the surge would automatically kick in during periods of high demand, and sometimes that high demand was like, you know, there was like, you know, something, a disaster happening or something like that.
[1433] And so they had a playbook for turning off surge when, in certain cases, and when the protest, when the taxi strike at JFK happened around the immigration ban, Uber ran that playbook and turned off surge.
[1434] And so we had said on the episode that they turned surge on and that's what people got upset about.
[1435] It was actually that they turned it off.
[1436] Because they wanted people to be able to, easily get to and from the airport, you know, without being seen as, well, frankly, without actually being mercenaries about it and making a bunch of money from people being in a dire situation, but also to avoid the perception of being seen as doing that.
[1437] But it turns out, I think partially, and several of the Uber people who wrote us said, you know, it was just the perception of Uber was so bad at this point, which was Uber's fault, you know, but that no matter what they did, they were going to get pilloried.
[1438] And in this case, people were like, you're trying to break the strike, you know, these people are striking to protest this horrible thing that happened.
[1439] Yep.
[1440] And you Uber, are you in league with Trump?
[1441] Are you trying to break the strike?
[1442] Blah, blah, blah.
[1443] So the way I would imagine this, if you want to make a metaphor, is they sort of like whittled a bunch of wood shavings on a hot day into a pile.
[1444] And then there were a sort of, you could throw one of two lit matches at it.
[1445] And they chose which one to have thrown.
[1446] No matter which way, it was, yeah.
[1447] Anyway, so we apologize for the error and wanted to correct it.
[1448] Yep.
[1449] All right, carve -outs.
[1450] Somehow I never did this carve -out.
[1451] I was looking back at ones I had previously done, but I watched all of the expanse end of last year on Amazon.
[1452] And boy, is that show a masterpiece.
[1453] I am extremely excited for it to come back.
[1454] Amazon bought the rights.
[1455] They did three seasons, and then it sort of pseudo -ended.
[1456] And then Jeff Bezos, I think, is sort of personally a fan.
[1457] And Amazon bought the rights and is developing season four.
[1458] But if you like really gritty sci -fi and you liked, like, like, like, like like Battlestar Galactica and shows like that, you're going to love the expanse.
[1459] It's a great sort of political drama in the gritty future in space.
[1460] Nice.
[1461] My carve out full disclaimer biased because we are investors at Wave, as everybody knows who are LPs.
[1462] Our portfolio company, Alma, Dan Hill, who was previously the head of growth at Airbnb, we had on one of our first LP episodes.
[1463] They just launched a big product feature previously on Alma.
[1464] you could use Alma to give to nonprofits through funds around causes.
[1465] So the wildfires in California were a big fund that they created that you could just give to the Alma fund for the wildfires and they would distribute that to organizations that were addressing the wildfires.
[1466] They just launched last week the ability to give to any nonprofit in America on Alma, which is super cool.
[1467] So now finally there's there's one place, Alma, where you can go and easily, discover and give to anything rather than having to wrangle with, you know, the website or Google the charity, you want, find their jank donation tool.
[1468] PayPal, if they even have it or whatnot.
[1469] So you can now give it to anything, which is really cool in and of itself, but also your, it all, you manage it all through Alma.
[1470] So you get your tax receipts at the end of the year, all in one place, all through Alma.
[1471] You can control the level of communication back and forth.
[1472] Very cool.
[1473] So encourage everybody, A, to give to great causes, and B, if you want a more modern way to do it, use Alma.
[1474] That's awesome.
[1475] Well, David, I have to thank you for bringing Dan Hill into my life.
[1476] Dan is awesome.
[1477] He's such a wonderful person.
[1478] He's obviously on the LP show, and we did a great episode on sort of dissecting Airbnb's growth and the tactics they used.
[1479] And we ended up actually doing a deeper dive with a bunch of PSL companies in him, having sort of a roundtable and talk sort of best practices for growth.
[1480] And, yeah, he's awesome.
[1481] Yeah, Dan and his co -founder, Michelle, who's also a PM from Airbnb.
[1482] They're in a part of many of the big moments and features that unlocked growth for Airbnb.
[1483] So, super excited to work with them.
[1484] That's awesome.
[1485] Our sponsor for this episode is a brand new one for us.
[1486] Statsig.
[1487] So many of you reached out to them after hearing their CEO, Vijay, on ACQ2, that we are partnering with them as a sponsor of Acquired.
[1488] Yeah, for those of you who haven't listened, VJ's sales.
[1489] story is amazing.
[1490] Before founding Statsig, VJ spent 10 years at Facebook where he led the development of their mobile app ad product, which, as you all know, went on to become a huge part of their business.
[1491] He also had a front row seat to all of the incredible product engineering tools that let Facebook continuously experiment and roll out product features to billions of users around the world.
[1492] Yep.
[1493] So now Statsig is the modern version of that promise.
[1494] And available to all companies building great products.
[1495] Statsig is a feature management and experimentation platform that helps product teams ship faster, automate A -B testing, and see the impact every feature is having on the core business metrics.
[1496] The tool gives visualizations backed by a powerful stats engine unlocking real -time product observability.
[1497] So what does that actually mean?
[1498] It lets you tie a new feature that you just shipped to a core metric in your business and and then instantly know if it made a difference or not in how your customers use your product.
[1499] It's super cool.
[1500] Statsig lets you make actual data -driven decisions about product changes, test them with different user groups around the world, and get statistically accurate reporting on the impact.
[1501] Customers include Notion, Brex, OpenAI, FlipCart, Figma, Microsoft, and Cruise Automation.
[1502] There are like so many more that we could name.
[1503] I mean, I'm looking at the list, Plex and Versel, friends of the show at Rec Room, Vanta.
[1504] They, like, literally have hundreds of customers now.
[1505] Also, Statsig is a great platform for rolling out and testing AI product features.
[1506] So for anyone who's used Notions' awesome generative AI features and watched how fast that product has evolved, all of that was managed with Statsig.
[1507] Yep.
[1508] If you're experimenting with new AI features for your product and you want to know if it's really making a difference for your KPI's Statsig is awesome for that.
[1509] They can now ingest data from data warehouses.
[1510] so it works with your company's data wherever it's stored so you can quickly get started no matter how your feature flagging is set up today.
[1511] You don't even have to migrate from any current solution you might have.
[1512] We're pumped to be working with them.
[1513] You can click the link in the show notes or go on over to stat sig .com to get started.
[1514] And when you do, just tell them that you heard about them from Ben and David here on Acquired.
[1515] All right, well, listeners, if you aren't subscribed and you like what you hear, you totally should.
[1516] We'll be continuing to cover all the big upcoming tech IPOs, and we will actually be rounding out our productivity trilogy with Zoom, Slack, and one more in the next episode.
[1517] And we will see you next time.
[1518] We will see you for the season finale coming very soon.
[1519] To a podcast player near you.
[1520] After you fell out the survey.
[1521] Thank you, everyone.
[1522] Later, David.