...as an aside, out of curiosity, does anyone know why the "...and the company goes away." phrase was linked to www.towerrecordsmovie.com specifically and how is that related to Fog Creek ?
I think it's a reference to how some businesses go out of business when the the theory that people will pay money for a product fails to hold true. Tower Records being one such example.
That said, the tenses and general grammar of the whole sentence don't really hold up to the usual Joel standards and sits very strangely.
Sort of, all the US based stores closed (went away) in 2006. Most of the international stores hung on, but most of them are now closed. So from their mega footprint in the 90's and 2K's to a few stores means they've gone away.
The surrounding paragraph gives a bit more helpful context.
"A typical startup is built around a single product, and some theory that people will pay money for that product. This theory eventually become false, and the company goes away."
The link to towerrecordsmovie.com is suggesting that many single-product companies outlive their product's usefulness, such as Tower Records. Tower Records was a large seller of physical music media (CDs, Casettes, Vinyl) which went out of business in the mid-2000s.
The author is suggesting that since Fog Creek is not a single-product company that this does not apply to them.
Incidentally, Tower Records didn't 'go away', it's still doing roaring business in Japan, specifically because Japanese like the tactile feel of CDs, and because Japan's anti-crime culture generally meant that file-sharing never became a thing [https://medium.com/cuepoint/the-tower-records-stuck-in-time-...].
A lot of the time when a product or service appears to have 'gone away' it's actually moved on to serving a different audience with different needs.
I'm glad to hear this. I'll never forget the hour or so I spent at the Tower Records store in Shibuya in the mid-90s. That store was gigantic! As a twelve year old with limited pocket money i had a tough time choosing the 3rd, 4th and 5th CDs I'd ever owned. Good times!
It's still a huge store, I was there last year. I thought it's a Japanese company in the first place, but apparently it wasn't? One of the biggest Japanese convenience store chains, Lawson, also used to be an American chain.
I don't know what amount of revenue they run, but tower.com is I believe the same company. They closed the brick and mortar stores in markets where those weren't competitive. It's good to be flexible. You can choose to follow a market and live or die with it, or you can choose to follow the customers to the new market. In the US, most of the market for certain items is via online orders.
congrats, Anil! Great news for Fog Creek, and a great opportunity for you to have a big impact on software development!
Hopefully one of the early things you'll be focusing on is the diversity of GoMix's technology (including building accessibility in by default) and its community. There's huge potential, and a great chance to address it at a relatively early stage.
Absolutely. Right now, Gomix is not very accessible (in terms of editing code, not mental model), but that's definitely an area we'd like to improve in. And more broadly, making the creation of tech accessible to a far larger audience is fundamental to Gomix's mission, and to Fog Creek overall. We're committed to getting it right.
Tower Records was so wrapped up in selling physical goods (CDs and earlier, LPs) that they were totally blindsided by the rapid change in listener preferences. They didn't react quickly enough, soon enough, as sales fell off a cliff thinking it a temporary correction instead of a fundamental change.
So the lesson is that once a company has "made it" they should diversify so that they aren't dependent on a single income stream.
You should totally watch the film, btw. I used to shop at the Watt Avenue store in Sacramento, and the breadth of titles they carried was amazing. Easily five times the product that a store like Sam Goody would carry, and in niches you'd never see anywhere else. If there were such a thing as Celtic Rap, Tower Records would have carried it.
My brother in law was a buyer for Tower. They weren't blindsided, they knew it was coming. The record labels at first refused to license music to them for digital distribution, then later insisted on onerous terms. Tower missed out on digital because of the record labels. It took Steve Jobs' force of personality to knock sense into the labels.
Interestingly, I recently browsed in the Dublin Tower Records store and the LPs now outnumber the CDs. Currently, they’re the largest specialist retailer in physical media (music, films and books) in Dublin.
As I understand it, the Dublin franchise became independent of the Tower parent company some years past – or at least they advertised themselves as being “independent”.
What are the pros and cons of Fog Creek spinning out its successful products into separate companies, rather than the typical model of simply keeping them?
A big benefit is that you can make decisions, specifically structural ones, that might be best for your product, but not for the parent. For example, StackOverflow's most likely revenue stream was clearly going to be ads, which was not only a different revenue stream than Fog Creek selling shrinkwrapped products, but could have put Fog Creek into a position where it had a conflict of interest (see Google and Microsoft for excellent examples of that at scale). Spinning it off avoided that. Likewise, it was obvious that Trello was going to need a lot of VC to grow fast, which (at least at the time) ran directly opposite Fog Creek's "bootstrap everything, grow organically, never sell" attitude, so spinning that off also made sense.
As a counterexample, while I obviously don't know that Kiln would have thrived if it had gone a different direction, I do know that being locked into FogBugz' business model, instead of having the freedom to innovate our own, did ultimately stunt our growth quite a bit. FogBugz charged by the user, and so Kiln had to charge by the user, but this really hurt us. First, our costs scaled by repositories' size and count, not by users (which were basically free for us as such), so our costs didn't necessarily have anything to do with our revenue. Second, charging by the user is a horrible idea for an SCM. FogBugz could reasonably provide a clear separation between tools using its data and people using its data (since the only UI was the website), but in the world of DVCSes, it's really tough to provide the same distinction. Many people happily just used things like TortoiseHg and treated Kiln as a code backup service, so having no-website accounts wouldn't have accomplished much. We ended up forced into an awkward position of creating deliberately hobbled modes of access for tools that would by definition be prohibitively obtuse for humans, or telling people to shell out an extra $15/mo just so Jenkins could talk to the thing. Kiln switching to its own business model while remaining at Fog Creek would've been theoretically possible, but when business decisions that might be good for your new product will hurt your legacy product, that's a really tough argument to make. Having Trello and StackOverflow at their own companies completely avoided that problem.
The downside, of course, is redundancies and frayed vision. In a world where Trello and StackExchange remained at Fog Creek, I can imagine Fog Creek being the productivity company, with all of these tools tightly integrated à la Microsoft Office and presenting a coherent vision of how to develop software. You won't get that if you're spread across multiple companies. And, of course, you can end up in situations where, specifically because of all the reasons I pointed out above, two of your companies are going at each other a bit (e.g. Trello vs. FogBugz), which, even if subtle (those products don't honestly actually compete much), means you're spending at least some money competing with yourself. And, of course, you lose out on being able to easily move employees from one company to another, reusing technology amongst multiple companies, etc.
I'm sure there's more, but those are ones I remember seeing while I was still there.
So at a typical benchmark of $500,000 of revenue per employee they should be making atleast $150,000,000 per year in revenue. I have to imagine that they are making much more than that, given that I've seen valuations of $500 million.
I'm going to give them the benefit of the doubt and assume that they are still very lean and well run, I'm a big fan, but the fact that they do data dumps makes me feel much better.
I've seen too many bankruptcy/wind downs of companies, and one thing you can usually bet on is that once that process starts, the data gets locked up and treated the same as any other asset, which is to say, sold to pay debts. Or put another way, once a company gets into trouble, releasing their data often gets taken off the table as an option.
Again as a reminder to startup employee's, the company was founded in 2008 and hasn't really had any talk about going public or selling, so always make sure you get atleast a market salary from any startup you join as your options even at a well run company could take more than a decade to provide you with liquidity.
>Wow Stack Overflow has 300 employees!! So at a typical benchmark of $500,000 of revenue per employee they should be making atleast $15,000,000 per year in revenue.
Isn't that 150 million a year in revenue? If so, 500M is about right assuming a 3x revenue valuation.
> founded in 2008 and hasn't really had any talk about going public or selling, so always make sure you get atleast a market salary from any startup you join as your options even at a well run company could take more than a decade to provide you with liquidity.
Is $500,000 a common benchmark? A couple of quick searches seemed to show $100-300K for tech companies with a $1 billion valuation. My current employer has something like 315 employees, and while I'm not privy to their financials or our investors projections, the thought of hitting that kind of revenue target is bonkers to me.
It really depends on the company. (Although $100k sounds really light as it would be hard to make payroll unless the company's employees are all in a low-cost country.)
TL;DR is most successful tech companies at scale make > $300k/employee in revenue.
Benchmark for the big-5 is about $1M/employee/year in revenue. FB is about $1.2M, GOOG is just over $1M, AAPL is $1.8M, MSFT is $740K, AMZN is $460K (but recall that they have a large warehouse staff and with their 3rd-party sellers, they only recognize revenue off the fees charged to the seller and not off GMV). That's how they can afford to pay $400K+ total comps, and why they're sucking the atmosphere out of the engineering hiring market.
It can vary a lot by industry. A number of recent unicorns are glorified restaurants or retail shops that happen to deliver their product over the web or smartphone. Their revenue/employee numbers are going to be much lower because they don't have any significant proprietary technology to build an economic moat around, and their employees cost less.
$ per employee varies quite a bit on industry, the type of employees, insourcing/outsourcing decisions, stage of company, etc. $500K per person assumes massive profitability. As recently as January 2015 they needed $40mm to help pay the bills. I think it's more reasonable to assume they are near cashflow neutral. Perhaps $150K per person is more reasonable? That would imply $45 million in revenue. Seems reasonable, but perhaps slightly high.
Then there's a question of what multiple should be attached to their valuation. That would be driven by growth, margins and customer retention. Since their revenue model is similar to LinkedIn, perhaps their 8X multiple would be correct? That would put the value at $360mm. Given the wide variety of built in assumptions, I'd put it anywhere from $180mm to $540mm.
One other way to get at their valuation... How much would their investors have demanded for their $40mm investment? If it's 20%, then you'd have a $240mm post-money valuation in 2015. Are things going better or worse than their investors would have planned for? I'm not qualified to make this judgement, but if you assume it's the same, then we're at the lower end of the $180mm to $540mm range.
Net - I don't agree with your math, but your conclusions are sound. :-)
I've followed Stack Overflow from before the site hit beta. At the time, Joel and Jeff put out a podcast every week talking about software development and the new "project" they were working on that ended up being Stack Overflow.
It still feels like they had just made their first few hires.
I'm completely blown away they have 300 employees now. The last I had seen they had what looked like half a dozen closed door offices for their developers.
Does Stack Overflow also encompass Stack Exchange or is that a separate entity?
Anil Dash is like the Kim Kardashian of tech. I keep having to look him up to find out what he created (I'm still not sure).
I followed him on Twitter for a while because that was the thing to do, but after (I felt like) he kept loudly co-opting celebrity deaths to increase his profile, I started to wonder why he had any to begin with.
I'm not trying to be mean, but I guess that's the disconnect between makers and marketers (evangelists?). He could be great for incubating ideas.
I've made a few things over the years; I helped create Movable Type and TypePad, which were sort of the WordPress of their era (used to publish Gawker, HuffPo, etc.) and more recently built ThinkUp and Makerbase with Gina Trapani.
I don't think I've coopted many deaths, but certainly there are folks who found it off-putting how much I talked about Prince after his passing. FWIW, I've always talked about Prince that much. :)
Fortunately, the Fog Creek team is hugely talented and there's immense breadth to the ideas and execution that everybody can pull off, so my shortcomings shouldn't overshadow their work.
I don't think this entirely deserves the downvotes it got. It made me finally look up what SJW means on Google/Wikipedia (I honestly didn't know), and pejorative meaning aside, I think Anil would be happy to be identified with "socially progressive views, including feminism, civil rights, multiculturalism". Some of the comments here have said Anil is 'outspoken', and it's probably fair to say he tweets a lot about politics.
The insulting component is that an SJW doesn't necessarily practice what they preach, but instead just uses any deviation from a very rigid (and not necessarily beneficial) behavior code to score points and undermine others.
Of course a lot of people called SJWs are not that rigid or controlling: the ones who call them that are possibly that [rigid and (con)trolling] themselves.
Of course! And a lot of people called motherfuckers do not fuck their mothers, and the ones that call them that are possible fucking mothers themselves. But I think the meanings of both "motherfucker" and "SJW" are broadly agreed upon.
I dunno, Kim Kardashian has brought multiple wildly-successful mobile apps to market with a consistency that exceeds most publicly-traded companies trying to make mobile games. As someone who aspires to make apps that reach a broad audience, I find no slight in being compared to her.
Exactly. I believe it was MEANT to be insulting, but Kim K isn't the joke everyone makes her out to be. I think she's a genius. Whether you agree with the image she puts out and how it may impact culture, she's totally in control at all times and has turned her fame into commercial success better than almost anyone in recent history.
I absolutely agree. She's done what many with far more resources and VC funding haven't been able to do. Just the way she was able to build her now incredibly strong brand over the years is a testament to her knowing exactly what she's doing. She then successfully leveraged that to build very successful apps and games that people love playing and using. Many many people with far more fame and brand appeal than her haven't been able to replicate this and probably couldn't if they tried.
She was briefly "famous for being famous", but not particularly famous at that point. Her empire was built based on her (and her family and her organization's) fantastic execution from then until now.
She knew someone was there was a leak to the yellowrags of some of her privately circulated pic, among friends and family. So when her baby was born, she sent different pics to different friends. She did the equivalent of <email>.$site@gmail.com to smoke out the sellout.
(My 2c: though I'm on the political right and agree that tech is becoming increasingly hostile to non-liberal ideas, I think many of Pax's ideas are quite weird).
We'll see! For what it's worth, I've been a CEO 3 times and have spent years as a management consultant with most of the clients being CEOs of multi-billion-dollar companies, so there are parts of my experience that aren't as obvious from my public persona.
More to the point, I've been on the board of Stack Overflow for some years now, and have known Fog Creek's business well for quite a while, so it wasn't too big a leap. Fog Creek also has a well-established and very talented team; I'm just augmenting their existing strengths.
Our industry is a lot smaller than it might seem. Anil has been around for quite a while and lots of us have had the pleasure of getting to know him to different degrees both online and off. He's generous, smart, and often outspoken but always thoughtful (IMHO). It's no accident that he keeps popping up on your radar.
Anil is a kind, thoughtful person and one of the few people in technology who will go out of his way to make time for you and genuinely get to know you without an agenda or need for a value exchange - I really admire that. We spent an afternoon wandering New York the time I met him, and he's truly constantly considering how he can impact the world in a meaningful way.
This is really, really cool, and he's a great fit for the job - along with the talented team over there. Super excited about this, personally.
I think of Dash as very respected in media, and I think of Fog Creek as being as the "engineers' company", so was surprised to see him CEO of Fog Creek and Joel CEO of StackOverflow. If anything, I would've thought it'd be vice versa? It's a testament to what a game-changer StackOverflow is in terms of information-design-for-humans that it could be capably headed by someone like Dash as much as Spolsky.
But if Fog Creek Software is tasked with coming up with more products and software-as-services, I would've thought that Spolsky would be more fit for that, even if StackOverflow is the biggest piece of the pie? (I admit to knowing little of how executive structures work though)
I think there's something in the idea of putting people in roles where they balance out other people's strengths and weaknesses.
Fog Creek probably doesn't lack for smart engineering people, but might benefit from someone who has a wider view on media and society in general - as well as enterprise sales, which Spolsky calls out specifically. Tech people and engineers, on the whole, loathe enterprise sales.
Dharmesh Shah and Brian Halligan at HubSpot spring to mind two guys who you might not ordinarily put together but they've both got a big role, and it's the same kind of deal. Shah is an engineer and a hacker, Halligan is more of an old-school sales guy.
The reason why this is a smart move is the same reason why it is counterintuitive.
Agreed, Fog Creek has been fascinating to watch over the years, back then (and now) they seemed to follow a completely different path to building a software company and the incredible detail they put into developer experience, I used to love watching their YouTube channel when it was about their internal setup/staffing etc.
Was always a dream to build a company like that, one day maybe.
82 comments
[ 6.0 ms ] story [ 151 ms ] threadThat said, the tenses and general grammar of the whole sentence don't really hold up to the usual Joel standards and sits very strangely.
Sorry to see that happen, but it does.
"A typical startup is built around a single product, and some theory that people will pay money for that product. This theory eventually become false, and the company goes away."
The link to towerrecordsmovie.com is suggesting that many single-product companies outlive their product's usefulness, such as Tower Records. Tower Records was a large seller of physical music media (CDs, Casettes, Vinyl) which went out of business in the mid-2000s.
The author is suggesting that since Fog Creek is not a single-product company that this does not apply to them.
A lot of the time when a product or service appears to have 'gone away' it's actually moved on to serving a different audience with different needs.
Hopefully one of the early things you'll be focusing on is the diversity of GoMix's technology (including building accessibility in by default) and its community. There's huge potential, and a great chance to address it at a relatively early stage.
It seems like the positions listed here http://www.fogcreek.com/careers are support roles instead of product roles.
So the lesson is that once a company has "made it" they should diversify so that they aren't dependent on a single income stream.
You should totally watch the film, btw. I used to shop at the Watt Avenue store in Sacramento, and the breadth of titles they carried was amazing. Easily five times the product that a store like Sam Goody would carry, and in niches you'd never see anywhere else. If there were such a thing as Celtic Rap, Tower Records would have carried it.
As I understand it, the Dublin franchise became independent of the Tower parent company some years past – or at least they advertised themselves as being “independent”.
http://www.towerrecords.ie/aboutus.asp
Which still makes Fog Creek pretty amazing.
https://en.wikipedia.org/wiki/Fog_Creek_Software
As a counterexample, while I obviously don't know that Kiln would have thrived if it had gone a different direction, I do know that being locked into FogBugz' business model, instead of having the freedom to innovate our own, did ultimately stunt our growth quite a bit. FogBugz charged by the user, and so Kiln had to charge by the user, but this really hurt us. First, our costs scaled by repositories' size and count, not by users (which were basically free for us as such), so our costs didn't necessarily have anything to do with our revenue. Second, charging by the user is a horrible idea for an SCM. FogBugz could reasonably provide a clear separation between tools using its data and people using its data (since the only UI was the website), but in the world of DVCSes, it's really tough to provide the same distinction. Many people happily just used things like TortoiseHg and treated Kiln as a code backup service, so having no-website accounts wouldn't have accomplished much. We ended up forced into an awkward position of creating deliberately hobbled modes of access for tools that would by definition be prohibitively obtuse for humans, or telling people to shell out an extra $15/mo just so Jenkins could talk to the thing. Kiln switching to its own business model while remaining at Fog Creek would've been theoretically possible, but when business decisions that might be good for your new product will hurt your legacy product, that's a really tough argument to make. Having Trello and StackOverflow at their own companies completely avoided that problem.
The downside, of course, is redundancies and frayed vision. In a world where Trello and StackExchange remained at Fog Creek, I can imagine Fog Creek being the productivity company, with all of these tools tightly integrated à la Microsoft Office and presenting a coherent vision of how to develop software. You won't get that if you're spread across multiple companies. And, of course, you can end up in situations where, specifically because of all the reasons I pointed out above, two of your companies are going at each other a bit (e.g. Trello vs. FogBugz), which, even if subtle (those products don't honestly actually compete much), means you're spending at least some money competing with yourself. And, of course, you lose out on being able to easily move employees from one company to another, reusing technology amongst multiple companies, etc.
I'm sure there's more, but those are ones I remember seeing while I was still there.
tl;dr - Gave investors the ability to invest solely in Trello instead of all of Fog Creek.
So at a typical benchmark of $500,000 of revenue per employee they should be making atleast $150,000,000 per year in revenue. I have to imagine that they are making much more than that, given that I've seen valuations of $500 million.
http://www.businessinsider.com/revenue-per-employee-at-apple...
I'm going to give them the benefit of the doubt and assume that they are still very lean and well run, I'm a big fan, but the fact that they do data dumps makes me feel much better.
I've seen too many bankruptcy/wind downs of companies, and one thing you can usually bet on is that once that process starts, the data gets locked up and treated the same as any other asset, which is to say, sold to pay debts. Or put another way, once a company gets into trouble, releasing their data often gets taken off the table as an option.
Again as a reminder to startup employee's, the company was founded in 2008 and hasn't really had any talk about going public or selling, so always make sure you get atleast a market salary from any startup you join as your options even at a well run company could take more than a decade to provide you with liquidity.
EDIT I can't math
Isn't that 150 million a year in revenue? If so, 500M is about right assuming a 3x revenue valuation.
Best advice that I've read here in awhile.
TL;DR is most successful tech companies at scale make > $300k/employee in revenue.
Here's something I saw a while back:
http://www.businessinsider.com/top-tech-companies-revenue-pe...
Also this more recent edition:
http://www.businessinsider.com/revenue-per-employee-at-apple...
and this:
https://www.statista.com/statistics/217489/revenue-per-emplo...
It can vary a lot by industry. A number of recent unicorns are glorified restaurants or retail shops that happen to deliver their product over the web or smartphone. Their revenue/employee numbers are going to be much lower because they don't have any significant proprietary technology to build an economic moat around, and their employees cost less.
Then there's a question of what multiple should be attached to their valuation. That would be driven by growth, margins and customer retention. Since their revenue model is similar to LinkedIn, perhaps their 8X multiple would be correct? That would put the value at $360mm. Given the wide variety of built in assumptions, I'd put it anywhere from $180mm to $540mm.
One other way to get at their valuation... How much would their investors have demanded for their $40mm investment? If it's 20%, then you'd have a $240mm post-money valuation in 2015. Are things going better or worse than their investors would have planned for? I'm not qualified to make this judgement, but if you assume it's the same, then we're at the lower end of the $180mm to $540mm range.
Net - I don't agree with your math, but your conclusions are sound. :-)
I've followed Stack Overflow from before the site hit beta. At the time, Joel and Jeff put out a podcast every week talking about software development and the new "project" they were working on that ended up being Stack Overflow.
It still feels like they had just made their first few hires.
I'm completely blown away they have 300 employees now. The last I had seen they had what looked like half a dozen closed door offices for their developers.
Does Stack Overflow also encompass Stack Exchange or is that a separate entity?
I followed him on Twitter for a while because that was the thing to do, but after (I felt like) he kept loudly co-opting celebrity deaths to increase his profile, I started to wonder why he had any to begin with.
I'm not trying to be mean, but I guess that's the disconnect between makers and marketers (evangelists?). He could be great for incubating ideas.
I don't think I've coopted many deaths, but certainly there are folks who found it off-putting how much I talked about Prince after his passing. FWIW, I've always talked about Prince that much. :)
Fortunately, the Fog Creek team is hugely talented and there's immense breadth to the ideas and execution that everybody can pull off, so my shortcomings shouldn't overshadow their work.
She knew someone was there was a leak to the yellowrags of some of her privately circulated pic, among friends and family. So when her baby was born, she sent different pics to different friends. She did the equivalent of <email>.$site@gmail.com to smoke out the sellout.
I am very tempted to write my own thoughts on the issue, but rather than start a political flamewar, I'll simply link to both men's side of the story:
http://anildash.com/2013/09/my-meeting-with-pax.html
https://paxdickinson.wordpress.com/2014/10/22/moral-panics-a...
(My 2c: though I'm on the political right and agree that tech is becoming increasingly hostile to non-liberal ideas, I think many of Pax's ideas are quite weird).
http://anildash.com/2013/03/what-its-like-being-verified-on-...
Though now I see the timestamp for that post is 2013 and I definitely knew of your work before then.
https://archive.org/details/crankygeeks&sort=-reviewdate?and...
More to the point, I've been on the board of Stack Overflow for some years now, and have known Fog Creek's business well for quite a while, so it wasn't too big a leap. Fog Creek also has a well-established and very talented team; I'm just augmenting their existing strengths.
Correction: he is generous to those who happen to hold and express the correct opinions. To those who do not, pure vitriol and unbridled hatred.
This is really, really cool, and he's a great fit for the job - along with the talented team over there. Super excited about this, personally.
But if Fog Creek Software is tasked with coming up with more products and software-as-services, I would've thought that Spolsky would be more fit for that, even if StackOverflow is the biggest piece of the pie? (I admit to knowing little of how executive structures work though)
Fog Creek probably doesn't lack for smart engineering people, but might benefit from someone who has a wider view on media and society in general - as well as enterprise sales, which Spolsky calls out specifically. Tech people and engineers, on the whole, loathe enterprise sales.
Dharmesh Shah and Brian Halligan at HubSpot spring to mind two guys who you might not ordinarily put together but they've both got a big role, and it's the same kind of deal. Shah is an engineer and a hacker, Halligan is more of an old-school sales guy.
The reason why this is a smart move is the same reason why it is counterintuitive.
Was always a dream to build a company like that, one day maybe.