Evolving a Mature Software Company | Anil Dash, Fog Creek | BoS USA 2017

No comments

Anil Dash, CEO, Fog Creek

Anil discusses the challenges of stepping into a leadership role in what is perhaps one of the best-known software companies the public probably haven’t heard of. FogBugz, Stack Overflow and Trello are all products that have come out of the Fog Creek stable. This year, they unleashed Glitch to the world. Anil is committed to putting impact and sustainability, both of business model and company culture, at the heart of their vision for the future. What does that mean in practice and what can you learn about your own vision?

“The Startup Thing” – Recent worrying trends in software

  • In the post-web 2.0 era, startups came as a response to the pop culture idea of what tech was
  • There was very little discussion on how to build strong core values and much more on how to secure funding from important investors
  • A pattern of speaking but not acting on values began to emerge
  • The tech industry was on a path to being the next Wall Street (and not in a good way)

Software Matters

  • It is more than just a piece of technology; it is a set of values and sociopolitical statements that are created into something real
  • All the choices you make about design, UX, UI, etc will trickle out into society – see the “It’s complicated” feature
  • Bugs are more than just a broken piece of code. The big bugs are often problematic at a societal level even when the software works as designed
  • Sentencing software recreates the socioeconomic and racial biases that people have reinforced through years of judicial discrimination
  • What if creating technology is a way to reckon with the changes of the world?

Video, Slides, & Transcript below

Business of software

Video

 

 

Slides


Upcoming Events

BoS USA 2018

Don't Miss a Thing - Get BoS Updates

Want us to let you know about new talk videos, speaker AMAs, Business of Software Conference and other event updates? Join the smart people who get BoS updates. Unsubscribe anytime. We will never sell your email address.


Transcript

Anil Dash: Wow, I get walk on music? That’s nice! Am I on? Everybody all right? Good.

Thank you for being incredibly accommodating – I know my schedule changed. Already having been here a day and feeling the community of peers and people we can learn from has been such a respite. I think from those of us who get to go to other events and other conferences, to be able to actually talk to people about what we do, right? Not the product or service we’re trying to sell. I think it’s such a meaningful community to find. I want to talk a bit about my story getting to the role that I have now, Fog Creek Software and how I got there because I think it’s a story a lot of you can relate to as well.

In the year 2000, I was in New York city trying to get myself established. I was always the tech guy and I wanted to pursue my passion and I said I want to go and be a part of the music industry. I’ve always loved music and maybe I can apply my tech skills there and that will be something that people will find useful. I set out to find what I thought was my dream job. I was working for a company promoting music and videos online and I would be the tech guy that was defined as the role was. And as you can deduce from the introduction, the music I love is funk, soul and RnB and that was what I grew up with, loved and spent all my time listening to. So naturally as soon as I got into this role, most of what I promoted was hair metal. And these were like – it’s a fine thing, but it was like their kind of has been bands, not super exciting. They weren’t even really into the work they were doing, they were going through the motions, it wasn’t a good time in the year 2000 to be prevailing hair metal online.

And then every once in a while we would get into the higher end and the better acts and they would be the heavy metal bands. The stories you would share online were like; Judas Priest at that time their lead singer, Rob Halford, had left and so a guy who had been a lead singer in a cover band of Judas Priest got drafted to be a member of the actual band – like living out his fantasy! And right around this time, they were putting out an album and he got kicked out of the band – they were like yeah that’s not a thing we want, we want the real deal. The fans rejected the organ transplant of changing the frontman on the band. And so I was really excited! I thought I was gonna be part of bringing rock and roll to people. What can be cooler than that? What I was actually part of was bringing bad software to people cause I was a lousy coder and didn’t really know what I was doing and was putting out stuff that nobody wanted. There were many other business problems there, but it was an ugly realisation that writing code was not gonna be enough to make sure I was doing something that was meaningful to people nor necessarily connect to what people wanted.

anil dash

So what had seemed like this dream job and this foot in the door, something I really loved, pretty quickly started making me miserable. And that was a shock because I think, all of us, the reason we create things, the reason we’re entrepreneurs is, we’re ambitious. And we head towards something and think if I just get there, if I do the thing nobody would let me do, then I will be happy and fulfilled, other people will like it and everything will go as I expect it. One of the things that happens when you have a job you hate is you get a lot of free time cause you’re not super engaged in what you’re doing day to day. I was really strong on the idea that you shouldn’t let your job get in the way of your career.

I started blogging in 1999 and that was early enough that, to give you an idea of context, we had a get-together meetup of bloggers in New York city, all of us, and we were around 2 tables at a restaurant. That was the social media sphere in New York city and Manhattan around the turn of the century. Sounds like 100 years ago. And through those meet ups with the early social media and blogging community I met this guy, Joel Spolsky. How many of you have read this book? Yeah, me too! And it was sort of a revelation. Here is this brilliant founder, at the time I didn’t know about Michael Prior even though now I think he’s pretty amazing. I just thought this guy is a brilliant founder and they invented these products and have this vibe to them and it’s amazing – they’re doing things right. And I knew where I was, we weren’t doing things right. So it was just this idea that maybe the problem wasn’t fundamental, it wasn’t that you could never get things right, only some places were able to get things right.

I remember distinctly sitting at this job and going through the motions and you have the other – we didn’t have tabs in our browsers back then so you had to open another window; and reading this blog post, the Joel Test. And in it, if you haven’t seen it, it was a set of best practices for creating good software. If you wanted to make good software, you would have a bug tracker which is pretty obvious today but not always. You would use version control which again these days seems obvious. But I was like what if version control is when you zip up your code and email it to yourself? People still do that – it’s terrifying. And so, this idea of how he described his team working together and help people collaborate, it felt like Sci-Fi. This is a great vision, I’ve never seen this any place I’ve ever been, I’m certainly not doing that. That would be great but is this how reality works?

I came to understand the idea of how Joel and Michael had envisioned making good software was grounded in the idea of making a good company. You have an institution which is run correctly and the artefact of the fact that it was run correctly was that it made software that was meaningful and useful to people and provided value and utility to others.

So that was how I discovered Fog Creek Software and I thought wow that’s pretty great. I became friends with the guys and got to visit their first office. They were in Joel’s grandmothers Brownstone, in a neighbourhood where there’s nothing else going on on the east side of Manhattan. And I thought this is pretty cool and we kept in touch. And after I thought about it for a while why I had such an affinity for the company, what I realised was what Fog Creek represented to me as a fan boy and reader of the blog was a set of values. It was this declaration, the Joel Test wasn’t just here’s a checklist of things to do when you make software, but everything Joel had written was this idea that you need to value the people on your team and respect your users and community. Things, when you say them in broad terms it sounds obvious but in practice it can be hard and can fall by the wayside. They had wild success after that and they created fog bugs, one of the first most popular bug trackers. Joel created stack overflow that spun out to its own company, and years later they created Trello. They were doing that stuff and I was watching and admiring from arm’s length saying that’s great! But my own path diverged after that.

anil dash

So after it had that initial sort of insight into the idea that we could make software that people loved and we can make good software and tech that is really meaningful, I doubled down on blogging. It’s what I loved and I loved expressing myself that way, I thought the technology is really interesting and decided to do the start-up thing. I helped some friends of mine that were making some early blogging tools. Does anybody remember moveable type and type pad? Couple of you, thanks. It was another company with a brilliant founder, this woman had this idea of how people can express themselves. The crystallisation to me was; I was like someday there’s gonna be a million people with blogs. And at that time blogs meant any photo sharing – basically everything we do on Facebook, Twitter, Instagram all of it together was what we consider blogs. And she was like you dumbass, there’s gonna be like 100 million blogs. What are you talking about? One million, you’re thinking too small! And it was like, this is what it’s like to be with someone who is a visionary. It was very clarifying. And we made good software. The products made, people liked. That was really great! This is what we’re supposed to be doing. In fact literally we became the first case study for Fog Bugs. Joel and had stayed in touch. We said we’re making great software, we’re using your tool, we’re gonna be on your website and we’d like to figure out how do you show off what we’re doing and we’ll show off how your tools work great. I thought this means we’re doing it, the guy who makes the thing about the good software said that we’re the case study for the software so that means we’re good. Everything is going great.

At this point, the company I was in started pursued funding. It got some outside funding. Started getting lots of help, whether we asked for it or not. And one of the things they said was our founder, the visionary that thought of the idea of the company and the product, didn’t have enough experience. She hadn’t run a consumer SaaS business. Of course this was 2003 so nobody had run a consumer SaaS business. So they brought in someone who had never blogged and never written a line of code in a blogging tool. And he had experience I guess. You know how this story goes, I don’t need to talk to this crowd about the way that played out. It’s something where I was proud of the work we’d done, I deeply cared about the community and found myself after a few years really miserable. We had made something that on the outside was successful, that people used the tools and they built amazing meaningful sites out in the world, things that were culture changing. And yet, I looked at the culture inside the company, the path we were on and I wasn’t really proud of where we were at. In fact, I looked at structurally what had happened and we had this young woman who had been the visionary of the company and she had very quickly got kicked upstairs to be a product person. We don’t want you to bother with actually running the company and was side lined in her own company. And it didn’t sit right, it didn’t feel to me like a thing I wanted to be a part of. It felt like we had lost our way. And I walked away. The company at its peak was valued at $100 millions and had 100s of employees. I was paid a fine salary for the years while I was there, but essentially even having been the first employee and written the business plan, my equity was worth nothing. It was something that it took a while to process because I had bought the dream, I had bought the idea that this is the path you go on, this is the way you do it.

anil dash

So then, this is about 10 years ago, I thought about what it means that maybe we actually were a case study, but a case study of how the tech industry was shifting and what the dominant narrative was going to be. I don’t need to tell you how the last 10 years have gone in the tech industry for better and for worse. But the start-up thing became, certainly during that post web 2.0 era, what the popular consensus, the pop culture description of technology is about. It’s what you’re supposed to do. The social network came out and everyone said that’s tech and software. There was very little discussion of the set of values in describing the ways those companies were created. The idea of doing things right was optimising return for those investors. And I started to look at the pattern of which founders got pushed out and got funded in the first place. I looked at who was saying the right things about diversity and inclusion versus where they were investing. I looked at patterns across all of the companies that were held up as the examples, in this most recent era, and I didn’t recognise it. It didn’t feel like the place that I belonged. It didn’t feel like I wanted to be part of anymore.

I left San Francisco and Silicon Valley. I went as far as you could go, which was to Washington DC, to go and work on policy initiatives as part of a non-profit which is a software free environment. And it was clarifying, there’s no ideal place but it was something that I learned a lot from. Because I realised I still love software and tech. So the problem wasn’t the apps, the problem wasn’t the lines of code but there was something in between here that had to be why it was calling me back. And one of the epiphanies I had was in getting distance from the classic tech industry, the main stream industries, the big companies. I tried to extrapolate what was the straight line projection of the behaviours we were seeing and I looked at other industries. Especially being in DC at the time, this is sort of – these things go back and forth but if you ask people what do they feel about politicians, they will tell you they don’t like them. That’s pretty universal answer. If you ask them what they think about bankers, they will say I don’t trust those people. If you ask them about media, they would say it’s pretty terrible. The thing is in all those disciplines – it was an epiphany especially for me in policy and politics – there were people doing great work. And if you say but that one person is ok, I know a great banker. They’d say but that’s the exception. They’re not really the thing. If you’re on Wall Street and you are nice and thoughtful to people, you’re not really Wall Street is what people will say.

And then I looked at tech and Silicon Valley and I thought oh sh*t! They’re gonna hate us! If we go forward 10 years like this, they will say tech is the same thing as Wall Street. The trends I looked at that drove that was; one, tech was changing things. It was really clear that everything was going to be transformed by the fact that we had these smart phones and apps in our pockets, this is the moment when that happens. And change alone is enough to make people mistrustful and feel unsafe. But also it was very clear already that what was gonna be sold as the gig economy was going to be something most people will see it as they’re threatening my job. What we saw as shortcomings or flaws in our privacy and security policies was people saying they’re creeping me out. Years ago, I wrote the piece pretty soon 100 million social security numbers are gonna leak, what are we going to do about it? I got a ton of pushback from people saying you’re an alarmist. There is no satisfaction as it turns out on retweeting yourself on Twitter saying a whole bunch of social security numbers are going to leak that after the Equifax leak, cause I’m like well we’ll all screwed! Yay! Like I wanna have that personality like I was right is enough. But I guess people get book deals and stuff, that’s what they do, but it didn’t work for me.

And then this idea that everything is tech. There was a company making mayonnaise, they said they were a tech company. There’s another one making a grill cheese maker and they said they were a tech company. Everything can’t be tech. Just because you have a website, doesn’t mean you are tech. There’s companies I like, there’s this company that makes mattresses and are on every podcast, and I’m sure they’re fine but you are not a tech company, you make mattresses. I don’t mean that in a negative way. I’m not criticising their existence. I’m saying these terms have to mean something. What I realised is the ship had already sailed. We’d already decided that a company that was about disrupting transportation systems and transit networks in metropolitan areas was also a tech company. If you say anybody that has a single IOS engineer on staff is a tech company, then nothing is a tech company. The term had become meaningless. There was no definition or boundary to the community. And the reason the definition and boundary of the community matters so much is because of people in this room. If you have a sense of community, you have a set of norms, a set of values, you can talk to each other about standards, you can look disapprovingly at someone who transgresses against them but if everything is tech and everyone is tech company and every company is in this industry, well then how do you have a corrective? How do you start to agree on what matters?

So this was an ugly realisation for me, because this was sort of my identity. When I was in the music business or working at newspapers – there’s the tech guy. That’s who I was. And now I was saddled to this thing that people, I was convinced, were going to see us as toxic, destructive, bunch of selfish jerks or at the very least kind of creepy. And so I was like I wonder if anybody has ever written down a set of values to sort of prevent that from happening? Has anybody ever documented a way of doing things right so you don’t have these negative feelings about the ways of technology is introduced into people’s lives? And I thought about could we change what the tech industry was? I became convinced that the answers is actually reforming the public perception of tech and certainly our own image as people that create software, our image of what we need to be. And I did this with blogging and started writing about here’s what’s wrong in the industry. And really, really promptly got a ton of pushback from people who had formally been friends in Silicon Valley; investors, VCs, former colleagues saying you are lost. Like variations on you’ll never work in this town again. Which was fine cause I was in New York so I was like I’m not in that town – I didn’t care that much. But it was clear. It was polite and nicely phrased but very clear.

anil dash

And then there was this thing. Egghead guy said this thing. I’m not going to say this thing. I don’t think software is eating the world is a good formulation. I don’t, I don’t. I don’t think it represent accurately the relationship that technology and software have with all the industries we’re impacting. I think a simpler more truthful way to say is that software matters. Technology today is a process of transforming other industries, a way of thinking about the way those industries run and increasingly, it is a set of political and social values. Some which we might agree with, some with which we might not. But we baked those in, software has values baked into it.

The example I always go to is when Facebook used to have the relationship status thing that said it’s complicated. And like yeah, if you’re at Harvard and hooking up with other undergrads it’s complicated is a very useful status indicator, that’s what it was made for. My parents had an arranged marriage, there is no it’s complicated. I mean it was always complicated but it wasn’t complicated in that way. Here you are, you’re married now! It’s not complicated. And so when you make a dropdown saying it’s complicated as one of the settings, you make a very strong declaration of who this platform is for. There’s a contrary example too, Facebook not that long ago, finally made the gender field on people’s profiles open ended so you could put anything you want in to the database which is great and it let people accurately capture their identity on the site. But these little choices, that’s one dropdown, one engineer that probably did that. One person made one dropdown and made that choice for a relationship status and it became so culturally relevant that all of you understood the relevance even though they took the feature out of Facebook 8 years ago, you all knew it’s complicated. These choices we make of interface dialogue, UX dialogue, percolate out into society.

The defaults on our apps for ordering stuff about whether there are tips at all and if so are they a percentage or an absolute dollar, and what’s the default? Those things used to be the point of organisation of entire labour movements. Certainly the subject of policy making and regulation around how people’s wages were handled. These are decisions by folks just out of their CS program. So software matters, matters really deeply but we don’t talk about it like it does. And what we talked about is bugs. I used to think about it doesn’t render right in this browser or we have a memory boundary issue and that’s going to be my bug. But like the big bugs are like we’ve got a criminal justice system that’s using software to suggest to judges likely sentences for people who are convicted of various ranges of crimes. There’s basically one software package dominant in that field and a lot of us make vertical software to serve certain industries. But in that case it was trained on the history of sentencing in this country which has had systemic biases, that I think everyone is aware of, for a long time and it reproduced those biases. So it was working as spec’d, it was working to function, but it’s reproducing injustices that have caused enormous pain and trouble in this country. And that’s not the category of bug that our bug trackers were designed to fix. I didn’t know how to log that one in the backlog. And that’s something we had to change at the much more fundamental level.

So in having this reckoning, again I looked out at what was happening in other industries and I looked at medicine, law, journalism. And when you transgress in law, you can be disbarred. When you cross the ethical boundaries in medicine, you can lose your medical license. And journalism is a little fuzzier, they yell at you for a while and then you get to come back. I think that’s how it works, I don’t really know. But in software, if you really make a serious bug or you were actually being a bad actor, there’s no corrective and in fact, most of our computer science programs don’t have any ethics training. A lot of us, we deduce it along the way and teach each other. Or if we had business training that came out of business school or something like that then that’s great. But you looked at who was getting funded in the main stream industry, and it’s like how could they make good ethical decisions if they haven’t been taught the basic context they were working in from a civil and societal standpoint? I don’t fault them, they just didn’t know. As a 23-year old kid right out of school given millions of dollars of resources to build some app. I wouldn’t be like let me hold on a second and go read my ethics text book, nobody is going to do that. So we had some built in structural problems around what they were encouraging and the worst part about that was that’s what everyone thought software was. So I was pretty disheartened, I thought maybe it can’t be salvaged. I thought maybe I’ll just go and work in other fields and have to say we can’t fix it.

And then about a year ago, a little more than that, because I’d kept in touch and I was still on the board of stack overflow and talking to Mike and Joel a lot, Michael Prior and Joel Spolsky the founders of Fog Creek came to me and said we got an idea. What if you came in and help us run Fog Creek because we’re busy. Michael was the CEO of Trello a standalone company and Joel was the CEO of Stack overflow. And at first I was like nope, it’s too big! That’s a tall order, I don’t know if I’m up to it and I don’t know if I believe in making software and tech this way anymore. I really don’t. I can’t believe I’m saying this, it’s an incredible honour, the kindest thing almost anyone said to me professionally and I don’t know if I can do it. Like I just don’t know if I believe that this is meaningful anymore. And , I’ll skip the internal process, but the long story short – because my wife is very insightful and a terrific entrepreneur – said what if this is the way that we can engage with these big questions? What if creating this technology at the company that inspired you in the beginning could be a way of reckoning with what technology will be in culture and the world?

So I said yes, and little under a year ago, I joined this company and I had revered it for so long. A lot of us looked up to it as this icon – they do things right. And I realised I was having my own heavy metal moment because it was like the guy in the cover band getting asked to be in Judas Priest. I was like, I love Joel, I love his blog, I’m a blogger, but I’m not that good. Do you get to go? Am I just like a bad cover version of Joel? And it was this thing like I’ve had dream jobs before, and they’ve been kind of terrifying and scary and which way does this go? And so I started to look at the state of the company. They made these products, many of you used them – I said where are they today? You couldn’t count Stack overflow anymore because they had spun out which is great but they’re a different company now even if they share the same founders. And Trello had spun out. And in fact a couple weeks after I took the gig, Trello got bought by Atlassian which is great! Good news for them and the founders. We still shared an office with Trello. That was interesting because the reason Atlassian could buy Trello was because their flagship product Jira had done gangbuster business and it was a direct competitor with this other product, Fog Bugs
which is our bread and butter. So we share an office with our biggest competitor right now. That was exciting and a lot were millionaires because they’d just sold their company. So lunchtime is fun, it’s really relaxing. No, actually everyone gets along great, we’re very lucky cause the culture was very healthy. But that was unexpected that’s not a thing – there’s zero case studies like this on what to do if you share an office with your biggest competitor. Very little blogging on it, weird! So I was like I wonder what I’ve wandered into. But at least we have this product that’s been around for 15 years and people like and realised it’s kind of not – it’s like this, it’s not taking off. People weren’t like wow Fog Bugs was exciting. Most of you were like yeah I remember that. That’s a bad thing for people to say about your product. People be like yeah! I think I remember! That’s bad!

So I was like oh shit, I did it again! I got my dream job again. And it was a little bit terrifying, and especially because I had to work with a team and these people are incredible, they are as good as you think. If you had the ability to recruit anybody in the world like Joel and Michael did, you get this incredible dream team and I was terrified they will see me and like yeah, this isn’t the brilliant founder we’re used to. We got like Joel-Lite. Michael-like. And that’s bad! You don’t ever want to feel that way.

And I sort of want to pause a minute cause we talked about imposter syndrome here. I’m not prone to imposter syndrome because I’m wired wrong. I also don’t get nervous speaking, my wife is like you’re chemically wrong. For whatever reason, I have inability to normally process things like imposter syndrome. But what I found was this was my version of it, which is totally reasonable – none of us are Joel Spolsky or Michael Prior. These guys represent something. I mean, I’ve seen their talks here and I know about this event cause I was like wow those talks are great. This is before I worked at the company. I saw them speaking at BoS and I’d be really inspired in my own work. I’m here and I’m nominally their successor. What do you do if you’re called on stage with Judas Priest and you’re like I was doing karaoke, I don’t really know.

Then what do you do about deeper problems? When we talk about technical level of 15 year old of code-base you have a lot of technical for a product that was made before AWS existed. They were perfectly reasonable choices back when we were buying Dell servers. This is a different world. And so like that’s normal stuff, you know how that works and that’s work. You have people, who are like those other products that spun out, those were the golden children. We know, they get the attention, for good reason. Millions use them, they’re useful but we’ve been neglected and we’ve been left behind. So you get these deep people issues.

I came in talking a good game cause I care about these issues of ethics and social responsibility in tech and I had people say you say all that stuff on Twitter, but what are you talking about for us? We don’t have salary transparency. We have an old blog post from Joel about how salaries should work or tech roles and we don’t even follow it anymore. Are you just sort of full of it and say the things to look good out in the world but is it real here? And we’ve a healthy culture where people say stuff like that to me and I have to cry quietly. Kidding – it’s very supportive, I cry on the inside. No. I’m kidding! You know what it’s like. You manage people. You want that healthy feedback, but you get all of that at once. And then you have just normal stuff like, we share an office with our biggest competitors and we will have to get a new office so that’s a big project. You have these functional things you got to do and it gets heavy because you realise there is no icon, no model company. There is no place that was this sort of shining light on a hill where everything worked great because they had great blog posts when I still had hair. A long time ago! And that was a little scary, but also freeing. Cause it was like there’s just us, there’s just right now. Doesn’t matter how good someone did things 10-15 years ago. I’m glad I was inspired by it, but it’s all gone. And the talent we bring in today, they aren’t like I read this blog post about 15 years ago about Microsoft Excel. They’ve never seen Excel, they don’t know anything about it, it doesn’t exist! They’re like your grandpa wrote some stuff on the internet too, so what? They read Hacker News. None of that stuff is relevant. So I thought what if I think about from the beginner’s mind? And one, I don’t bring the cynicism that I had sort of cultivated in my critique of the major parts of the industry. And instead refocused on what had inspired me in the first place which was maybe we didn’t have to concede that tech was just defined by the big companies, by their worst excesses of those big companies.

I went back and I re-read the Joel Test and some of it wasn’t exactly right anymore. Like it’s talking about fix bugs before you ship software. Yeah, he was shipping software on physical CDs that we were mailing out to people. Yeah, that’s one way to work but we got a regular SaaS service, you can just ship updates, that would be all right. So they were things that had to be modernised but I lost the reverence for the text. This was just a blog post. And then all of a sudden, I was thinking what did he mean? Why did he say we will do these things? Joel very famously said we will give a private office with a door that closes to every coder. We still do that which is a great thing to do if you’re looking for real estate in Manhattan is tell a real estate agent we need a lot of private offices but it can’t look like a law firm and they’d be like that doesn’t exist! Great! And then I thought what he meant for good software, and it wasn’t the features, the way you test it or the design, but it was those starting principles of building a good company. We realised we didn’t have the answers back then – there was never a point where it was solved or everything was working right, it was only ever continuously fixing these bugs. But the values hadn’t changed, first and foremost taking care of our people, employees, communities and customers. Then trying to be a bit of a model for how we can get things right.

This was where it turned the corner, I realised what we had advocated and lost was being a symbol to others of aspiring to get it right. That was what had caught my eye, my heart in the first place – not that we got everything right, not that Fog Creek was getting everything right back then but that it stood for the idea that you can make an institution that stood for the idea. That you should be trying to get it right and you should talk about that. And infused with this new insight that software matters in big ways, not in little bugs. With little bugs we know what to do, you fix them or you don’t. That’s pretty much it – those are the two options. And you handle that on a schedule and try to be clear and communicate about it but that’s a solved problem, you know how to do that. What we aren’t as good at yet is talking about the bigger problems that technology is revealing, galvanising, accelerating, causing. You know, I think we underestimate how much we may be thrown into the same basket as the worst in our industry. I look at whatever, every day there is some story about someone at Uber is doing, what someone at Facebook is doing or google and they are great people, they make great products. These stories are complicated. But with good reason the worst problems there get a lot of attention. And our futures, our ability to attract talent and to sell our products is bound to the reputations and actions of those companies. It’s not fair nor ideal, but it’s true. It’s just true, it just is.

So how do we reckon with the really big bugs? Not the little bugs, not software bugs, but culture and social bugs. I thought about this in terms of what the team had done at Fog Creek before I joined. I get to talk about it, but it was one of those things where I looked into the product – it’s always been a company that cared about product and they launched a thing called glitch. And it’s a collaborative coding environment and you can sort of multiplayer code in real time. It’s a great product, you can try it out. But the thing that caught me was what the team came up with in the process of building essentially a standard coding environment. We’ve all used that stuff. And they built this feature that let you raise your hand. So you’re in the midst of coding, you catch an error, you flag it and you can just raise your hand and say I need help. If it does that, you go to the homepage and it says Joel is coding this thing, do you want to help him? It’s a really nice sort of social experience. It’s still early on and rough around the edges and more stuff has to be developed on it. But what I had realised the team had done, intuitively, there’s no top down direction on it, was made a place where you can be vulnerable and say I don’t know this, or if I got this right, something is wrong and I need help.

And that was where I saw the intersection of having a set of values of about what we are in the world, what we stand for in the world and a couple pixels on the screen and a button to click. It was this idea that if I feel our industry is not inclusive and diverse enough, then what we can do fundamentally is remind people you’re allowed to raise your hand, you’re allowed to not know. If we feel like a lot of our problems with the way the tech is perceived in society or because we are stepping in domains that we aren’t fluent in and trying to use software to change them, not necessarily respecting the constraints that brought them to where they are. And I think about what would it be like if you could raise your hand and say I don’t know how you work, but I want to have technology help you work better as opposed to disrupting it.

anil dash

And then I thought about myself, I care about tech and want it to be better and the industry to change and I’m not going to do it alone, it won’t happen cause some guy blogs about anything, right? We have to be able to ask each other for help. The thing I thought most deeply about because both founders of our company, they had been here on this stage in front of all of you before and I can’t fill their shoes. That’s not like a false humility thing. I’m good at the things I’m good at but I can’t do what they do. But I can have a slightly different mission which is to say I can be at an institution, company, an organisation that meant something to a lot of you, to me and I could ask for your help, this is what I need. As I hope we’re able to work together to remind the rest of the industry, the people who don’t come to these conferences, who aren’t independent or trying to do things their own way, that it’s important we do things right in our companies and the impact we have on the world. Each of us is that case study, every single one of our companies, somebody is looking at it. Our employees are now going on to be entrepreneurs. You know that feeling, when they graduate? I fell like I got to leave, well where are you going? Well I’m gonna start my own thing. It’s like the greatest feeling in the world, if you’ve ever had it, there’s nothing like it. The sense that they can do this thing. They will model what they see from us or if they go to another company, they will bring with them the lessons they learned. What does it mean to be the case study for having a set of values and being thoughtful about our impact on the world?

The biggest thing I see is that all of us intuitively know what I’m talking about. I don’t have to explain any of this to you. You’re like yeah, there is a way to do tech thoughtfully. But what’s telling is we don’t have a hashtag for the concept, we don’t have a name for it, we don’t have a way to say we want you to participate in this movement that all of us stand for. We have these vague terms like tech industry. If I say I want you to be ethical and humane, who will be against that? It doesn’t mean anything, it’s not specific enough. So the biggest thing I want to ask for your help is how do we describe this ethos that binds everyone in this room, in this community, together and carry that forward to be something that new people who come in who never read these ancient blog posts, might not have met any of us in person, can be inspired by the work we do and understand there’s thought and heart behind it and it goes deeper than the features we see on the screen? I’m hoping we can do it together going forward. Thank you!


Upcoming Events

BoS USA 2018

Don't Miss a Thing - Get BoS Updates

Want us to let you know about new talk videos, speaker AMAs, Business of Software Conference and other event updates? Join the smart people who get BoS updates. Unsubscribe anytime. We will never sell your email address.


Q&A

Mark Littlewood: We may not share valuations, but I think we share values. Questions! There!

Audience member: I don’t know how much you can answer this as the CEO, but I’m interested you said Fog Creek had a great team on Fog bugs and it was a successful and inspirational product, but it’s nowhere near turning into a verb like Trello or Stack overflow. Why and if it matters one is moderately successful and the others conquered the world?

Anil Dash: I have a couple theories. So first I should talk we’re doing a lot of work to modernise it, I think it will regain its relevance through that work so I don’t want to short sight what the team is doing. How did it get adrift? I think there is a couple of things. The first thing is the company took a very successful business and invested that, the profits of that business, in building these other products like Trello and Stack overflow. It was a deliberate strategy and it made sense. The world where you can blog your way in to having a product accepted doesn’t exist anymore. That just isn’t a relevant thing so the idea of using that to sustain growth is sort of not real. It’s interesting cause I had someone, a young person I was mentoring, describe Joel’s software as content marketing to me and I was mortified and I was like I understand intellectually why you would say that, but that was not the goal. This was close to therapy for Joel than it was to content marketing. But their only understanding why a company founder would ever blog is it’s the top of your lead acquisition funnel. It has been but isn’t that anymore. That thing changed. I would say to describe the professionalism of that discipline so how you get software, what you use it for, that project management of software development became an entire industry in a way that wasn’t conducive to a little company serving it. You had to have all this other infrastructure and things around supporting consultancies and this sort of – the substance of the whole field that was not really aligned to this idea if you’re going to go to this one weird company with this terribly named product and install their new server or whatever the model was. So I think those things were all there and the other thing, candidly, there was lost the plot from a technology standpoint. There were a ton of features and they were writing their own compiler and things like that and undone a lot of that technical debt but there was a wandering in the woods period whilst the company strategically focused on the other products. All these things contributed. I look at the future direction where we will be closely aligning with what Fog Bugs becomes with what Glitch does in terms of letting people code and collaborate. That feels to me much more coherent. I’m excited to see if it’s even possible to re-engage with that community in a way that grows it again.

Audience member: Thank you for an inspiring talk! My notes on your talk open with the Joel test and they close with the Anil test.

Anil Dash: No, don’t do that to me! That is not the move! It’s very flattering so thank you! I actually think I’m a very strong believer in leaderless movements, I look at social justice and civil rights movements and it can’t be about a person. And Joel was lucky in that he wrote something broad enough that people could interpret it in different ways. I do feel very strongly it has to be something that is community grounded, but I take your point.

Audience member: I appreciate that too. But I was hoping..we don’t need to call it that. One of the things about the Joel test when it came out is that it was deliberately over-specific. It was trying to convey a series of values but by describing a set of much more specific steps to a point where they might not be as valid 10 years later. I don’t know whether you – how much you’ve given thought to this question, but what would be your list of the falsifiable of the specific that don’t need to be the platonic embodiments of the ideals but are the specific ideas – maybe that’s just you’re doing with Fog Bugs today.

Anil Dash: Yeah, there’s a couple things I see there. One of the things I think about a lot is people give a lot of lip service to diversity and inclusion. And even the big publicly traded companies that are legally required to disclose their numbers, one resisted doing so and two don’t talk about goals. They don’t talk about metrics. So if you say this is something you value, the example I always use is Facebook, they clearly care about messaging. WhatsApp started taking off, and they said we have to own WhatsApp and I think I read in the press took 4-5 days for Zuckerberg to decide to acquire WhatsApp, the conversation happened less than a week. And they are publicly traded so you have to go to your board and stuff like that, extraordinary. And at the time, the offer was made it was a $18 billion acquisition, by the time it closed, $22 billion acquisition. So if Zuckerberg cares about something, he’s able to deploy 22 billion in 4 days. And he says he cares about diversity and inclusion. So I think one billion would be an indication. And like dollars aren’t it, dollars doesn’t magically make everything happen but it doesn’t hurt. And like he won’t be like there’s too much money. I think if I saw from the big companies a commitment of dollars or numbers, so in I think it was 2-3 years ago, in absolute numbers, the number of African American women hired in technical roles at Facebook in a year was 1. So set the goal next year at 2. Right? It doesn’t have to be – I want it to be higher than that, but the point is if there are things where they publicly announce metrics, then you can test against them. If someone on my team was like we don’t have enough uptime on this service and we will increase it to this other number and then 6 months we check in and they’re close but not quite there, ok we can work on it. You can troubleshoot it. Or we overshot, we did great – we have like 10 9’s. Great! That would be amazing! And if someone came to me and says we’re gonna fix this thing, we really care about it. Well, how are you going to improve it? It will get better. How much? Some. I don’t think you’re serious. That’s a really clear example around something but I think if we all came up with this test of where are the ways that we’re measuring these other impacts, I think one of the things that – I don’t have an articulation for it in a metric driven way yet but I think about industries we impact. Say, I’m going to disrupt taxis in a town, you think what the impact will on that community? How do we bring them along and make sure there’s service to them too and it’s upgrading them and not just displacing them? Those are concepts that exist, social responsibility exists, we don’t have to invest it from scratch. It’s just something we have a sort of a stance of neutrality, tech is neutral, anybody could do anything. And then you get these photo filters that don’t work for me and I’m like yeah, did you test it? Could you look at it? Was there anybody my shade that you tried it on? No. Come on, let’s fix it! They’re fixable problems if we say in advance we care about them. And I do think it’s possible to come up with a best practices checklist, I’m a big checklist fan and I go in and just like set up a set of things. I think it’s a valuable way to start.

Mark Littlewood: Just before you ask the question, you were here yesterday when I said thank you to you and Jenn for covering the scholar’s accommodation. That an incredibly kind and unexpected thing that came out of a separate conversation so thank you!

Rob Allen: You’re welcome! So as a follower of yours prior to your ascension to Fog Creek –

Anil Dash: Ascension, I like that! I should have made a slide that just said ascension.

Audience member: Some of the work you’ve done with some of the talks you guys have done together since then, I think Joel made the right choice. You’re going in the right direction and I feel what you’re talking about so my question comes down to how can I, who has less than 1000 Twitter followers, my team is 6, my entire company is smaller than divisions of some of the companies represented here. How can I help? Where do you want me?

Anil Dash: That’s a heavy question. First of all thank you that’s very kind! To put me in the same breath as Joel and Michael, I still – there are very few things that make me nervous. That’s heavy! What do we all have to do? I’m lucky I have a platform and voice and get to be at a company I love, but we all have to do the work. And one of the things, I see this a lot, is people ask me my team is not diverse enough, I want it to be. And I can’t just go and tweet out to the world, hey unrepresented folks come join me! One of the advantages we have in being independent is like everybody here is one of the reasons we guard that is to be able to work on different time scales and not doing having to do quarterly results for investors and not have to do these artificially aligned deadlines. A lot of this work whether it is how do we make sure we are being thoughtful about community impact or how do we bring different voices into our companies, products and communities is about investing time. The number 1 thing I tell people – how can I hire people from this category? Have you spent time in that community? They’re like no, but we sponsored an event. There’s sort of this – there isn’t some shortcut or some magic where everybody in a community you haven’t had any relationships with is gonna suddenly be like we think you’re ok and one of the things to be mindful of is every company says the right thing about inclusion. Nobody says we’re against it! There might be some, I shouldn’t draw those lines. Everything is possible these days, but for the most part there aren’t. And so if everybody says it but nobody is doing it then something is disconnected. I think it’s the follow through which is all of these things, whether we’re talking about changing ourselves, the companies, our cultures, taking on these big initiatives around our values are long and slow processes and so it necessitates making a habit. So the thing I ask of you is what I ask of myself which is every day am I doing a bit on this list of values that I say I care about? I have had people hold my feet to the fire and say I care about this and salary transparency is valuable and empowering to underrepresented folks that they are being compensated properly. And I talked to Leo and Joel at Buffer and they are amazing and like yeah, they’re great! But they started from scratch and it’s easy if you had it at the beginning. Our company is 17 years old, we had people that had been there for 10 years, what are you gonna do? We are gonna try to get there and I said that we are going to try and get there. And someone on our team said, well you said and where is it? I’m like I haven’t done a bit each week. I wouldn’t presume to give you advice, cause I’m still learning. What I see and where I’ve learned to get better at is if I haven’t moved the needle a bit each week I will never get there. So I decided if there’s one thing about compensation about being fair for every group and every identity – that’s where I have to make progress every week or reporting why I haven’t and being accountable. And that’s something where I’m lucky because I have a public platform. I try to use that to do that myself and I reveal the work and it’s something I learned from Michael and Joel, being transparent about it all and it makes you accountable. The biggest thing we can do is find the peers who care about it. If we hold each other accountable and get in the Slack channel and do a check in with each other, how are you doing on these things that we say we care about, that’s it. Then the other part is identifying which of those things you care about. You can’t fix everything at once so picking which of those you want to do first is really important.

Audience member: Thank you! Real quick, put you on the spot a bit following up what you said here. Can you give us an example of a number you are aiming to reach to improve any aspect of this?

Anil Dash: Yeah, I’m trying to get above 0 for revenues on glitch. We just built the thing and …I’m being glib! I will tell you something – we build a coding environment, the idea we have of how and who it will serve is really like Dev rel and people – and the vision I came in I was like ok we will have this product and it will help the people sell their APIs to developers and be able to measure how it works and it’s a great idea! And what we’ve come to find out is that most dev rel people on a product stand point is most have budgets for stickers and t-shirts, they don’t have a tools budget – they haven’t bought software before. Whatever the marketing team has Hubspot and sales team has Salesforce, whatever the tools they are using and like some CRM tool and our dev team has nothing. And that was a struggle, it was really hard to say we have this vision of this market we’re going after. And we’re credible, people in Dev Rel know Fog Creek so we aren’t starting from scratch; and this idea well you don’t actually have customers who have the ability to spend money. That was a fun conversation to have with our sales team. I sit with our sales team and when you get into month 2 of we don’t know if this market exists, how do we see this and build this? That goes from keep trying, sport, to let me see some numbers. So the thing that I think we’re translating into is business fundamentals are really key. We have to have that accountability and that was instructing cause where else are we at zero? Obviously that’s a sales thing and we’ve all struggled with these issues of how do we grow this channel? We do have parallel issues around, and a particular one for me was communication. Our team is ¾ remote and I was still doing walking around management in our HQ so at max, 1/3 of our company would know what I was talking about. And it was my own lack of experience from knowing how the company culture worked, that conversation should happen in slack or hangout or whatever, this sort of group environment. Part of it was being tasked with making sure I’m doing it at least once a week, a really broad strategy, what are we doing and how has the progress been made, where are we seeing successes and failures? And it’s funny to have the number be 1 but that was really it, once a week, being able to do that. It took me a long time to get in that habit, I’m still not perfect at it. A lot of that was I hadn’t listened enough to what the culture was telling me and it’s interesting cause there are positives to that too. The remote culture works really well, but I wasn’t keyed to it because I was new. And that was one of the things that was challenging and became a clear – it wasn’t just measurable but binary, did you do this this week or not? That was very clarifying for me.

Audience member: Hi! I’m the one who is trying not to fan girl too hard, but I thought it might be a great finishing point to recommend to a lot of white people in this room who are 3 women of colour that they should take out on their phones or computers and follow.

Anil Dash: That’s a great question but it’s hard! The first one that comes to mind is Tressie McMillan Cottom. She is an author, her last book was called Lower Ed and it’s about the sort of commercialisation of education and its privatisation. Very thoughtful book, but she is among many other things a digital theorist and is thinking about how technology is impacting society. That’s a really great starting point, I think Tressiemc is her Twitter handle. I’m sure if you search for it I’m sure it will pop up. So she comes to mind. I think everybody in tech should be watching what Ellen Pao is doing, she is EKP on Twitter. A lot of people will know her lawsuit for discrimination against Kleiner Perkins. And then her tenure as CEO at Reddit. Her book came out. She was an incredibly valuable mentor to me in terms of teaching me how these things work. I think one of the things a lot of people care about values and inclusion and we care about the ethical course of the industry and we’re uncomfortable talking about it. I know I don’t have to ask for a show of hands, I know I talked to a lot of you about this. I care about it but I don’t know how to say it, I don’t want to say the wrong thing. I’m trying to be welcoming but I know that it’s really easy to get one word wrong in a tweet and then I’m being dragged for 3 days online – I get it. It can be very fraught and contentious to dig into these things. I think people like Ellen are great leaders that give us the space to have those conversations and her book is a good starting point but even the conversations around her are a great way to start. There’s so many. I try to narrow it down. So anyway we will Tweet the follow-up and the other is Tracy Chap. She had been at Pinterest and started an effort amongst mostly women but workers at a lot of different tech companies to document the number of women in tech organisations. It was very simple, like how many coders are in your company and how many are women? This is the power of a spreadsheet, I’m a believer in them. She made a spreadsheet and crowdsourced data for it and it’s been transformative. A lot of companies had these conversations about sharing representation and then compensation and all those other issues catalysed by this simple effort. I think again, just such a thoughtful and simple solution to like the more we can talk about these things, the more honest we can be, the more we can have a dialogue, and move the ball forward. So I think those are all great role models for the conversation and a great place to start. Thanks for the question!

Mark Littlewood: Thank you! Anil, thank you very much indeed!

[Boilerplate]