Welcome to the 19th episode of our podcast!

You can download the podcast to your computer or listen to it here on the blog. Click here to subscribe in iTunes.

CP19: Talking with Avi Cavale from Shippable about Entrepreneurship & Shipping Coder Faster


Martin: Hi folks out there! If you are interested in starting a company or you are a developer and really working on your next product, this podcast interview is for you. Today we have Avi with us. Hey Avi, who are you and what do you do?

Avi: Hi Martin, this is Avi Cavale, I am a co-founder and CEO of Shippable – we are a continuous delivery pipeline company for containers.

Martin: Great! When did you start this company and what did you do before?

Avi: Shippable, the genesis was around late 2012, physically incorporated and everything in 2013, early 2013, February. It’s been about 3 ½ years so far. Before that I worked at Microsoft and briefly at Amazon and I was there at Amazon for over 11+ years, worked on xBox, Office 365, Azure and Kinect, the xBox live products.

Martin: How do you go from gaming to shipping code?

Avi: I mean, as recently said that “Software is eating the world” and pretty much everything is software. And even though we were at gaming side of the company, that’s what the consumers saw, but behind it, it was all pure software technology. While I was doing that, I was managing about roughly 150+ organizations across 3 different countries (China, India and the US) and at the end of Kinect when I was going through my review process with my boss, I realized that over 50% of my entire organization spent their time on none feature work. I mean they were doing all sorts of plumbing, deployment, code auto-pilot, code all sorts of things. And that’s kind of as you look at it we worked on Kinect for 2 years, 150+ people, that’s almost 150 person-years being wasted on things that should have been actually big new platform. And that’s basically what our motivation was.

We said: “Hey, every single group inside Microsoft is doing this and imagine every single company in the world. What it should be is more available as a platform where developers are using it to build features as opposed to kind of building the plumbing work for deploying the features that they have built. And that’s basically what Shippable was.

Martin: Cool. Avi, what was the next step? So you left the company and then what happened?

Avi: I left the company and I had never done a startup. Biggest thing was I left it in a very impulsive way because I was so excited that this is a problem that I could go solve.

I think for about 3 weeks, I was kind of… I mean I left in late 2011 saying that I want to go do this company, and I meandered around for about 3 or 4 weeks. I realized that I just didn’t have the skills to do this at all especially founding a company.

Martin: Why?

Avi: And let alone do it alone. I mean that’s the biggest mistake I did was, I said: “I am going to do this alone.”

And then I said: “Okay fine, I am going to go… I kind of parked my idea a little bit” and I said: “Okay, let me go work on some modern technology.” And I started working on a company which was working with Cloud Foundry, and I learnt all about how startups work, how actually this thing goes on, and that’s basically, I kind of trained myself on the job for about 10 months, always knowing that I will eventually get back to this idea that I was going to go do a company about.

Martin: And how did Cloud Foundry prepare you for the entrepreneur route?

Avi: I think the biggest thing that happened was, at Microsoft, being in kind of like somewhat of a leadership role, you don’t realize how easy it is to get meetings, and so you lose the hustle of your life, and things happen very easily because of the background and the brand that Microsoft has.

When you start trying to do it on your own as a company, it’s impossible, I mean you have to really hustle to get these meetings and get these people to help you out. And what Cloud Foundry, this startup that was working with Cloud Foundry did was kind of make me get used to that kind of mode as opposed to sending an email to a company and say: “Hey, I want to talk to you for half an hour”, and I would get that meeting set up because of just the fact that I was working at Microsoft.

Martin: Good. So once you have acquired some kind of knowledge from Cloud Foundry, what did you work on then? And did you find a co-founder?

Avi: Yes, I mean the other thing I did was I realized what my limitations were, and then I started finding folks who would counter balance the limitations that I had.

One was I was purely a techie person; I needed somebody who has a little bit more of a business background. We are still a very technical company so they need to be still technical, but they need to have more business exposure than I did.

The second thing was, even the temperament. I mean I kind of think in big vision, kind of goal setting views, whereas you also need to have someone who brings you back to the ground and gets some execution plan to get towards that big goal that you have. And so I think Manisha was kind of my obvious choice. I convinced her that she should quit Microsoft and do this. She quited Microsoft in late 2012 and we started Shippable in 2013 in February.

Martin: Did you know Manisha before and why do you think she qualified for a co-founder role?

Avi: She had worked with me at Microsoft back when I was with Office 365 and that is where I had interacted with her. She was part of a team that my team actually worked with.

There were 2 things: one was that she was super smart and I knew that with all my interaction at Microsoft. The second things was she was very practical in how she thought about things and I was kind of the big vision that she was a very practical person and I thought that that’s the right mix that I needed. And of course she had an MBA degree from Berkeley which made it even more easier for me to say she has to be the person who is the co-founder.

Martin: Great. So now you have assembled the 2 of you and starting out building on the product version I guess, so what was the next step then? So did you work on the product or did you try to acquire some kind of beta customers or did you already talk to investors? How was it like?

Avi: So I think we kind of did a few things. I mean one thing that she kind of said is: “We’ve got to have some kind of a framework of how we are going to go about doing this.” I mean I wasn’t too keen on having this big MBA like frameworks but I wanted something that offers structure.

So what we did is, we started looking at accelerators and we filled out their application forms. And we didn’t want to apply to any accelerator, we just wanted to fill out the application form because some of these accelerator application forms really make you think about what your business is. And that’s what happened.

So we started off with Y Combinator and we filled out their whole application form and then we also accidentally got in touch with a few folks from Techstars and this was due to some random events that happened. And then, I mean I would say we were lucky meeting the managing director of Techstars. And then we ended up taking Techstars as an accelerator.

And by that time, we had kind of built a prototype just to prove the technology can be done. And we knew that the customer problem existed, but we hadn’t done any formal customer development and those kinds of things. And at Techstars, when we went to the program, they had a lot more structure on how we go about doing it and it helped quite a bit in terms of how we did the rest of the company.

Martin: Okay. How did you go about the customer development, once you have been into this Techstar program or afterwards?

Avi: I mean it was hard for us to accept that customer development was very critical.

Martin: Especially as a techie, right?

Avi: Yes, I mean we thought we knew more. And actually in retrospect if I go and look back that was probably the best thing that we ever did at Techstars. Andy Stark who was the managing director of Techstars Seattle at that time, he almost had a stick for us to say: “You have got to do customer development.”

The easy thing for us was that finding developers in Seattle was not very difficult. And so what we used to do is we used to go out in downtown Seattle where Amazon was, there are all kinds of food trucks where people are standing in line for food during lunch, and we would just ask them saying: “Hey, I’ll buy you a soda if you answer a few questions.” They were all techie, so we could easily get 50-70 interviews a day done in relatively like an hour and a half over time, which was kind of unfair for the rest of the companies in Techstars because our customer base was a developer and we could find them so easily.

Martin: What did you ask them?

Avi: I think Andy’s guidance on customer development was – you can never tell them what you are building, and you have to somehow ask them questions which don’t tell them what you are building. If they don’t answer the core fundamental value proposition that your product does as their main problem, then your product will not actually sell. I mean that was kind of the philosophy.

So all our questions were all about behavior; what do you guys do on a daily basis? What is the one thing that you would want to do less? So it’s more of open ended, trying to drag out what they really are doing and trying to see whether our value proposition actually stick in terms of what the pain point or what we are really trying to solve with our product.

Martin: Okay cool.


Martin: Avi, let’s talk a little bit more about your company Shippable. Can you briefly explain how the business models work? You briefly touched on this, so what are the customer segments? How are you making money, and what type of value proposition are you offering those people?

Avi: I think the basic idea is that our product is a premium model, so it’s like so as like pretty much everything that’s done in these days. You have a basic value proposition which solves some problems for maybe an individual or a small team like 3 or 4 people, and then the moment they grow beyond that, then you need to start buying in order to actually get them to start becoming a paying customer. I mean that’s basically the get up model that is basically pretty much a lot of these online services model. That’s how our basic fundamental, what we call as demand generation.

And then with that what happens is, you don’t have to do traditional enterprise sales. So you get a lot of people getting ground up, and then eventually they will get to some size at that point they become a sales qualified lead to us, and we kind of sell them a more of an enterprise kind of product on top of it. That is basically the evolution.

So you come in as freemium; you become a paying SaaS customer and eventually, you will end up buying it for your department or maybe for your organization within your enterprise. I mean that’s kind of how the business model works.

Martin: What makes Shippable unique in the market place? Because I mean you need to think, I guess your co-founder as well because she is having an MBA, what are the competitive advantages over your competitors?

Avi: I think the most important thing is it is about efficiency, right. It’s like if you really look at it in today’s world, every single company has to out innovate the competition. And that’s basically the only way you can actually survive. Otherwise your features are going to get copied very quickly and you lose out on the differentiation. So that’s a platform that every single company that’s out there needs, I mean, if you kind of look at any mobile app kind of a company or anything, they have to constantly keep adding features that makes them more valuable and have more unique features than their competition.

So we are a very interesting product. I mean our product actually helps other customers to do this. So what we do is, we use the same platform to build Shippable. So what we are doing is, we are helping engineering organizations to become more efficient. In other words, we call it ship code faster kind of qualifying it a little bit more, its ship quality code faster and repeatedly, and that basically is continuous innovation. That’s what our product does. Hence, and the best thing for us is we are the number 1 customer for this product because we are trying to out run everybody.

So if you go and look at the last 3 months, we have added over 25 different features that none of our competition has even added even less than 5. That’s basically what makes our product more complete and we are constantly innovating using our own platform to kind of help other customers innovate faster on their whatever product that they are working on.

So that is basically the uniqueness, so it is a continuously evolving platform that helps developers become more efficient to build software.

Martin: Okay, cool.


Martin: Over this 4 – 5 years, what have been the major learnings and mistakes that you have seen or done yourself which you can share with our people interested in starting their own company?

Avi: I think some of the mistakes are like very specific to our business, I mean our customer base, our developers. So a lot of mistakes that we did around that was not listening to developers closely enough and so they were asking for something and we were trying to build something else. So that was one mistake that we did, 12-18 months ago. And in the last 6 months we have been super focused on our developers and our customers, that has really changed how we actually are perceived by our customers. I mean that’s one thing that we never want to do; is alienate your customer base.

Even though you might think this is the right way to do it, you have to always have your pulse on your customer base, that is one thing that we did about 12 months ago, the wrong thing we did. Since then, we have post corrected and we have become a completely customer centric company at this point of time.

Martin: Was there a key even where you said to yourself: “Oh wow, we are really missing a point here, we need to change”?

Avi: I think what happened was 2 things. One was that we ended up opening up our customer support queue completely to the public. It was a decision because internally I had to change the culture of the company, and what ended up happening is the moment we opened up, kind of airing your dirty laundry out first, I think that’s the first thing that we did. And that was a very risky thing to do because it pretty much told our competition what the problems we had were. And so the moment that happened a whole bunch of customers started actually commenting, cross commenting on it because now everything was open and that internally changed our entire teams mind set. So that was a turning point that basically said: “I mean I know everybody has dirty laundry, but we want to keep them as clean as possible as quickly as possible.” So suddenly everything became customer centric as opposed to hey we will address it when we get some time, that attitude changed, I mean that’s basically what happened from inside the company.

Martin: And did you use a tool for this?

Avi: I mean we used Github and we are a developer platform. So it was very easy, we just opened up. It was private, we made it public, that’s basically what we did.

Martin: What other things did go well or did not go well which you can share?

Avi: There were  a few other things. I think hiring is very important. And sometimes I think you need to hire for attitude as opposed to aptitude, and most startups end up making these mistakes where they hire for aptitude as opposed to attitude. And when you are this small, like when you are like 5 person, 7 person, 10 person company, if 1 or 2 people don’t fit, it really causes a lot of problems within the company. I mean that’s another mistake that we did where we got carried away by people’s aptitude as opposed to their attitude and whether the culture and the fit is going to be right.

So I truly believe that in the initial product of your company, you should be super focused on building the cohesive team, even if it is not the best super star team that you have. You probably want to have a team that works as a team, as opposed to an individual excellence. That was a couple other mistakes that we made. Since then we have post corrected that and that’s the hard part; it’s letting go of some of the earlier employees that we had hired because they were just not the right fit. I mean it’s kind of like my board once told me that if you don’t start being a CEO then we’ll start finding a CEO. So I mean I was kind of the message to say this is part of being a CEO, you have to make these hard decisions. That was a wakeup call for us which kind of re-jiggled the company into the right direction.

Martin: Great. Also a good example of how the board can also help advise the founders, which is great.

Avi: I mean this is the other thing for a first time entrepreneur, like this is the first company I ever created, and even for Manisha, we both were completely novel at this. You need to understand, and I think for whoever is trying to do this, they need to understand that there will be highs and there will be lows, and there is no company out there that hasn’t gone through that floors. And you have to be completely okay with that.

That’s something that people always remember the great home runs that you had, but they don’t understand that there is a whole bunch of failures also internally happening. So that’s normal part and parcel, and that’s where the board can help you because they have seen hundred of companies go through this. And so you have to use your board quite a bit, as opposed to just trying to do these things in isolation.

Martin: You said very nicely that you need to hire in the beginning for attitude and less for skills and aptitude and so forth. I mean the CV and drop test for testing people for skills are very well defined. how would you test for attitude?

Avi: I think what you do is, I do the special thing called case based interviews. I mean I don’t ask people how to do things, I ask them why they do what they do. I mean it’s very easy, if I ask you to sort these numbers in the fastest way, everybody will tell you how to do it. Very few people can answer why one technique is faster than the other technique.

And so it’s kind of like my personal favorite of what I am saying is I don’t really care is people tell me what to use when we need it. What I really need is people to tell me when not to use a particular technology.

So when you do case based interviews, you start seeing that whole whatever the candidate does pros and cons, whether they are kind of looking at it more holistically or are they kind of getting carried away by one single piece of information. So that gives you more of a well rounded approach of what the candidate is doing, as opposed to just asking them a few technical questions and seeing how deep do they know the syntax, or how deep do they know sorting algorithms. I mean, that can be all found on Google, you don’t really need to know all those stuff. What you can’t find on Google is when do you use Non-JS as opposed to use goal or vice versa. I mean that is a much harder question to answer than just saying: “Hey, we should use goal.”

So I think that’s the kind of questions you want to ask more as opposed to just asking very tactical problem based questions.

Martin: Avi, if you look back from today to like 5 years ago, what would you have liked to know before you started the company, which would have helped you to become even better?

Avi: That’s a hard question. So what would I like to know, what information I have to have 5 years ago? I think what I would have done a lot more was. It’s basically a very hard question to ask because everything happened for a reason, and it’s all about how do you react to the things that happened as opposed to trying to control things not to let it happen. So I don’t think I have anything that would have changed my game, I like the journey I went through and it made me who I am today, so I don’t think I would change too much.

Martin: How did you change over those 5 years?

Avi: I have a lot of respect for people who have built companies. I mean if you would have asked me 5 years ago what is the role of a CEO, I would have probably not been able to answer. I think just having empathy towards the different roles and the different skill sets that people bring to the table is something that… I mean I was too much of an engineer, I only valued engineering skills and I kind of discriminated on the rest of the skills others had.

So I think, being a CEO for the last 3 years, trying to create a company from scratch, I have a lot of empathy for pretty much every single skill set and job role that’s out there and people who actually do that really well. I mean, you need every single aspect of it, like from people who can do content writing to people who can actually code, to actually market. I mean you need to have a well rounded team, and I have developed a lot of empathy towards the rest of the skills that people have.

Martin: Great. Avi, thank you so much for sharing your insights.

Avi: You are very welcome, I hope it was useful and it will be useful for some folks who want to listen to this.


Thanks so much for joining our 19th podcast episode!

Have some feedback you’d like to share? Leave a note in the comment section below! If you enjoyed this episode, please share it using the social media buttons you see at the bottom of the post.

Also, please leave an honest review for The Cleverism Podcast on iTunes or on SoundCloud. Ratings and reviews are extremely helpful and greatly appreciated! They do matter in the rankings of the show, and we read each and every one of them.

Special thanks to Avi for joining me this week. Until next time!

Comments are closed.