By Leah Culver (Founder & CEO, Convore) For my second startup Convore, I applied to Y Combinator because I wanted to be part of their alumni network. It's a great way to test out a product -- with thousands of Y Combinator alumni. Convore was accepted and part of the Winter 2011 group.
My motivations now are very different than when I just started doing startups. I'm not in startups for the money -- there are a lot better things to make money on.
The reason I am doing startups personally is that I kind of see programming and creating software as an art form -- I want to make something really good, and really interesting, and the making money part is a side effect of that.
Lessons Learned People underestimate how emotionally taxing startups are, especially your first startup.
At Pownce, I was always stressed out -- it was a lot of pressure. You also don't want to let people down, and you don't want to let your users down. One of the things to think about is if you want to have that kind of pressure.
My second startup is much less stressful because I know what is important and what is not, and what to stress over and what things will work themselves out in the end. I think this is why a lot of investors will invest in a second-time founder; if you are willing to do it a second time you've adjusted to that lifestyle and your company is more likely to work out.
Learn not to add too many features right away, and get the core idea built and tested.
People can't handle too many features at once and you need to get your core idea. We struggled with that at Pownce and we also struggled with it at Convore. It's important to have a core use case.
As an engineer, I struggle with narrowing down the scope of a product. Anything is possible, so I build anything I want. It's not, "Can I build it?" -- "Should I build it?" is the better question. Making simple products can be difficult for engineers.
It takes a long time to build a company and to get people used to a product.
I often say that people won't like something until the third time they try it. It's exciting to launch but you need to find a way to get people to come back and keep using your site. That's where startups fail -- between launch and success, it's hard and painful and seems like it has no end.
There are some companies like SoundCloud that have been around for ages, and they've only been popular for a year or two. Foursquare (as Dodgeball) has been around for 3 or 4 years before it really took off. Sometimes it just takes time for a product to mature.
The Beginning of Convore
I have been thinking about the idea behind Convore for 3 or 4 years -- a really long time. I really wanted to create a real-time chat application. From Pownce, I saw that people had a desire to communicate with each other in real-time. Users would start a comment thread, and spend the entire night talking with each other just in this one thread.
So I wondered if there was a better communication solution out there. There are tools that allow you to chat in groups on the web but I don’t think the space has had a lot of innovation. Things like IRC are fairly outdated.
We built Convore in just 2-3 days and we got a handful of friends to test it out. Of course it didn't work properly, but at least the idea was there. Then we added a really rough design and took it to Y Combinator. It was 2-3 weeks old and at that point, we were able to show people outside our group of friends and get feedback.
We kept adding more stuff, and at some point we released it to our Y Combinator group. We later launched Convore with all of the Y Combinator alumni, and by the time we launched publicly we had 500 users. Paul Graham tweeted that we were really good for Y Combinator, which was really nice because he almost never tweets!
At each stage, we added a certain set of features, like email notifications, a better design, Twitter login, etc. Then we gathered feedback to validate. Luckily people left a lot of feedback in Convore itself. Getting feedback from a lot of different people was super helpful. I wouldn't do invite-only again like I did with Pownce, it was more helpful to launch in stages like we did with Convore.
For tracking, we added a page right away that showed us metrics and stats that we were interested in, like how many people were logging in every day. We also talked with users in person to get their feedback. We use Google Analytics for basic stats and created custom graphs with the Google Charts API.
Getting Started Programming
I was a programmer first before I was an entrepreneur. I've been a programmer now for 10 years, and I don't even feel that old! The first few years I was programming I didn't anticipate running my own business, but I always did side projects. My first side project was a HTML game. I recently put it on GitHub so people can look at bad my first application really was.
You always make fun stuff on the side, and then at some point the side stuff becomes more important than your day job. That's how a lot of programmers get started on their own business. It comes from wanting to make really cool stuff instead of making other people’s software all the time.
There is a San Francisco Django Meetup that happens once a month, and there is pretty good free documentation for Django. Other languages, not so much. I wanted to create iPhone applications and I ended up buying a book for it because I couldn’t find enough good documentation online.
There's a difference between learning a language and learning programming though. Learning your first programming language is really hard. It's very different from what you're taught in school, and it is difficult to grasp the core concepts. I don't think you can learn programming in one day. There's a whole spectrum of things you need to understand to create successful software.
I majored in Computer Science in college, which forced me to learn on a wide variety of programming topics. I don't think it's easy to dabble in programming and it takes a long time to teach yourself. However, once you learn the basics and begin to love programming, it changes you forever.
Editor's note: Got a question for our guest blogger? Leave a message in the comments below. About the guest blogger: Leah Culver is the Founder and CEO of Convore, a real-time group communication application. Leah was formerly a co-founder and the lead developer of the social network and micro-blogging website Pownce, which was acquired by Six Apart in November 2008. She has co-authored both the OAuth and OEmbed open API specifications. She developed an iPhone application for Plancast. Leah promotes open source, APIs, and the Django web framework on her blog at leahculver.com. Follow her on Twitter at @leahculver.