Tag Archives: startitup

How to Catch a Technical Co-founder + How Paul Graham Asked Me to Build Pinterest in 2009


We’ve launched StartitUp for just over 2 months now, and we are now helping more than 1100 startups build and grow, and we’re growing quite fast. However, it took me 9 years and 1-and-a-half failed startups to get here.

This is a story that I’ve never told before, but I thought I should share it just to document my failures and what I’ve learned from them.

PS: the post is quite long. Will take at least 5-10 minutes of your time :) If you don’t want to read about my awesome story, skip to the end and read the summary.

Wolverine vs. Pussy Chinese Dude (on Halloween)
Wolverine vs. Pussy Chinese Dude (on Halloween). Me on the right.

In the year of 2004, I had just graduated from University of Connecticut and had just moved to Palo Alto for my first job in the bay area. At first I was pretty oblivious with the scene there, as I had graduated with a major in electrical engineering back in UConn. I was still thinking hardware, and for people who did EE in college, you know that even though we wrote programs in Matlab or something, we really didn’t understand much of “software” in terms of internet or web applications that are responsible for the popularity of Silicon Valley.

Well, being in the heart of all the innovation, I soon started my path to become a fanboy in the startup business. First it was just TechCrunch. Then it was Mashable. Then I needed Google Reader to help me keep track of all the glorious battles that are happening on the internet. I had Firefox (no longer Internet Explorer – if you know what I mean) set up to open Google Reader right after I open Firefox.

The more news and acquisitions I read about, the more that ticked my inner entrepreneur. By 2007, after all lines of ideas I’ve written up in my notepad, I finally made up my mind to do something about it. Just to let you know, at this point, I didn’t know any functional programming, but I’ve had a few years of using Dreamweaver, so I can put together shitty and static landing pages.

So, I started out with my best idea, which was an add-on to extract the conversation between people on Skype and analyze it to deliver contextual ads using Google Ads in real-time. So lets just say if you’re talking about cars, then at the bottom of your Skype window you’d probably see a BMW or Jiffylube advertisement. I was so excited, I even wrote a patent about it. I didn’t have much money so I had to write the patent myself, which later was rejected by the USPTO (US Patent Office), to which I appealed but was rejected again. BTW the process took years, just in case some of you are thinking about it.

Also, a little secret is that I left my patent printed out on A4 30 pages stapled together at the door of an important Google senior executive’s doorstep with 10 dollars (trying to be witty and humorous), hoping that my patent would be read and I would miraculously get some kind of offer or funding for it. Well, I didn’t. You just kind of start to think there’s this kind of shortcut or fairy tales where you’d just make it big. Well, you don’t.

Anyhow, soon I started looking for a technical co-founder who could make the idea work out. I asked a friend (whose username is pimpnutz) who worked at Cisco, and he was pretty interested in the idea. So the division of the work was that I would handle the front-end design, and obviously he’d do the heavy lifting. We held weekly meetings at his house and would set weekly goals for the following week. So things kind of went on like this for a few months before I realized that we actually never got anything done. My friend finally told me that he was very busy at work, and he just got a promotion and he was making enough money so he didn’t have his mind on building a startup that would be a lot riskier. The truth is, I think the balance of responsibilities was kind of off, and I could feel that it wasn’t right as well.

I really didn’t blame him, because I knew that I really couldn’t offer him anything, and he would obviously feel pressured having to do all the important work while I kind of just kept piling on features and deadlines on him. So at the end, I realized that I needed to depend on myself, just in case shtf. I finally decided to pick up PHP. I downloaded a bunch of open source scripts like PHPMyDirectory, SocialEngine, and a bunch of other clone scripts to which I tried to modify to become what I wanted to build. Well, I didn’t learn it learn it. But I already knew which part’s which and I could copy and past or edit some small stuff to make the revisions I wanted.

So after that I kind of launched a lot of things on my own, to all which I got very little traffic for, and people were using it, but I just kind of drifted away because I couldn’t grow them any further, and they would kind of grow and then they’d just die.

I’ve also bought this proxy site (nomorefilter.com) for $250, which was making about $50-$80/month through Google Adsense. So I kind of played with that for awhile, finally getting a task of making money on the internet. But then of course, I didn’t make the website, so I kind of just bought that experience.

Anyhow, I learned a lot on my own just playing with different scripts and just reading a lot of stuff, but I wasn’t really good enough to build something on my own, or attract a rock star programmer to join me and build a startup with me. I really still didn’t have what it takes to build something.

Fast forward a year (2009), and I’ve moved back to Taipei, Taiwan – my hometown where I grew up. I then begun to work on Swagly.com, which was an in-image advertising network where you can tag products in images and then corresponding advertisements on those products would show up. Again, I also spent a few weeks writing a patent for it (which was also later rejected), and I started building up a mockup of the whole website, which included the landing page, the sign-in page, the inner admin panel, the widget that you would plug into 3rd party websites using a Javascript. So at this point I kind of had the ability at least to put together what my idea is so someone could build it. However, at the point, my design skills are still far from being worthy of mentioning.

Swagly's product flow
I made this with Photoshop, and you can see how shitty it was back then.

At this point, I started looking for a co-founder again. I asked around and I found this guy who a lot of people wanted to hire, and who actually is pretty good at programming. So what happened was that this guy actually did outsourcing for companies, and he was working on a lot of projects at once, so when I mentioned about my idea, he was interested but wasn’t sure if he would have enough bandwidth to build it for me. We had a great talk, and I had put out my best sales skills and asked him to partner with me, and that it’s so much better working on his own product than slaving away for other companies.

A week passed and we talked about it more, and he agreed to work with me. However, he mentioned more of an outsourced kind of partnership where I’d pay him around USD$10k to develop this, and he would own a very small part of the company (I can’t remember, but probably like 10%?). Okay, so we started to build Swagly.

So during the development, it was pretty hard for me and him because he had a lot of stuff he had to do for his existing customers, and as a startup, we couldn’t really wait for bugs to be fixed, or when we just needed a new small feature quickly. Also, I originally had envisioned the advertisement to appear IN the picture as a mouse-over popup, but my co-founder mentioned that it was too difficult or too much work for his team and we’d rather just deliver the ads in a widget box at the bottom of the image. Well, I thought okay, as the MVP that should be fine. So due to me not being able to do more on that end, I had to agree to the external widget that took up more of the publishers’ real estate (which customers don’t want).

So what happened was that we spent like 3-4 months building a demo that wasn’t scalable. Something we could only use to show to people, but it wouldn’t be good enough to deploy to customers. I actually didn’t know the app was so inflexible until we had gone into details of their reporting. My bad too. Having no choice, I decided to ask the team to scrap the demo and restart the whole project to make it into something that’d scale. I did this because I actually was sorta already talking to customers, and we had this validated and we needed a working product, no longer just a demo.

After another 3-4 months we had something sub-par but at least we could show it to the customers that I have been talking for months. So actually at this point, I was doing a lot of customer development. I was actually talking to MySpace, MTV, and some major entertainment blogs on partnerships, and most of them were pretty excited about using it, because they knew they needed a way to monetize their images, or just simply monetize.

Swagly even got onto Techcrunch on an occasion when they came to Taiwan and picked out only 6 teams. So that really helped us too. However, we got thousands of visitors that week, but the traffic quickly dropped off, leaving us with a forceful disappointment. We even had many investors call us who were interested in investing in us, but I turned them down because I was a firm believer of bootstrapping until proven product. Bad decision on my part as well.

So okay, the product was ready, and I called all the customers again, and they were excited to see the product. We passed the products over to the customers to ask them to test it out. They liked the idea, but there were many thing they wanted changed before they could use our widget/plugin. So I had sent the request back to the team, but then the team was too busy working on other projects that would actually make them money, so my customers had to wait. The wait took weeks, and even though I had kept the customers updated weekly, by the time we had the features fixed, they were pretty lukewarm about the idea already. So okay, I lost all of my customers as they saw we really couldn’t respond fast enough.

I never brought this up to my co-founder or the team because I thought this really was what I chose to do, and I didn’t have the skills to build a product of this scale. It was really my fault.

It’s okay I thought, my then co-founder and I got an invitation to go interview with Paul Graham at YCombinator in 2009. They gave us 15 minutes, and it was just me talking even though my co-founder was there because his English wasn’t good enough to be helpful in the interview. So I think PG actually liked our team, but he was very doubtful about the idea (he was right). He even asked to speak to us again for a 2nd round, an opportunity that nobody else in the room got. The 2nd time he basically told us that they like us, but asked if we were willing to pivot and work on a project that could let people post stuff that they like on their personal catalog and show it to people because it somewhat overlaps with Swagly, and we could use our existing database and program to build that. Pinterest anyone? There you go. PG asked me to build Pinterest in 2009. To my utter stupidity, when he asked me what I felt about that, I responded that I obviously felt a bit disappointed that we had to change our idea, but if possible we’d still like them to reconsider our idea for the bootcamp.

Later in the evening, we got this email from Paul Graham:

“I’m sorry to say we decided not to fund you guys.  It was a hard decision, because we liked you a lot.  The problem was that we didn’t buy into your vision of having users tag stuff in their personal photos, and when we tried to talk you into the alternative plan of more deliberately organized shopping guides, we could tell you weren’t really into it.  We’ve learned it doesn’t work to push people to do things they’re not genuinely excited about.  But we definitely thought you have the qualities to be good founders, so we’d like to hear from you again in the future if you have new ideas or continue to evolve this one.”

(Actually, as I went back to the email to copy and paste it, I realized that PG had intentionally formatted the width of his email to be narrower than the usual full width so it’s easier for people to read. Wicked.)

At that time, I was like…oh well, what does Paul know, and then I kind of went home happy that it’s a good thing that I didn’t go join an incubator that didn’t see how awesome my idea was. Boy was I wrong. Well, I later learned that you should always be open-minded, and always be open to pivots. Also, ideas are bullshit. It’s really the details and executions that the team brings that really contribute to the success. Fuck me.

Okay, so we went back to Taiwan, and I was running out of money. I was already quite sick of the whole Swagly situation where I had no control over it, and I was just kind of waiting for things to happen. I decided to shut it down, and just cut my losses after about a year and a half of being a complete retard.

I picked up a new job (2010-2011) at famous server manufacturer, a competitor to the company I was working for when I first moved to the bay area. Anyhow, at this point, I was more modest (I still need to work on that), and I was learning a lot about cloud server infrastructure on the job, and I even built a 2-node cloud cluster in my room using Eucalyptus that was basically the same as what Amazon’s AWS cloud hosting is using. I also had time and gathered my patience to learn PHP with CodeIgniter as the framework. I also read a couple of books on Java, and Ruby on Rails. I was really learning more than ever. I also became a SEO Jedi by learning from this SEO black hat/white hat master who owns one of the most popular SEO forum on the internet.

So actually during my venture with Swagly, I actually just came up with the Lean Startup Methodology on my own too. Obviously I didn’t call it anything, but in short, I knew the right thing to do is the direct opposite of spending a shit load of money to hire a bunch of engineers to build a product that you don’t know if people want. So, at that point (2011/2012), I was actually quite well-learned. I spent years doing sales and management. I finally know how to code, and I do a decent design using HTML/CSS/Bootstrap. I know how to build a startup from ground up, and I could bring traffic via SEO or growth hacking.

During the years of working at the server company, I also mentored a few companies, where Pinkoi, etsy in Taiwan/Asia, is the most notable startup that I helped, and who was generating more than 100,000 uniques per month.

At this point I think I was ready to start again.

In 2012, I picked up the position as the CEO of National Division over at Groupon Taiwan, and I did that for about 8 months before the entrepreneur urge took over me again and I decided to leave and build a group dating website. I was pretty close with some of my direct reports, and I would share my startup ideas with them. So one day Harry, a sales who later became my close friend, told me that his brother (Yitao) is also thinking about building something too, so I quickly scheduled something with him to see if there would be a good match. So just to mention, at this point I already have a semi-working prototype of this group dating website that I built from scratch using PHP.

Anyhow, we sat down and we both had the same end goal in life, which is helping people succeed and revolutionizing education. Yitao wanted to help people in Taiwan learn English because he feels that that’s the barrier to learning for Taiwanese people, as the best content are all in English. For me, I have always been a shitty student because I never thought the kind of education I was getting was suitable for me. I believe that everyone learns differently, and all are all meant for different things. I don’t believe in the education system that we have in place now, so I want to help people discover what they want to do, and help them get there.

Anyways, we ended up partnering up.  I would be the CEO, where I would do all the front-end design, marketing, content-creation, SEO, growth hacking or what not, and Yitao, the CTO, would handle all the back-end programming and IT administration. That works out pretty well. We first built the group dating website (hellolets.com) for the Taiwan market. Anyhow, the problem was that while people LOVED the group dating idea, most users couldn’t get their friends to wing them.

After that, we decided to pivot into a dating website just like HowAboutWe, where the singles can post event ideas and other singles who are interested can contact the event creator to go on a date. We then encounter the notorious problem of the growth rate of horny guys > growth rate of female users, where the guys would just spam every girl on our website to have one-night stands. That was a problem because the rate of horny guys scaring away our female users away is much higher than the rate of our Google Adwords campaign or our blog is bringing in new female users. We were growing a lot of users, but to be honest we wanted to solve the problem of girls being spammed. We could do it by making users pay, but Taiwan users are usually quite frugal and would just resort to another website where they can continue their spamming.

So we’re like fuck this, and we pivoted into Hellol.com, which is a singles pinboard where you’d play matchmaker and pin your single friends onto the board or introduce friends from within 3 degrees away to each other. Singles can just come here and get to know someone too. The cool thing about this is that since everyone you see are someone that you are somewhat connect to, most people would probably not try to spam.

Anyways, fast forward. One day I was just kind of thinking about what else we need to do for Hellol in able to validate, grow, and just do all the right things. I had all my bookmarks open, and Evernote open. I was looking through all the notes and realized that it’d be really nice if we had this kind of guide where it ordered everything chronologically in a way where you’d can follow the guide and get the most solid advice and tricks to build and grow your startup. I told Yitao about this, and we were just both super excited about this because naturally, this is actually what we both want to do. Helping people succeed and empowering everyone to be the best version of themselves, which is exactly what being an entrepreneur is.

Summary is: So, at this point, I’d like to say that while it didn’t work out with my first startup, I’d say that I have learned to be a solid entrepreneur from all the things I did wrong in the past. I’ve taken everything I’ve learned in the past + learned everything I could from all the best startup experts and applied them to StartitUp – which I really hope can help those that are just starting out or trying to get back on their feet.

Successful startups are the result of a series of failed attempts. Angry Bird was Rovio’s 52nd game. It took Pinterest 2 years to get where they are now, and it took JK Rowling 7 years to write Harry Potter. If you’ve faced failures, do not fret. Failures are just the process of getting there. If you are determined, it’s about when, and not about if.

Anyways, sorry for the long post, but I hope you enjoyed it. Now, as someone who’s gone through many failures and have learned it the hard way, I’d like to make a few points.

About finding a technical co-founder:

First of all, I am in no way suggesting that you cannot be a lone founder who doesn’t know how to code. We’ve seen enough founders like that succeed to know that it is doable. The problem with my case was that I didn’t communicate correctly, and I had the wrong expectations. I probably should have found a full-time tech co-founder when I had the MVP build, and was already talking to all these huge companies for partnerships.

Second is that while I am in no way suggesting that you need to code to become successful, but you could try to learn some so that when you are talking to your co-founder, you can let him/her have a piece of mind when you know what he/she is talking about. You should also be able to give good technical advice when he/she is stuck. Coding is cool. Sitting in front of your laptop is lame, but creating something is cool. Coding has become so easy now, and the interwebs has everything you need to learn how to become a coder. Ruby on Rails or Bootstrap are so easy they should be illegal!

So, I’d say, if you want to learn, use the Michael Hartl’s Ruby on Rails tutorial. Don’t use it. Just read it, get the general idea. Ask a friend to help you. Sometimes having a friend explain to you what’s what will make your learning 10X faster. And if you want to program something, don’t follow the tutorial. Actually start coding on your project, and just read the parts that you need. That’s the proven best way to learn.

Therefore, the secret to catching a good tech co-founder is not to find or try to convince them, but to attract them. You need to have the skills that compliments them, or also be a coder that they can depend on as well. The important thing is not to think you’ll get lucky and land an awesome tech co-founder that will do all the work for you. Look to be good at front-end design, or marketing, or just something that can be a big contribution to the startup.

We’re here to help startups around the world succeed, and we’d are trying to be as helpful as possible. If you need anything, please let me know! at ed@startitup.co. Peace out.

Update (4/25/2013): Read this post on how to build a business as a non-technical co-founder.


Why Did We Build The Step-by-Step Startup Guide?


So one day, we’ve just begun brainstorming for Hellol.com, and while we were preparing, I was feeling pretty frustrated. It was frustrating because we were just kind of doing things or thinking of things as they go. We have to do everything all over again. We have to remember what we did right, and what we did wrong. Coding and all is all quite easy. Yitao does the back-end, and I do the front-end. However, what’s really hard is all the other stuff. How do we validate our MVP again? Where can we find our early adopters? How should we create our interview script? How about content marketing? When and where do we launch our MVP. The list goes on.

I’ve done it before, but where did I put all my notes again? I have a bunch of bookmarks of great articles from the internet and notes I’ve taken while reading great startup books by Eric Ries, Ash Maurya, or Steve Blank, but they’re just all scattered everywhere. I’ve read and saved great growth hacking trips from people like Mattan Griffel, Noah Kagan, or Andrew Chen. And while each piece of information is great, starting a startup is requires knowledge from all of those experts.

Information Overload
That’s what I feel like when I’m going through all my bookmarks and Read Later saves.

Also, we needed a way to keep track of what we’ve already done, and what else can we do to make our startup more awesome. We didn’t know if all the tricks and knowledge I’ve saved are still current. All those frustration piled up and that kind transformed into an idea. 

So…we have a few problems here:

1. We do know how to start a startup, and the team does have prior experience to building and growing startups, but most of our actions are more like reactions, and not proactively planning on action items before we realize we need something.

2. We do have these knowledge and tips saved somewhere, but we’d have to look at them separately. Also, the collection of them is not organized, and most of the knowledge are macro, not micro. They focus on telling, not showing. We needed action items that can give us immediate results (for things like Growth Hacking, getting early adopters, or getting feedback, etc)

3. We have no way to know which stage we’re at, or where we should be at. We don’t have a way to track what we’ve done, and what more we can do to improve.

4. Startup knowledge gets outdated quickly.

5. It wasn’t easy to just get a bunch of early adopters to test out your service or give you feedback on how you can improve your startup. Yes there is reddit, hacker news, quora, or what not, but most of them are not for this purpose.

So I said to Yitao “I wish we had this step-by-step guide that we could follow to help us do everything that’s right to building a startup. Lean startup shit, and it’ll have all the tricks to growth hacking and stuff.”

Enter StartitUp. So we decided to make something that we would love to use ourselves. Therefore, we ended up creating a startup guide, with constantly updated step-by-step action items collected from the best startup masters, where a startup can track their own progress, and get feedback from a community of other startups. We also want to make this guide free, so we can get the best information to entrepreneurs everywhere. We want this guide to help people succeed. Starting something you are passionate about is just about one of the best things anyone can do in their lifetimes…except for Ice Cream and perhaps sex (I love Ice Cream so much it’s a pronoun for me). The point is, we want people to be happy. Yitao and my dreams are to help people get the information they need, in the way they want it.

Ok, so other than that…I think it’d be interesting to discuss about why we think we’re qualified to create a guide that claims to be the ultimate startup guide. The truth is, we don’t necessarily think that we are. We’re just the guys who consolidates and organizes all the best startup knowledge from the people who ARE the shit, and organizes them. However, that’s not to say we’re not at all qualified. We’ve build and failed with startups before, and while we haven’t had a hit startup, we have learned a lot from the mistakes we’ve made, and we definitely have learned from that.

Jimmy Wales of Wikipedia doesn’t know every single history in the world, but he collected all the information and made a platform that made it possible to create the most complete encyclopedia in the world. Like Jimmy Wales, we’re merely consolidating and creating a platform to collect the best startup knowledge. And we’re creating the Ultimate Startup Guide.