…and mine too

Disclaimer: This isn’t a short read and I’ve been told my grammar isn’t great.

I was sitting at work when I got a text from a friend saying he saw my website featured on a big tech blog. I’ll never forget that feeling: something that, little computer nerd me, had built was being written about. I was always the one reading about other websites and apps launching on websites like Techcrunch, Mashable etc. dreaming about one day having one of my projects actually take off.

Twitpic was born out of another image-hosting website I made called Echopic. Back then, when Digg was the front page of the internet, image posts that made it to the front page of Digg would often have their servers taken down by the “Digg Effect”. I would host a mirror of the image and post a link to it in the comments so it could still be viewed. I think I made a few bucks off of Adsense, but I was more excited to have something I made being used by other people!

Then in early 2008, I made a Twitter account for the second time. Before that none of my friends were using Twitter so I never had a use for it, but now Twitter was starting to gain traction outside of the valley. Twitter was text-only then and people used Tinyurl to share stuff so they wouldn’t go over the 140 character limit. That’s when the idea for Twitpic hit me: A simple way to post a photo to Twitter with a short URL

The original Twitpic homepage in 2008

The name was created from half of the words Twitter + Picture: Twitpic (keep this in mind, it comes into play later). I spent the weekend ripping the code out of Echopic and hacking it together to work with Twitter’s API. Another big “aha!” moment was when I hooked the photo comments of Twitpic back into Twitter as @ replies to the original poster. This was huge in making Twitpic grow virally. (I was on my second Red Bull when that idea came to me 😜). By that Monday it was up and running with myself and a few other people (mostly friends I think) posting photos. Then Tuesday, February 5th 2008 (10 years ago as of this post) it was written about on Mashable and the snowball effect had started. I had no idea how fast, how big and how global Twitpic was going to grow and be used after that weekend.

I was a full-time computer programmer for a financial firm and worked on Twitpic during the nights and weekends. It ran on a small dedicated server I used for all my little projects and it worked well for the time being.

A few months later I got a call from the CEO of a large online shoe retailer who wanted to fly me out to Vegas to meet and talk about Twitpic. Of course like any sane person I jumped on the offer and flew out, all expenses paid. They showed me around, toured their offices and sold me on their culture. It was a great company and the culture matched. On the last day they offered me six figures for Twitpic and a nice signing bonus to come work for them. It was a great offer and I would have jumped on it except I thought: If someone was offering me this amount of money for Twitpic now, what could it turn into? It was growing quickly. I was 23 at the time.

Twitpic hummed along for the rest of 2008 growing steadily. It didn’t hit any major scaling issues until that December. I remember visiting my parents during Christmas and the Twitpic server had went down. My parents lived out in the country in Oklahoma and didn’t have internet (this is 2008 remember, dial up was the norm in rural areas). I drove to a nearby apartment complex with my laptop in the passenger seat, slowly driving around the parking lot until I found someone’s WiFi that wasn’t password protected. Yes, I stole someone’s WiFi on Christmas to fix Twitpic from my car…thank you to the person who didn’t put a password on their Wifi!

Old servers in my apartment

When Twitpic outgrew it’s little server, I purchased old Dell servers off of eBay for cheap. I set them up and tested them in my apartment. I would then load them up into my parents van and drive them to the datacenter that was about five hours away and rack em up!

Parents van loaded up with servers headed for the datacenter

Me racking servers in the datacenter 2009

This worked for a bit, but by 2009 I couldn’t keep up. I switched to leasing hardware from a provider who would build the servers to spec and host them.

Photo credit Janis Krums — twitter.com/jkrums

It was in January 15th, 2009 when I got my first glimpse at what Twitpic was turning into and the power of social media. I tried to load Twitpic in my browser that afternoon but it failed. I fired up my terminal to access the server itself to see what was going on, but that wouldn’t load either. It was when I went to Twitter that I saw the Twitpic of the plane crash in the Hudson River being retweeted over and over. I saw then how breaking news and information was flowing through social media and I saw the power of Twitpic.

Keep in mind that in 2008, sharing photos with your phone wasn’t mainstream. The iPhone was barely 6 months old, the App store didn’t launch until 6 months after Twitpic and social media was nothing like we know it now.

March 2009 was when I decided to leave my full-time job to pursue Twitpic. It was making just enough money to pay for the hosting costs, but not enough to pay myself a salary. With a couple thousand dollars in my bank account I took the leap. At the same time I decided to move from Oklahoma to Charleston, South Carolina after watching the movie The Patriot — I felt led to go there, so I did.

Also around this time I was invited to speak in San Francisco. Me, being the country boy I was, couldn’t believe it. San Francisco, the mecca of tech and I was asked to speak there. The icing on the cake was when I got an email from Biz Stone, one of the founders of Twitter. He somehow saw that I was going to be out there and invited me to do a presentation of Twitpic at the Twitter offices! You could have knocked me over with a feather, these people were my heroes! I had a blast, I was star struck.

While in SF, I got an email from a popular image-host asking me to lunch — If you looked up the word naive in the dictionary you would’ve seen my photo. We had lunch and they offered 7-figures for Twitpic. I declined the offer, but little did I know they had already built a clone. Not long after I declined they launched — I say clone because it copied parts of Twitpic in verbatim, including text that still had my spelling and grammar mistakes in it. I don’t fault them, they taught me a valuable lesson — the business world is cut throat. Another way competitors tried to undercut our lead was to pay 3rd party Twitter clients to be the “default” photo-sharing option in their apps and many times the user’s settings would be changed automatically without them knowing. I couldn’t in good conscious pay one app developer for integration and not another…that would be a long, back and forth (losing) battle with heavily funded SV competitors. Our strength was in the brand “Twitpic”.

Mom and Dad started working for Twitpic that year to help offload things from me — I think they felt bad for me and how overloaded I was. Dad took care of accounting and advertisers (stuff I hated) and Mom did customer support and content moderation. One time Mom banned the rapper 50 Cent because of a photo that violated our content policy…she had no idea who he was — He was unbanned shortly after. I can’t thank them enough for how much they helped me and how hard they worked on Twitpic — I was blessed with great parents. The first employee, outside of my parents, was hired in late 2009, 18 months after Twitpic launched. We maxed out at seven employees total during the company’s history. Very small for the “size” of web app we were running.

Between 2009 and 2011 was a whirlwind of growth, learning, struggles…and stress. “Twitpic” became a pop culture term used from rap songs to Cosmo articles and of course synonymous for photos on Twitter. Our users ranged from the President, celebrities, musicians, parents who were keeping in touch with their kids at college and everyone in-between. The money was pouring in from advertising, so much so that after all our expenses (salaries, hosting costs, company trips, etc) we still had over 60% remaining. At our peak Twitpic had 30+ million users and 80,000 more were joining every day.

To the world I was the happy, successful entrepreneur and that’s the image I conveyed at public speaking events. But inside I was miserable, stressed to the gills and imploding. Please don’t take that to mean I was ungrateful, I was very grateful, but I didn’t know how to handle the stress of trying to keep Twitpic running and be a good leader. I had millions of dollars in the bank, but I was ready to end it, the company, my life. I lost a close friend to a car accident in late 2009 which led me to self medicate with drinking. I found I could numb myself from stress and other unpleasant feelings through that. The drinking lead to partying, the partying lead to drugs and a very unhealthy life style. I used whatever high I could from alcohol to relationships to try and make myself feel better, but really I wasn’t. I was borrowing from tomorrow for a quick fix today. If you see someone struggling with depression or addiction, I encourage you not to be quick to judge. The mind and body are powerful at controlling someone. I thank God for getting me out of that lifestyle and restoring my Faith. I’m also thankful for a praying mother (who blames some of her gray hair on me 😉 )— love you Mom and Dad.

Ok, enough of the dark stuff…

Dick Costolo (really nice guy btw) speaking at Chirp 2010

Twitter’s attitude toward developers took a big shift in 2011. Before then they were welcoming to all developers who built on their platform and much of the functionality that is found in Twitter today was filled then by 3rd party apps. Twitter even had a developer conference in 2010 called Chirp which we attended. They stated their intent was to keep the ecosystem open and have transparent communication with developers, but as history shows that’s not the direction things went.

In mid 2011 Twitter launched their own built-in photo sharing and also began tightening down on access to their API. From limiting tokens for 3rd party Twitter clients to cap their growth or outright blocking apps, the “cleansing” had begun. I don’t fault Twitter and honestly I understand from their point-of-view why they went that route. They needed the user’s “eyes” on their official apps so they could monetize with their ads. Is it the route I wanted? No, of course not and for many other developers too. The preferred direction would have been to keep the API open, integrate Twitter’s ads into the outside apps and do a rev-share or even charge for API access. But it was Twitter’s platform and they had every right to do with it as they wish.

Twitpic made money from ads and if I can help it, I never want to run a company based solely on advertising again, it’s not fun. Advertisers do not have your user’s best interest in mind and I cringe at the types of ads we had to run sometimes to stay afloat. We ran Google Adsense as our base network — which got shutdown without warning at one point and our google rep “disappeared” (Hey Google, you owe me $70k still 😉). We then had other specialized networks for the countries we had high traffic in. Unless you control the ad ecosystem completely (i.e. native ads) I don’t recommend it.

2011 started the down-turn for Twitpic, mostly due to Twitter’s own competing photo sharing, but also for not innovating in the direction social media was going (i.e. Instagram). However Twitpic was pretty much married to Twitter from the beginning.

Twitpic was almost acquired in 2012. An unsolicited offer come in so we hired a firm to help us vet it and also seek out other interested parties since we were now an acquisition target. I had many meetings with many companies ranging from some you’ve never heard of to the every day household names. My one stipulation was that I wasn’t going to go with the company long-term. I would stay to help with the transition and some months after to get them setup for success, but that was my line in the sand (Twitpic didn’t need a huge team or resources to operate). After all the meetings and vetting, we landed on one company who was ready to lay down a nice offer BUT I had to commit to staying with the company long-term (golden handcuffs). I declined and once the phone conference had ended, the firm we hired told me to just agree to stay on long-term and “just quit” the company early. My conscious wouldn’t let me do that and would leave a bad taste in my mouth if I had. So acquisition failed, back to running Twitpic.

We tried for many years to trademark the term Twitpic, but because I was late in filing, other “Twit<insert-term-here>” had been approved and were blocking ours. Even our trademark lawyers thought it strange we couldn’t get our term to go through since our first use date pre-dated all the other applications. Regardless, years later we finally had overcome all the hurdles and re-submitted our trademark application to finally get ours approved.

Not longer after I received an email from Twitter’s legal saying that if we continued with trying to trademark Twitpic they would cut off our API access — I was livid to say the least. I thought: “What were they thinking?! Do they think they own the term Twitpic?!” — I wasn’t going to give into their demands! Yeah! fight the man! viva la revolucion! — but if they shut off our API access that would kill us.

After a day or so of many emotions and questions (I still don’t know their thought process on this honestly) it was decided to shutdown Twitpic. We couldn’t fight Twitter, a company with over a billion in funding versus Twitpic which was struggling to make ends meet. It was one of the most gut wrenching times of my life. My identity was so wrapped up into Twitpic that I didn’t know how to separate the two. If Twitpic failed, I failed as a person.

I didn’t think an acquisition was an option at the time because of our previously failed attempt, but I knew I couldn’t just let Twitpic and all the photos disappear. Surprisingly we had one come through that almost succeeded but failed at the last minute. Most of the other offers either weren’t legitimate or they couldn’t be trusted with the data. I went to sleep that night in turmoil after all the solutions had failed wondering how in the world did I get here. I woke up in the middle of the night, drenched in sweat, and it popped into my head to contact someone who I had met many years previously, but now worked at Twitter. I sent them an email that morning with my idea and that got the ball rolling.

In the end, we gave Twitpic to Twitter on October 25th, 2014. I wanted to be sure the photos lived on in safe hands and who better to do that than Twitter themselves. It took about a week to get all the paperwork done and hand Twitpic over to them, which is really fast for that kind of deal. I got so choked up on the last phone call with Twitter when I was officially “handing over the keys”. I told them to “take good care of it”, but I could barely speak — I was pretty embarrassed.

This was a seven year journey for Twitpic and me. I made many mistakes along the way and I learned more than I can put into words. I am so thankful to have had the opportunity to work on something that was used by so many worldwide, to have been able to work with such great people and to have learned so so much. I wouldn’t trade it for anything, the good or the bad.

Sarah and I on our wedding day…I call that move the “James Bond”

I’m now happily married to a beautiful woman who stayed by me even when I was going through the bad times. We have an english lab named Lou, who is basically our first child, (and now a baby boy, who is the joy of my life!) and she’s starting her own business — I’m also working on a new startup…

…I guess I didn’t learn my lesson 😆

Feel free to reach out to me here on Glue (@noah), Twitter (@noaheverett) or email me at noah@ark.fm

Notes: This post was originally posted on Medium February 5th, 2018. I'm migrating some of the posts from there to Glue. — I made two updates to the original post: 1) Referencing our new baby. 2) Updating my "new startup" to Glue (which this post is written on) it used to link to Pingly (which has an upcoming refresh).