Tyler Swartz is a graduate of General Assembly and the creator of Bar Roulette. You can follow him @TylerSwartz on Twitter.
Bar Roulette is my first web app. Last year I decided to learn how to code beyond the free code academy courses and signed up for front-end web development at General Assembly. I loved it and as soon as I finished I took the back-end web development course to learn Ruby on Rails. My decision to build Bar Roulette was a result of Uber releasing a new ride request API and the need to pick a final project for my General Assembly class. Uber also had an online hackathon and I thought that deadline would help give me an extra push.
By the end of July, I had gotten Bar Roulette to a reasonable minimum viable product. I had the app fully functional and a nice landing page for it with some screenshots. There were a number of things I still wanted to improve, but I kept thinking of a quote from Reid Hoffman, “If you are not embarrassed by the first version of your product, you’ve launched too late.”
So I kicked off August by submitting it to a number of startup launch sites such as Startup List, Geekopedia, and Killer Startups. I also sent emails to Mashable and others. The only response I got was from Killer Startups, who said that Bar Roulette would be posted in about 60-90 days unless I wanted to pay the premium fee of $399 to have it posted within 48 hours. I declined.
Product Hunt and Hacker News were on my list, but I was intimidated because I hadn’t gotten a response from any of the other sites. I’d also read a number of posts about how to “hack the system” in your favor by posting to Product Hunt and Hacker News at the optimal time. However, I decided not listen to any of that and posted on a Wednesday night. I sat there looking at Product Hunt’s upcoming list with Bar Roulette and about 100 other products all with 1 upvote. The Product Hunt team selects the best products from this list to be featured later in the week. Not feeling very optimistic I decided to tweet at Product Hunt and Ryan Hoover (founder of Product Hunt).
A few minutes later I received a notification on Twitter that Ryan had favorited my tweet (nice!). I then refreshed my Product Hunt posting. The first thing I noticed was that the subheadline I had written for the post, “Web app combines Yelp and Uber to take you to a highly rated bar” had been changed to, “Go bar hopping with the help of Uber + Yelp.” Hmm…that’s odd, I thought. Then I saw that Ryan had upvoted Bar Roulette. I tweeted a thank you to Ryan and got confirmation that Bar Roulette would be featured tomorrow. I was ecstatic!
The next morning, I awoke early and checked Product Hunt and saw Bar Roulette was listed on the day’s featured list. It already had some upvotes and comments. I immediately dived in and began replying to people in the comments. I thanked them for the feedback and answered questions. A key learning for me was that it’s important to join the discussion and talk to everyone in the comments. The more active the discussion, the higher your product climbs on Product Hunt’s daily list. I was way above other products that had more votes because there was ALSO an active discussion happening on Bar Roulette. This discussion helped drive additional views and upvotes.
I also began actively engaging with people on Twitter. I searched for “Bar Roulette” and then continually refreshed the search. As people shared Bar Roulette’s post on Product Hunt I favorited and replied to them. Even when someone only shared the link with no additional comments, I would still favorite it and then reply to them with a simple thanks and comment. That ranged from “+1 🍻” to “Thanks, let me know how it goes when you try it!”. A big thanks to my cousin @graydayshop who reminded me to add “Creator of Bar Roulette” to my profile. This enabled me to interact with people who otherwise wouldn’t have replied to me because they wouldn’t have realized that I had built it. Some of the people that I thanked for sharing were genuinely pleased that I had favorited their tweet and followed up with congrats or feedback. I think my engagement was essential to drive traffic and buzz because I was genuine and it showed there was a real person behind Bar Roulette.
Throughout the day, Bar Roulette was continually shared on social media. You could say that it went somewhat viral. This buzz led to Business Insider, TechCrunch, NextWeb, Eater and many other sites writing about it. It’s clear that many of these sites use Product Hunt as a curator and source for new tech stories. Because I was active in the Product Hunt discussion from the beginning the writers were able to take my comments and quote me in their stories which gave them more substance.
One of my favorite exchanges about Bar Roulette on Twitter was between Yelp Columbus and Uber Ohio.
By Friday evening, I was seeing a large spike in traffic and I was beginning to get a new feeling…helplessness. I had scaled up additional dynos and that helped with the response times. But, as I sat there looking at my Heroku server logs via Papertrail I was seeing them return errors to users. I was doing my best to interpret everything they were saying, but I was having trouble understanding if the errors were minor or serious. Finally, I became concerned that the Uber ride functionality had broken and that it was returning errors midway through the trip.
My wife and I decided the best way to alleviate my stress and concern was to simply use Bar Roulette to make sure it was still functioning under the traffic. If it broke when I tested it, then I’d get back home and get to work. My wife used her phone, confirmed our location, the Uber showed up and drove us to Hayes Valley. We got out of the Uber, holding our collective breath. The app refreshed and showed us the final screen with the bar name. We embraced in relief and then went into the bar. Bar Roulette was still up and running under the server strain.
I don’t think a beer has ever tasted so good.
I’ve now had some space from the initial success of Bar Roulette to put my thoughts down. Here are the key points I took away from posting on Product Hunt and increasing awareness of my side project.
1. Be genuine.
Use your personal brand, people respect you for putting yourself out there and being vulnerable. Be genuine when you ask for feedback on your product by actually listening and acknowledging users.
2. Engage meaningfully with people.
Don’t just favorite a tweet about your product—reply to every tweet by asking for feedback and/or thanking them for sharing your product.
3. Include your product name in your Twitter bio.
While you know that you built something, most people don’t know who made the product they are tweeting about. When you interact with people on Twitter they will often check your bio. You don’t want to be that random person favoriting their tweets for what appears to be no reason.
4. Make sure to keep some perspective.
On Thursday morning when I left for work I told my wife that I’d be happy with 25 upvotes on Product Hunt and 500 visits to Bar Roulette. I ended up with 240+ upvotes and 10,000 visits. By keeping some perspective, you can celebrate the successes as they come. There will always be more you can do, another goal to be reached, but make sure that you take moments to celebrate what you have achieved.
5. Use emojis.
Well it depends on your product 😉. Bar Roulette is about having a fun night out and emojis can be a good way to keep tweets light and upbeat.
So, what’s next for Bar Roulette? I’m currently working on a native iOS version. If you have used Bar Roulette, I’d love your feedback and any feature suggestions for the app. Follow me on Twitter for more updates and product musings.
Turn Your App Idea into a Viral Success
Originally posted on Medium by Tyler Swartz on October 13, 2015.