Product Hunt’s Rise
An Overnight Success 1,834 Days in the Making
The overnight success is a myth that has been thoroughly debunked over recent years, especially as it applies to tech. Knowing this, it’s still hard not to think of Product Hunt as an overnight success given how rapidly it spread throughout Silicon Valley. Practically all major product launches are now posted to Product Hunt. In fact, the product’s makers are almost always present as well, commenting in realtime with everyone else. From the outside, Product Hunt’s rise felt like magic. As if this awesome community simply sprung into being. Naturally this led to a slew of Product-Hunt-for-X copycats as people who were inspired by the site attempted to build communities of their own…
Unfortunately as these would-be founders quickly discovered, there’s a lot more to starting a community site than the idea and technical execution. Regrettably, that’s why I don’t see much of a future for Telescope, which is a site that allows you to “create a Product Hunt / Hacker News clone in 1-click”. Looking through the showcase and seeing nothing but expired domains of dead projects and ghost towns with zero comments, it’s abundantly clear that the tech behind community sites isn’t what moves the needle.
Even OpenHunt, the response to Product Hunt’s insider controversy which hit the front page of Hacker News twice with heavy mentions in a 3rd front page thread is now dead. In the post-mortem thread the founder shared a screenshot of OpenHunt’s Google Analytics account:
Close to 30,000 unique visitors and yet the site still flopped. So clearly even a (humungous) one-time traffic spike isn’t enough to keep a community site afloat.
As a software consultant, I get pitched community-based business ideas left and right. Each time so far I’ve declined though, because after drilling down into the details with the founder it’s clear that they are out of their depth. After the third or fourth time this happened it got me thinking — what does it take to successfully launch a community site? Being a fan of internet archaeology, I decided to do some metaphorical digging into Product Hunt and Ryan Hoover, Product Hunt’s founder.
Internet archaeology step 1: find the oldest thing online about the subject that you can (oftentimes hilarious MySpace pages). In Ryan’s case, I settled for the very first post on his blog: “Techniques for Getting Traction with your Social Site” from March 3rd, 2011 (AKA 1,834 days ago and this post’s namesake). While hindsight bias is both real and problematic, it’s no coincidence that this was Ryan’s first post. It clearly demonstrates his obsession (or at least disciplined study) of the mechanics behind social/community sites and how to make them successful over the course of years. Specifically, Ryan’s post was written a full 979 days before Product Hunt’s pre-beta launch on November 6th, 2013. In the post Ryan describes tactics including:
1.) Target Influencers
2.) Indulge Early Adopters & Listen
3.) Make it Useful w/o Users
4.) Create Exclusivity, Scarcity, Urgency
5.) Give Users Tools to Evangelize
6.) Seed Content & Communities
While these points are nice, it’s unlikely that they made much of an impact on anyone back in 2011 when they were published. The reason is simply that Ryan had written them before he had ever actually built a successful community. Now that he has, we’re going to walk through each tactic step-by-step and show exactly how Ryan used these techniques in 2013 to successfully launch & market Product Hunt.
And now, without further adieu, here they are:
1.) Target Influencers
Ryan in 2011
- Focus on influencers within specific communities to drive WOM and evangelism of the product
- Almost every product starts with a core audience (Crossing the Chasm)
Ryan in 2013: Ryan says: “I seeded the community by inviting a few dozen founders, investors, and startup folks I knew. To my surprise, people really enjoyed the daily email and the subscriber base grew organically.”
Because Product Hunt has sequential User IDs starting from #1, with a bit of legwork we can actually see who the very first Product Hunt users were. Here are the bios of the first 20:
#3 Entrepreneur, now at Sequoia Capital
#4 Startup Edition account (explained below)
#5 Polymath Engineer / Designer / Speaker / Digital Nomad
#6 Y Combinator Founder + Angel Investor
#7 Founder of a startup acquired by Yahoo!
#8 Founder of CoffeeMe
#9 Author of Hooked, +33K Twitter followers
#10 Founder of Quibb
#11 Serial Entrepreneur
#12 Founder w/ successful exit
#13 Growth Marketer & Consultant
#14 Head of Product at Lost My Name
#15 CEO of MonkeyInferno + Blab.im
#16 Designer at Envoy
#17 Founder of Speak
#18 Product Engineer
#19 Venture Capital Dealmaker
#20 Serial Entrepreneur + Angel Investor
(The list goes on…)
Ok great, but how do you actually get a few dozen influential founders/investors/startup folks to use your site? Sure most people have hundreds of Facebook friends and Twitter followers but we’re talking about dozens of influencers, not just regular people. Getting a network like this takes focused work and it won’t “just happen” organically. While I would love to hear directly from Ryan his thoughts on how he built these relationships, as an outside observer I noticed this:
Startup Edition (archived copy), which is now offline, was an email list with “Weekly wisdom from founders, hackers & designers” organized by Ryan. Each week would pose a new question that all the influencer participants would then write a blog post answering. Examples include “How did you choose your pricing model?” and “What advice would you give to young entrepreneurs?”. With Startup Edition, Ryan had an “excuse” to contact influential people he didn’t know and build a relationship with them. The key is that this “excuse” needs to benefit the influencer as well to get them interested. In the case of Startup Edition, the benefit to the influencer is more publicity for their post plus getting a unique chance to compare & contrast their ideas with other influencers that they might not know either. In essence, this made Ryan into a connector and helped him rapidly build his network.
Another person who has written about this idea is Tim Ferriss. He often repeats that volunteering for a Silicon Valley charity got him an “excuse” to engage with influencers, in his case to organize events involving the influencer. Whatever the specific method, the key is getting your foot in the door so that future interactions won’t be a cold-open.
Other possibilities include running a podcast (no coincidence there is now a Product Hunt Podcast), doing traditional written interviews, or freelance writing for an existing publication (added bonus: getting an automatic boost from that publication’s credibility). Know any other tactics? Mention them in the comments and I’ll add them here.
Then once you’ve built out a group of +20 influencers in your domain (the Product Hunt private-beta had 30), the next step is to start building the audience for your site. Which brings us to step 2…
2.) Indulge Early Adopters and Listen
Ryan in 2011
- Promote early adopters and make them feel special
- Engage with early adopters and listen to their feedback wherever they leave it (ex: Twitter, forums, blogs, Get Satisfaction, etc.)
Ryan in 2013: In his excellent write-up of Product Hunt’s origin, Nathan Bashaw writes: “Ryan personally emailed and tweeted at hundreds (possibly thousands) of people to make this work, and it would have been a lot harder if he didn’t put in a lot of work over the previous year building his online network through his blog.”
The importance of this point cannot be overstated. As many have riffed on before, “build it and they will come” is BS in today’s information overload era. With millions of apps and websites competing for attention, standing out from the crowd is major feat. Ryan’s blog, tagline “startups, product & personal growth” was started a full 979 days before launching Product Hunt. Though he didn’t know it at the time, Ryan was building an audience of startup enthusiasts critical to Product Hunt’s early traction. But Ryan went well beyond the typical “blog & pray” strategy that is often seen…
Specifically, in my research I was shocked to learn how Ryan leveraged a simple (but not easy) tactic to more than triple his Twitter follower count (1219 to 3868) over the course of the 8 months leading up to Product Hunt’s public launch.
It all started innocently enough for Ryan, in a Twitter conversation on March 1st, 2013 with Nadia Eghbal:
The idea is simple. As a small-time blogger occasionally people will share your articles on Twitter. The default response most bloggers will take is to like/favorite that tweet. Ryan decided that instead of doing that default action he would actually reply to their tweet with a tweet of his own like so:
While most of these tweets would simply disappear into the ether (and perhaps leaving some warm fuzzies in the recipient) occasionally the sharer would respond to Ryan’s thank you. At which point Ryan would “strike”, inviting the user to join his email list. As he describes in the below article, the user is perfectly primed to join your email list since they literally just shared your content:
While this is a very calculated tactic, it comes from a good place. As Ryan explains in the below article, the idea is to grow the small kernel of engagement in the retweet into something bigger — possibly going so far as to create a real relationship with the reader. He wrote about this concept further here:
Now let’s zoom out and see the macro effects of this tactic. Using Twitter’s advanced search, I looked month-to-month to see exactly how many of these tweets Ryan was sending out (link to the search here):
In total Ryan sent out over 1,300 individual “thanks for sharing” tweets to people on twitter, 1,073 of which occurred from March 2013 through December 2013, when Product Hunt launched. For fun, I compiled a 13.3MB 594 x 154590 pixel image of all 1,327 tweets. I use it to inspire myself to power through the dull-but-necessary grunt work required to achieve anything significant. If you’d like to do the same you can download the image on my website. Just make sure to open it in Preview (or your OS equivalent) since it’s humungous size is liable to crash Chrome/Firefox/etc.
Next, using TwitterCounter.com I purchased the historical Tweet/Follower/Following dataset for Ryan’s account. Here it is graphed over the same time period as the “Thanks for Sharing” graph:
From the source data (downloadable here), we see that from March 1st, when Ryan started this technique, to December 4th, when Product Hunt publicly launched, Ryan more than tripled his follower count from 1219 to 3389. During this same period he tweeted 12,382 times where 1,073* (or 8.6%) of which were the “Thanks for Sharing” tweets (*if we included follow-up tweets this would be even higher). Seeing as prior to March 2013 Ryan had tweeted 10,801 times and only acquired 1,219 followers it is clear that this period of tweeting was dramatically more effective. Though we don’t have the data, it’s likely that Ryan’s email list at least tripled in size over this period as well.
The end result of all of this is that on day 1 when Ryan tweeted announcing Product hunt he got over 150* people to click through and check out the Product Hunt email list beta (*across https://bit.ly/1hMVzkQ+ and https://bit.ly/1iMUK9q+). Since this coincided with a post on Ryan’s blog, he definitely emailed his list as well. At a 58% open rate and 23% CTR on average for Ryan’s list, we can speculate he easily got a few hundred other visitors to check out Product Hunt on day 1. Ultimately this led to over 170 people subscribing within the first two weeks and the clear momentum to build out the full-featured version of the site.
The power to cheaply and easily get your message in front of hundreds of targeted people cannot be understated. Buying ads on a large platform like Google AdWords or Facebook to achieve the same result could easily cost a few thousand dollars each time. While this isn’t a big deal if you’re sure about your current idea, in reality your first idea is almost certainly wrong (Ryan talks about a few of his earlier unsuccessful ideas here).
Now that we’ve shown the importance of engaging with your early adopters to build an audience before you launch, the next step is to make sure your earliest users get solid value from the site before it becomes popular. Which brings us to step 3…
3.) Make it Useful w/o Users
Ryan in 2011
- Early adopters’ friends won’t be on the service so it must be useful w/o them
- The initial user experience may not be focused on friends (Hashable as an address book, Foursquare as a game, Instagram as a way to publish photos across various social networks)
Ryan in 2013: In Ryan’s blog post called “Product Hunt Began as an Email List” he writes, “I was reminded of Linkydink, a link-sharing tool by the friendly folks at Makeshift. Simply create a group and invite people to share links with other contributors and subscribers. Each day, the collection of posts are emailed to the group.”
By starting Product Hunt off as an email list rather than a full-blown comment-enabled community site, Ryan was able to make the site useful without users. Even if he couldn’t find outside contributors to submit links (which he did), it would have been simple to find all of the links on his own each day. Definitely much easier than creating fake accounts and fake comments like the founders of Reddit did in their early days.
What I’d like to emphasize was that making Product Hunt email-only to start wasn’t a divine stroke of genius that hit Ryan as he conceived of Product Hunt. Instead, it was merely another tool in his toolbox of techniques that he was able to pull from. In his post “Email-First Startups” from March, 2013 (6 months pre-Product Hunt) he lists other email-first companies including AngelList, iDoneThis, Timehop and Thrillist as examples. In the front-page Hacker News comment thread, Craigslist, the grand-daddy of them all, was also mentioned as possibly the biggest email-first startup. And as a sidenote, not all of Ryan’s Hacker News posts hit the front page:
Tying back to technique #2 (Indulge Early Adopters & Listen), this points to the volume of writing output that building an audience actually takes. And while there are certainly some super-humans who can find the time to write regularly & voluminously despite being full-time employees (one of whom is Ryan), anecdotally I’ve noticed that more often it’s consultants and bootstrapping entrepreneurs who tend to publish a lot of high quality content online.
As stated earlier, with the audience that Ryan had built over the previous year it took him two weeks to reach 30 contributors and 170 subscribers. This demonstrated that Ryan wasn’t a ‘wantrepreneur’ and provided the key momentum necessary to get Nathan, his designer/developer partner, on board to actually build out the full-blown community site.
From here, the next step was to figure out how to bridge the gaping chasm between an active email list and a thriving community site. As we saw, avoiding the zero comment ghost town is no easy feat. Which brings us to step 4…
4.) Create Exclusivity, Scarcity, Urgency
Ryan in 2011
- Private and invite-only betas increase user’s desire for access and can help generate some hype
- Emulate a land grab by providing users the ability to claim “land” on a first-come-first-serve basis (ex: About.me’s vanity URL registration before launch, Convore’s chat room creation/moderation)
Ryan in 2013: Nathan, Ryan’s early collaborator, wrote: “It helped a lot that we had an invite-only model, because people love hooking their friends up with exclusive access to new things. By creating artificial scarcity, we had something at least semi-valuable to offer: an invite.”
Invites are a tried and true tactic — everything from GMail, Pinterest, Spotify, to even Medium itself began as invite-only sites. While it sounds like a “well duh” tactic, in practice it is painfully hard to actually execute.
It’s when your community has just a sprinkling of comments and is walking the knife’s edge between success and failure that your mettle will be tested. Though it can be painfully tempting to abandon the invite system and open the traffic floodgates, that would be a mistake. In Product Hunt’s case we see the number of daily comments (in the 7 highest upvoted stories) range from a high of 42 to a low of 9 during the latter half of the first month that Product Hunt existed as a full-featured community site.
Despite the roller-coaster engagement graph, Ryan and Nathan stayed the course and maintained their invite system. Ultimately this keep comment quality top notch and the signal-to-noise ratio high.
With that said, Ryan didn’t just leave things to fate (recurring theme) by simply hoping his early users did a good job inviting their friends. Instead, he proactively solicited invitations from his early users like so:
This is a perfect example of Paul Graham’s “Do things that Don’t Scale” concept which he popularized in July 2013. While Ryan wouldn’t be able to keep this up forever, in Product Hunt’s early days it was essential to tilt the scales in his favor.
Overall with this turbo-charged invite tactic, the site grew from 400 users at the end of day 1’s press launch, to a group of 2,000 handpicked superusers 20 days later. And that is the type of community that could really thrive & grow. From here, continued growth would require bigger forces than even Ryan could muster on his own. Which brings us to step 5…
5.) Give Users Tools to Evangelize
Ryan in 2011
- Embeddable widgets that allow users to distribute content across the web, particularly powerful when its their content (read: expression)
- Make sharing easy, fun, and intrinsic
Ryan in 2013: By tying Product Hunt logins to Twitter handles, Ryan firmly associated Twitter as the primary sharing medium of the site. Today you see this yourself by simply searching “we’re on product hunt” on Twitter.
Like all community sites, the fact that more upvotes leads to more views for the product heavily incentivizes the creators to share with their audience (though the rules do forbid explicitly asking for upvotes). This allows Product Hunt’s marketing to scale with its userbase, which is how the site was able to skyrocket into the top 3,500 most popular websites in the USA (per Alexa ranking).
And just in case the founder wasn’t aware their product was on Product Hunt, Ryan set up the @producthunt Twitter account to automatically tweet out each submission like so:
In his own words: “This extends Product Hunt beyond email and the site itself to another communication channel. Submitters receive feedback for their contribution, @producthunt followers are reminded of the service, and often founders of the products shared discover the service after monitoring mentions of their product on Twitter.”
In 2015, close to +40 Products are debuted on Product Hunt each and every day. This leads to a diverse set of easily over 100 people eager and ready to share links to their product on Product Hunt. Further, many of these makers are outside of Silicon Valley or even the USA entirely, thus facilitating the spread of Product Hunt across the entire world.
Finally, for the top performing products on Product Hunt each day Ryan and his team personally email the founders encouraging them to write follow-on blog posts about the results of their success:
This leads to an “aftershock effect”, leading to further publicity for both Product Hunt and the product creator after the initial launch on the main site. It’s a win-win for everyone and is a key scaling technique for Product Hunt to get exposure across a broad set of audiences.
Now that Product Hunt is being spread to a diverse set of new users each day, the final piece is to make sure those new users (who may be the first in their network on Product Hunt) have a great experience as soon as they register. Which brings us to step 6…
6.) Seed Content and Communities
Ryan in 2011
- Seed user profiles with personal content to add immediate personalization and usefulness (ex: fflick aggregates Twitter user’s movie mentions, Squarespace provide blog importing tools)
- Create communities around a specific entity (person, brand, etc.) to encourage those entities to claim ownership and engage (ex: Get Satisfaction, LinkedIn company profiles)
Ryan in 2013: Nobody likes an empty activity feed. In product design parlance, this issue is referred to as “Solving the Empty Room Problem”. Again, Ryan elegantly bypassed this quandary tying ProductHunt logins to Twitter accounts. This enables Product Hunt to auto-follow all the people on Product Hunt who you are already following on Twitter.
This kicks off your activity feed so that without any further work on your part you will get informed about the recent activity of people you are interested in. Further, it dramatically increases the number of times Product Hunt can email you about relevant updates:
This is huge because as Ryan puts it:
Habits don’t form overnight. It takes several days, often weeks for a product or service to earn unprompted user engagement, triggered by people’s day-to-day emotions. Consider your use of Facebook, Twitter, Pinterest, or other popular, habit-forming products. Engagement starts with external triggers that inform the user what to do, driving the desired behavior.
In fact, Ryan has written extensively about habit forming products, with at least five separate blog posts on the subject and even collaborating on the book “Hooked: How to Build Habit-Forming Products” with Nir Eyal (PH user #9). Once using a product becomes a habit, that user is going to stick around.
And ultimately that is the key for Product Hunt’s continued long-term future growth.
Ryan puts it best here: “The term ‘startup’ is deceiving. Successful companies don’t start up overnight; they are founded upon years of experience and help from others that must be earned.”
In Ryan’s specific case, we’ve learned that he:
- Grew his network of community-specific influencers using Startup Edition, his “excuse” to form a relationship with many varied influencers.
- Tripled his audience over the course of the 9 months prior launching Product Hunt by authentically engaging with his readers on Twitter.
- Earned critical early engagement in Product Hunt with an email-only beta he built himself. Enough to get a programmer excited to join him.
- Cultivated the early seeds of his community through high-touch interactions with community members and an invite-only system.
- Facilitated continued growth through deep Twitter integration and high-touch interactions with community makers via follow-up blog posts
- Streamlined further scaling through Twitter auto-follows and a habit-forming product design based on his expertise in habit formation
Most startup origin stories start with step #3 (when active work on the project began). Hopefully by showcasing the things that Ryan was doing over 2.5 years before founding Product Hunt, I’ve shown a new side to his story.
Please email me at firstname.lastname@example.org or tweet @theunixbeard with your thoughts and questions about this story. As inspired by Ryan, I reply to every single response 😎
If you made it this far, I’m guessing you like Internet Archaeology. Join the club now to get involved with our next dig, spots are filling fast.