8 Lessons from Durable's First Product Hunt Launch

Yesterday we launched the Durable AI Site Builder on Product Hunt. 

If you’ve launched on Product Hunt in the past or put any research into it, you’ll know that it's can absolutely game-changing for your product. And it's pretty simple and accessible! Build something, post it, and see what happens!

But for all that simplicity, it's also a bit of a blackbox. There are endless reams of content out there selling their methodology for hitting that top spot. Tips, tricks, snake oils, you name it. Everyone’s got a theory of how to make it work, and how to reach that golden #1 spot. 

We didn’t crack the code, but we did pretty well. After duking it out for 3rd place throughout the day, we ended up 4th. And we put our campaign together in under a week. So, all things considered — not bad! 

We also saw record sign ups and sites generated for one day. Sweet. So, clearly — *some* of the things we did worked. But within the first few hours, we had a list of “things we’d wish we’d done,” too. 

Find below our biggest takeaways from the launch.

‍

But first, the wins 🎉 

‍

#4 Position on Product Hunt 👍

414 upvotes and 154 comments. We’re pretty proud of that! But we *really* wanted that #1 spot, and placing *just* outside of the top 3 kinda hurts. 

‍

Featured in the October 20 Product Hunt newsletter 🌊

Nice. Any product placing in the top ten products from the prior day gets featured in the newsletter. Worth it! The traffic today has been awesome. 

‍

‍500 Signs Ups since 12:01am October 20th 🏒

To reiterate: 🏒🏒🏒🏒🏒🏒

5000+ Sites Generated 🤖

Yep, y’all put our robots to work yesterday. Perfect. This brings the lifetime total of AI sites built to around 15,000 

‍

10,000~ Site Visits 📈

A big day for us, and we're looking to hit the same traffic the day after. We almost 2x our traffic from our soft launch two weeks ago.

‍

Big Referral + Direct Traffic, with Social Helping Out 📣

Unsurprisingly, the majority of our traffic was from Product Hunt itself. That said, we saw good traffic from social (especially considering we focused solely on Twitter). Next time, social will put in more work!

But our launch wasn’t flawless. Product Hunt is still a total black box — we can’t be certain that any of the below learnings would have rocketed us to first place (or if they were solely responsible for our 4th place finish). Hopefully this is helpful. 

‍

What we learned:

1. Include a video and/or GIFs. We were the only product in the top five to not feature a video or GIF. We were torn on whether or not to post a video, and decided not to (part of that rationale was "there's some wow factor in using the product, let's not spoil it with a video"). If we could rewind to Wednesday, we'd include a video. 
2. Go big with your 12:01am voting network. We were a bit gun shy here. There are a ton of posts about how too many upvotes at midnight from new users can ding your score. I'm sure that's true! But looking at the #1 product from October 20 had a huge influx of votes at midnight. Our take: so long as you don’t have 100s or 1000s of new users upvoting at once, you're probably fine. 
3. Don't give up after 12:15am. Our midnight traffic was okay. We were hovering around the 5th to 7th position. But on Thursday morning, our West Coast wave hit! We rocketed to 2nd. The rest of a day was a battle for 2nd and 3rd place. Long story short, keep grinding in the morning and throughout the afternoon. The steady effort makes a huge difference. 
4. Start conversation threads. Engagement is crucial and a no-brainer (respond to every comment!). But where possible, start longer conversations/threads with individuals. We saw a ton of back and forth with our main competitor right at 12:02AM. We’ll do the same next time we launch.
5. Create some hype beforehand. We didn't do this (except for messages to people the day before launching). Next time we'll invest in a Product Hunt preview page, and promote on our social media.
6. Don't lose your momentum. Waves throughout the day are key. Again, this might be old news if this isn't your first rodeo, but just because you have the most votes doesn't mean you're #1. Take the whole damn day off from your normal work simply to reach out to people in your network. We were bumped from 3rd to 4th at about 8pm. A huge bummer, but a valuable learning — it’s not over until its over. 
7. Tweet! We focused on Twitter for supporting the launch (both from the @durableteam account, and from James' personal account). We were lucky enough to have some tech influencers and straight up legends retweet us. We hadn’t planned any of this beforehand which is a good and bad thing. Good because it genuinely felt “organic” and in the moment. Potentially bad because we were relying on the algorithmic gods to smile upon us. Big learning here: don’t sleep on Twitter, and put serious effort into it throughout the day. Build up your following as much as you can beforehand, tweet your launch, update your followers on progress, retweet comments from users/supports, tap big accounts on the Twitter-shoulder. 
8. Add some value for your supporters. We kinda felt like a jackasses asking so many people to support us. It feels tacky. So we wanted to give *something* back to people who took time out of their day to hit a triangle and make a comment (as my one friend put it after pestering him all day: “Here! Take your damn triangle!”). That’s why we’re creating a Product Hunt Learning Network (name pending!) for anyone who supported us (or anyone who's interested, honestly). Whether or not this made a difference, who knows — but it makes me feel like less of a tool for bugging hundreds of people! If you want in, send me an email (jonathan@durable.co).

‍

What we did to prepare for launch:

Some companies take months to prepare their Product Hunt launch. We gave it a week. Here's what we did:

1. Made 20+ updates to the Durable AI Site Builder based on feedback from V1.
2. Radically improved/simplified Durable's onboarding. This was big. And while not every company will have this flexibility, do everything you can to simplify your onboarding!
3. Secured a network of about 100~ supporters for launch. Also big, but we could have gone even bigger here. Tap your network, and find as many existing Product Hunt users as you can. New users are good too, but don't rely on them 100%.
4. Found a solid hunter to post for us. We went back and forth on this, but eventually decided to have a very solid hunter in my network post for us. We had the luxury of knowing someone beforehand. Do what you can to find someone!
5. Launched at 12:01am. It makes for a late night, but just do it! 
6. Produced dedicated social and Product Hunt images. Put some effort into designing something unique (and beautiful!) for your launch. Don't just rely on hastily thrown together screen shots with comic sans captions.
7. Posted on LinkedIn. We posted via our company account, and had most folks who work here share the post. We saw some good shares from larger LinkedIn influencers that produced good traffic.
8. Tweeted progress throughout the day. We provided updates on our position, progress, etc throughout the day. Worth it! Don't rely on one or two tweets to put in a full day's work for you.
9. Posted to Reddit in the evening. This doesn't seem to have generated much traffic, but we tried. Next time, we'll try this again, but earlier in the day. Note: add value here! Don't just post "hey look at us on Product Hunt!" You'll be eaten alive if you do, and ruin your reddit reputation (which matters! Reddit is slept on gold).

That about covers it for now! Product Hunt is a weird, fickle beast. I won't say we're pros at it yet, but I'd call this launch a success regardless. Would love to hear your thoughts, and your own findings over on Twitter, LinkedIn, or via email. 

‍