How to launch (and not launch) a product on Product Hunt
Last week, after many months of development, I was happy to launch Haystack into the world finally. We decided to do a Product Hunt launch, and I wanted to share everything we did well, poorly, what I’d do differently and also the results of our launch.
Quick Background on Haystack
Haystack helps people with their “where is that document” problem, by searching across all your cloud providers (Multiple Drive, Dropbox, Slack, Code, Notion…etc). It’s a browser extension (currently Chrome but other browsers on their way), that has a simple keyboard shortcut to bring up, search and then use your keyboard or mouse to quickly go through the results to find the right document/file. It saves the average office employee 20 minutes a day.
Quick Background on Product Hunt
Product Hunt was created in 2013, and was made to be a place for people to share cool new products with each other (hence the ‘hunting around’). Since it’s creation, it has become stable for (mainly software) products to launch to a tech-centric audience. It’s full of early adopters, other creators and people that love giving feedback.
Product hunt is available on the website, a Chrome Extension that takes over peoples New Tab, apps and also a daily newsletter that shows the top 5 products from the day before.
A Quick note on the Product Hunt Algorithm
Product Hunt (PH) doesn’t just work like Reddit, where it’s 1 person to an upvote and the highest wins. It takes into account comments, reviews and how long/active people are that upvoted. That is why you will often see higher numbers lower down in the results on PH. It’s not an exact science (nor publicly revealed), but in short, they are trying to reduce people gaming the system, by creating lots of fake accounts or getting their friends to upvote them (when they don’t have accounts). It can definitely still be gamed (which we’ll talk about), but that’s any system.
What we did before launch
In short… very little for the launch. I was spending 95% of my time on the Product, I had set a goal of launching before the end of the month, and I was working my arse off to ensure that it worked great for people that accessed it. It’s often a tough decision as to where to focus your time. In short, the advice is if you have a great product then people will share it with others, but if no one first uses it then who will share it (“build it and they will come” is a lie).
What did I end up doing then?
- 2 weeks before launch, asked an advisor if he would be the one to ‘hunt’ us (so that his network would see it and increase views). Thanks, Eric Friedman!
- 2/1 week before launch I emailed and shared on social media asking if people would do a usability study/spare 15 minutes to try something I’d been working on.
- 2/1 day before launch, I shared the tool with people across my friend network, investor network (Expa), Slack groups (xoogler startups, Startup Leadership), Email groups (Founders Roundtable) — telling them we were launching and I’d love their feedback.
- I also asked this group, if they liked it to review us in the Chrome Webstore.
What I would have done differently
- 4 weeks before launch I would have searched for people that wrote about Chrome Extensions, Productivity and our competitors (Blogs, News, Vlogs), complimented them on their last articles and offered non-Haystack suggestions for great tools I loved.
- 3 weeks before launch I would have shared it with 50–100 people asking them to try it out, give feedback. Write on Social Media that I had a new tool that solved X, and would anyone be willing to try it.
- 2 weeks before launch — Those that gave positive feedback, I would have asked if they would provide us with a review in the Chrome Webstore (turns out that reviews can take up to a week to appear, so on launch day Haystack said 0 stars and 0 users, even though we had 10 reviews and 50 users).
- 2 weeks before launch — Emailed the people from week 4, saying that I was actually launching a tool they might like in a couple of weeks, and I’d love to share it with them to see if they’d like it.
- 2 weeks before launch — Shared early access links with Haystack Across my Social Media, Slack and Email groups telling them we are launching soon.
- 1 week before launch — Check-in with people that have been using it, check-in with the press. Tell friends I’m launching in a week, and I’d love it if they could share a post on Social Media when we did.
- 1 week before launch — Directly email 500 people from my network sharing it with them.
- The day before — Thank everyone for all their feedback and advice and again say we’re launching tomorrow at X time.
The content of the launch
Sharing on Product Hunt requires you to create content in a particular format with set word counts (Name, Description, OneLiner). I’d written this a few days before and gotten the hunter to read through and check it over. There is also the comments that you’ll write afterwards, so he had written his proposed comment, and I’d written my reply. We checked each others words over for this and gave feedback.
I’d also created particular images that illustrated what Haystack was through a story (the first image being an explanation), I also created an animated logo to try to attract peoples attention (It’s annoying, but I think it really helps).
What would I have done differently?
- I think Videos tell a better story than flicking through images. I did create one auto play video for the first page that people went to after Product Hunt, so I could potentially have created a version here too.
- Only chose one category: We chose 6 categories, and for some reason Product Hunt chose ‘Safari Extension’ as the one that showed up (which we had yet to support fully) which I think would have stopped some potential users. You can add more categories later.
- Been clear with the order of the pictures, alas I didn’t give clear direction on these, so the ordering was off so didn’t tell the story how I wanted to.
The launch itself
So as with all things, there is going to be a little bit of an anti-climax when you actually launch but the day itself was very full!
What did I end up doing?
- Eric Friedman and I got up and jumped on a call at 8AM EST (5AM PST) to submit it. We later realised that Product Hunt now lets you schedule launches so we could have chosen any time.
- We both then added our comments and shared on Twitter details about the launch asking people to check it out.
- I also shared on Facebook, LinkedIn a similar message.
- Emailed on the previous email chains that we had now launched and I’d love them to check us out on Product Hunt and reshare on Twitter the tweet I’d written.
What would I have done differently?
- Launched much earlier in the morning! Product Hunt works on a PST timezone which is fine if only the US looks at it, but it’s a global site these days, which meant that Europe had been checking out the days’ activities and voting for hours! The entire top 5 were submitted way before us! Most of the blogs have given the advice for 5AM PST, but I don’t think that is right at all — There is a ‘latest’ tab in Product Hunt, but it’s so small that I don’t think anyone clicks it except the people submitting that day to check out the competition!
- Changed the share message from ‘I’ to ‘We’. The message I shared on Twitter, was “I just launched”. Now, in all honesty, right now it is just me working at Haystack, but regardless I feel it comes off poorly. Either people thought I was being unmodest and not reflecting my team, or it seemed like a small side project that was going to close a week later. Not great either way (and a few folks messaged me about it too!).
- Only asked people to do one thing (share on Twitter!). People don’t like decisions, especially if it requires them to do things. By asking people to share on Twitter (and providing a direct link to retweet like this), it was easy to do, I was taping their networks, and if they liked the message & new Product Hunt, they would have gone in and upvoted themselves. I think this is a crucial hack for PH — You can’t ask “If you have Product Hunt, please upvote this”, but asking people to retweet means that generally only Product Hunters will click on the upvote and you get much more visibility.
Throughout the day
Apparently, you get a bit of an initial hit, but then you start to see numbers go up very slowly. You also see the other products start to go up to, which is scary because the higher up you are, the more chance people will give you a shot (and getting into the top 5 has enormous benefits). We started off in 15th and after a few hours had gotten to 10th due to the steps above, but there was much more to be done.
What did I end up doing?
- Throughout the day I refreshed the Product Hunt page and replied to every comment and review.
- 1st: I messaged Friends & Colleagues directly over an email asking them to check us out
- 2nd: I went to my sent email for my work and just started emailing everyone I had ever messaged asking them to tweet the message.
- 3rd: I wrote a blog post about why we created it (which took 1–2 hours up!)
- 4th: As time got close to the end, I started messaging people on LinkedIn, first in London then San Francisco then Australia (around the world time-zoning). I asked them how they were (I hadn’t spoken to these people for a while and I was genuinely interested), told them I had just launched a product and would love a retweet if they wouldn’t mind.
- Finally, I went to bed about 11.30PM (still technically 3.5 hours to go on Product Hunt time), but after about 2 hours sleep beforehand I was too dazed to be effective.
What would I have done differently?
- The day before, ideally I would have had a list of 500+ people that I wanted to get back in touch with, had written 4 blog posts (Why we created it, How we built it, The benefits of a tool like ours, where we are going next with the tool).
- Have emailed all the press/influencers letting them know we had launched
- Had done the LinkedIn ‘hack’ every hour depending on the country/timezone. I have 3,500 LinkedIn connections, and I really didn’t use that enough — plus I liked reconnecting with people)
- Gotten more sleep! I was so dazed from 2pm onwards that I could barely think.
- Written out additional ‘channels’ that I could use (I did HackerNews to mainly no result).
So what was the result?
We ended the day in 8th place, which isn’t bad at all! Especially as I don’t have a large Twitter or Product Hunt following.
So over the 2 days (Feb 27–28) from Product hunt, we ended up with:
891 Hits > 629 Chrome Hits > 200 Cloud Integrations (Approximately 150 unique users)
The 2 days later (drop off expected):
201 Hits > 170 Chrome Hits > 73 Cloud Integrations (Approximately 45 users)
We also gained about 100 emails from people requesting access on the homepage (Which was surprising given that you could click on the direct link in Product Hunt to get instant access!).
Other things I would have done differently
- Made sure we had a way to identify users uniquely
- Made sure to add much more tracking in the tool (anonymous tracking, just to show usage, clicks, errors…etc).
- Added tracking in the onboarding to work out where people dropped off and what screens people didn’t like
And most of all, just planned & slept more. Eric Friedman very wisely said I should enjoy it… And there were smiles through the day as the reviews came in, but I was definitely exhausted and worried about what I should be doing with the rest of my time!
Wrap up
Overall, launching on Product Hunt was really great for us, I have many people I’ve never spoken to before reaching out with thanks and feedback for the tool!
I want to thank everyone for all their help, reviews, feedback and love through the launch (including the before and after). In particular Paige Birnbaum (Thanks for putting up with me!), Eric Friedman (thanks for the advice and hunting us), Andrea Funsten (Thanks for forcing me to set a date and keeping me sane), Tony Peccatiello (Thanks for the positive enforcement), LukeGilson (Thanks for the beers) and Ayla Lewis (Thanks for all your work and help) There are many others, sorry if I didn’t mention you (I’m still tired.. OK!).
I’m still over a week later trying to catch up on the sleep I lost for the week before launch, but I’m so happy to have it in the world and people loving it!
I’m happy to offer advice to folks about launching on Twitter and in the comments below too!
PS
If you’d like to try Haystack yourself (it’s free after all), you can do so here.
