Let's get this party started! 🎉
Some appetizing data from our launch 🚀
Hello there! Excited to share some preliminary stats from this week.
Here’s the launch thread that accompanied our App Store release. Fun fact: we are now 4 for 4 on viral Reddit threads.
Remember when we predicted this was a “button” we could push for hundreds of new users at a time? It’s now clear this is a pretty reliable way to kickstart the flywheel. Stats from the post continue to confirm a highly favorable response (99% upvoted) and unprompted word-of-mouth (112 shares).
We have bigger “buttons” we’re planning to push too, so stay tuned for those. This week allowed us to kick the tires and iron out some bugs, and the fun’s just getting started!
The stats
So far this week we’ve had over 750 installs on iOS alone, mostly from Reddit. The first day alone brought more downloads than months on TestFlight. Our Android app isn’t fully instrumented yet, but we’ve interacted with over a dozen Android users and there are many more who haven’t talked to us—more data there soon.
Our primary goal at the moment isn’t to juice the number of downloads, but to keep shipping experiments for improving acquisition, activation, retention, and referral in the app.
Here’s the first measurement of import: onboarding efficacy. This release (called Barracuda1) achieved 97%, up from 87% in our last launch2. This is a huge improvement and means we’re less likely to lose users before they’ve even played with the app.
If we extend the activation funnel out a bit further, we can take a look at progression through the first two levels. Here we see that we’re losing about a third of users between finishing onboarding and completing their first level (457/689 made it).
This is some low-hanging fruit to fix, so we’re actively working on it—because once people connect with the app they get cruising!
250 users (about a third) completed 6 or more levels—our current threshold of significant engagement. This fraction is around the same as our last test, though it’s probably an underestimate since it’s only a few days since they installed the app, and we see some folks come back and pick it up later.
But, the folks who really get into the app, are getting much further. Our most prolific users have played through levels 80, even 90 times. Wait, what? There are only have 54 levels total at the moment! These folks are replaying levels, most likely in the quest to 3-star all of them.
We can’t wait to see what happens with further work on retention and finding ways to re-engage users who haven’t quite connected with Parsnip yet. We’ve barely scratched the surface there.
Some folks, though, don’t seem to need our help! Here’s a someone who crushed all the levels in Parsnip in a few hours, on the pre-release Android app no less.
We have lots more data behind the scenes, so what would you like to see? Would love to hear from the product minds and growth hackers out there.
Surprise and delight
The best parts of this launch, though, weren’t the installs going up or the metrics that were improving. It was seeing the magical moments of surprise and delight from users who connected with Parsnip and realized what it could do for them.
Here’s an Android user on Discord, after we helped them install the beta APK:
All the Give Feedback buttons in the app send users’ ideas directly to a Slack channel via Segment. This person was so touched that they felt compelled to tell us exactly how they felt, twice!
The first quiz questions that any user sees are multiple choice, but this was a sequencing question: ”put these steps in the right order”. What we think happened here: seeing this novelty triggered a moment of delight as it revealed that Parsnip is far more than just a quiz—it’s a game!
Asks
Everything’s moving in the right direction here, but there’s much work ahead. We know that building a great product isn’t about getting it perfect out of the gate, but about relentless iteration to kindle these sparks of joy and connection into a roaring fire of enthusiasm.
A few things we’d love your help with…please e-mail me if you have any ideas:
Until we launch either community-driven social features or ways to help users further in the shopping & cooking process, our main bottleneck is content3. Know any food writers, culinary instructors, or anyone who is interested in teaching cooking to a huge audience? Please introduce us!
We’re exploring ways to get Parsnip in front of bigger audiences. Can you think of any content creators on social media who have big followings, or are awesome at creating viral TikTok videos? We’d love to meet them too!
Finally, help us build our community by joining our Discord and stirring a more active discussion around cooking. You can post about what you’re making, what you’re eating, and request help & advice. We’d love to help you on your cooking journey even outside of the app.
If you liked this post, subscribe to see more updates as we build Parsnip! Don’t worry, they won’t all be filled with nerdy product data 😄.
Release codenames for Parsnip are named after fish that don’t contain “fish” in their name. So, Anchovy (our first iOS release) is kosher, but you’ll never see “Parrotfish”.
For comparison to our last release, check out this previous post.
Any cooking product is requires creating content, but Parsnip is different because the tech tree we’re creating is essentially the set of LEGOs for cooking knowledge (game levels) rather than what you can make with the LEGOs (recipes). The number of things you can make with a LEGO set is exponentially bigger than the variety of the LEGO pieces themselves—or put another way, Parsnip requires exponentially less content than any recipe-based platform to cover the same number of dishes. This creates incredible leverage, to be explored in a future post… here’s a sneak peek:
