I launched one week ago, here is all launch and usage data, I'd love to know what you think

cah5896

New member
One week ago, I launched talknative.app, an app to practice a language 24/7 by talking with a native speaker created with IA.

Here is a few numbers.

Launch and usage metrics
  • 87 votes on Product Hunt, 44 comments, 21st day rank
  • 905 visits on landing page
  • 428 unique downloads (47%)
  • 282 accounts created after downloading (66%)
  • 242 used the app after creating an account (86%)
  • 50 used the app more than once (21%)
  • 15 DAU
  • 12 subscribed, among those 7 used the app more than once (so 14%)
  • 40€ MRR, 60€ if the ongoing free trials convert
https://preview.redd.it/db4jxxmtlol...bp&s=6afd0f221477a70dac2c495308f27c5595724eb8

https://preview.redd.it/m28wmf0vlol...bp&s=830f94fffa7761e7055458dbd387d4d82252bd08

https://preview.redd.it/jeks6w6wlol...bp&s=28a335afc17c1b3af45087122616fd29b82c6ebf

My insights based on those metrics and feedback I got
  • It’s definitely possible to launch a product for free by leveraging your network (LinkedIn for me)
  • When you learn a language, the goal is to talk with ease to a native speaker, and TalkNative innovates with its “learn by practicing” approach, unlike traditional apps where you spend months learning grammar, vocab and sentences while being comfortable speaking/writing stays a long far away dream
  • Onboarding is simple, and the 1st exchanges create a Waouh effect
  • Having to create an account before testing pushes people away
  • Despite the Waouh effect of the 1st exchanges, 80% don’t come back the next day, because (1) the app doesn’t create a habit and (2) full beginners quickly feel overwhelmed by the conversation
  • The app has a good market-fit for people above beginner level, but less for beginners, who represent 60% of users
  • 14% of active users subscribe, and the average number of messages sent per day varies between 4.6 et 8.2, the 10 free messages per day limit is too high
Still hard to say if the idea is validated or not 🤔

I’m thinking I should try to improve the experience for beginners as well as retention before deciding.

What do you think?
 
@cah5896 The first thing the app asks me is to enter my email address and unless I open the link in the email, I can‘t even enter the actual app. At least let people try it out or have a look at it.

Apart from that the logo and the screenshot on your website suggest it‘s an app to learn Japanese. Had to check twice to confirm that other languages are also included.
 
@mak123 Definitely should let users test before signing up

Yes indeed I need to make the app look more international. It was for Japanese only at the beginning but not anymore.

Thanks for the feedback!
 
@cah5896 Congratulations! This one looks interesting to me. I am curious what kind of tool did you use to track the numbers and make the graphics above? I just want to learn. Besides Linkedin, did you try other channels? I am new to this area, thanks for sharing your experience.
 
@deelo00 Thanks!

For launch data, it’s a mix of analytics from the platforms I use: Vercel, App Store, Play Store..

For usage data, I use Metabase on premise (I host it on Railway). Metabase is plugged to my database and I can build graphs from there

I tried LinkedIn, Twitter, WhatsApp, Instagram (private account), Reddit, Snapchat. Basically all apps I used.

LinkedIn was the best by far because that’s where I have more people, and they all work in tech so they like to try new products.
 
@actionjaxon Still early for me to say, I'm still trying to find the right balance between which model to use (using 4 now, maybe 3.5 is enough for the job?), costs, subscriptions, and how much users on the free plan can cost me. To give you an idea, on launch month, I paid 50$ for 250 users. Expensive, but I didn't pay it, I have 2500$ of free credits granted by the Microsoft For Startups program.
 
@cah5896 That's interesting! I usually don't go through all the legal paperwork and fees until my idea receives consistent good feedback, so that's nice to hear.

I'd need to apply with a linkedin profile, but I don't use that site much.

Do you need to have a strong linkedin profile?
 
Back
Top