Yeah that’s pretty much where I see the main value as well. The aggregation makes it easier having everything in one place, but the main thing I’ve been working on is getting it to actually learn the individual user.
It starts with onboarding to give it an initial idea of your taste, then saves, skips, watched titles and feedback continue changing the recommendations. I’m testing that quite heavily at the moment because I want the difference after 20/50/100 interactions to actually be noticeable rather than just saying it
That’s a useful test. The 20/50/100 interaction checkpoints should make it much clearer whether personalization is creating a genuinely better experience or just adding complexity.
Yeah exactly, that’s what I’m trying to prove now. The biggest issue I’m having at the moment is actually getting enough testers before launch. I want to get real people using it so I can see how the recommendations change at 20/50/100 interactions and improve the system before putting it out properly.
Friends and family only get you so far, so getting completely new users with different tastes is probably the main thing I’m struggling with right now.
That makes sense. Since getting enough new testers is the main bottleneck right now, I’d be interested in continuing the conversation beyond the thread — would you be open to sharing the best email to reach you on?
You can reach me at hello@beryxa.com. Feel free to email me there and we can continue the conversation.
About
I started working on WatchSmart because I was fed up with spending more time scrolling through Netflix, Prime, Disney+ and other streaming services than actually watching something.
There are thousands of films and shows
7 Comments
The recommendation side seems like the real product here, not the aggregation.
Curious whether early testers notice the recommendations getting meaningfully better with use, or whether the value is already there on day one.
Yeah that’s pretty much where I see the main value as well. The aggregation makes it easier having everything in one place, but the main thing I’ve been working on is getting it to actually learn the individual user.
It starts with onboarding to give it an initial idea of your taste, then saves, skips, watched titles and feedback continue changing the recommendations. I’m testing that quite heavily at the moment because I want the difference after 20/50/100 interactions to actually be noticeable rather than just saying it
Yeah exactly, that’s what I’m trying to prove now. The biggest issue I’m having at the moment is actually getting enough testers before launch. I want to get real people using it so I can see how the recommendations change at 20/50/100 interactions and improve the system before putting it out properly.
Friends and family only get you so far, so getting completely new users with different tastes is probably the main thing I’m struggling with right now.
yes could you provide me with an email I’ll drop you an email as it will be my personal one.