Victoria White 0:00 Good thing we're yappers. 0:01 Honestly, it's very fast, so. 0:03 That's great. 0:03 This is maybe ideal. 0:06 Da da da da. 0:07 Oh, let me go present. 0:09 Oh. 0:10 Can I plug this in? 0:11 Yeah. Speaker B 0:12 Let's do that. 0:13 For the people at home. Victoria White 0:15 Hey. 0:16 What do they want to see? 0:18 Uh. Speaker B 0:19 Wow. 0:20 Extended, extended, yeah. Victoria White 0:22 Extend that display. 0:25 Bada bing, bada boom. 0:26 Pretty cool. 0:28 Presenter. Speaker B 0:31 Yay! Victoria White 0:32 This is the presenter. 0:39 Oh, good. Speaker B 0:39 Yeah, it's doing. Victoria White 0:40 Okay. 0:43 Boop. 0:44 Okay, you guys. Speaker B 0:44 Thanks, Tori. Victoria White 0:45 Yeah, thank you. 0:47 Sorry for being late. 0:48 Thanks for waiting for me. 0:49 Hi, everyone. 0:50 I'm Tori. 0:51 Welcome to my talk. 0:53 It used to be what 350,000 users taught me, but hell yeah. 0:57 Now it's 400,000, holy crap. 1:02 What I'm hoping to do today, if, and I'm hoping it's gonna be useful, is just distill some of my learnings and Reed's and I's learnings from this past year and package them up here for you today, um, with the hopes that you don't make the same mistakes that we made and, oh, how do I get over there? 1:26 Oh, there we are. 1:27 Wait, sorry, you guys. 1:28 Hold on a second. 1:30 My mouse is— there she is. 1:33 Was I over there? 1:34 Yeah. 1:35 And that's on me. 1:37 But also with the caveat that we could be completely wrong about the things that we're suggesting, so do what's right for you, okay? 1:45 So this is the Skylight Squad. 1:48 This is Reed and I, and we built Skylight Social. 1:51 Skylight is a video Instagrammy-style app on the protocol. 1:56 We started building last year in January 2025, right before the big TikTok ban. 2:02 We launched in April and like I said, have since grown to over 400,000 organic downloads, no paid ads, insane. 2:09 And that's Skylight right there on your guys' right. 2:14 Oh, she's beautiful. 2:17 Oh, and there's transitions on this. 2:19 Whoop, okay. 2:20 So with the help of my co-founder Reed, I've kind of distilled my learnings into these three main pillars, which hopefully will help organize this thing. 2:30 Prove yourself wrong, we win together, and share your story. 2:36 So prove yourself wrong. 2:37 Right now, a lot of you are builders, so you probably, like we did, have a hypothesis of what you think users want, and your new north star is to try and test whether that hypothesis is right. 2:47 Not to build every feature, not to fix every bug. 2:50 God knows we still have bugs in Skylight. 2:52 But get as quickly as possible to the answer to this question, I think. 2:59 Hot tip, the first thing you build is almost certainly the wrong thing. 3:03 This doesn't mean your insight was wrong. 3:05 This just means that maybe the solution that you built, users don't actually want to use it. 3:09 Our first version of Skylight, over 10,000 people entered, and all of them Left. 3:15 Not ideal. 3:16 I don't recommend. 3:17 Makes you sad on the inside. 3:20 In fact, we've made at least 3 versions of Skylight's interface. 3:24 We even dabbled in a horizontal swiping interface. 3:28 Again, spoiler alert, that absolutely did not work and people literally hated it. 3:34 The comment you see here, yeah, there was a lot of them. 3:38 People on the internet are sometimes mean. 3:41 Be nice to me in my comments, you guys, OK? 3:43 But don't crash out if your first version doesn't work. 3:47 I think we all have a couple of shots at bat. 3:50 The first couple people who see this— you know, there's 7 billion, wait, maybe 8 billion people in the world. 3:55 You have some opportunity to try again. 3:59 And ironically, our best version of Skylight wasn't the one that users asked for. 4:04 And on social, it often isn't. 4:06 So see what they say with their feet and with the time they spend on your platform. 4:13 And to that end, I highly recommend PostHog for analytics. 4:17 This really saved us. 4:20 Again, people will tell you with their actions whether or not they're enjoying what you build. 4:24 Warning, don't use analytics to convince yourself it's working. 4:27 You're not trying to get retention from 3.2% to 3.4%, especially in the beginning. 4:34 Yeah, try not to chase these numbers. 4:36 Just use it as a gut check. 4:37 Are users liking what you made? 4:39 This was really helpful for us. 4:41 They have a— PostHog has a free startup program. 4:44 I know some— a lot of people here, when I've told them about this, they're like, oh, I built my own analytics. 4:49 And I'm like, wow, you're a badass. 4:51 But yeah, we really were thankful for that when we got hit with a fat bill. 4:58 Woke up freaking out and then got an email like, hey, you should really join the startup program. 5:03 We were like, oh wait, that would be really great. 5:07 Yeah. 5:08 OK. 5:10 Pillar number 2, we are in this together. 5:13 The people in this room, the people at this conference, they are going to be your greatest strength. 5:17 I know that they were ours. 5:19 And I truly believe that our communal collaboration is going to make or break the things that we build. 5:26 Hot shout out to these 5 folks and, gosh, honestly a ton of people. 5:31 They have talked me off the ledge. 5:33 They have physically come into my codebase and written things. 5:37 We have collaborated with them. 5:39 Some of the things that they have built now exist inside of Skylight making it better. 5:43 Yeah. 5:44 Black Sky, Graze, Streamplace, Germ. 5:47 I just can't thank these guys enough. 5:49 And that's Jimmy. 5:50 He lets us use his space. 5:50 Base. 5:52 Yeah. 5:52 So talk with the people here. 5:54 Try and get a group together so that when you're inevitably crashing out in the middle of the night, you have someone to call. 6:03 This is a big one for us. 6:05 The app protocol is a resource. 6:07 I highly recommend using it. 6:09 The way that we chose to use it was to utilize Bluesky's lexicon so that a Skylight post is the same as a Bluesky post. 6:17 And this was very intentional on our part because we wanted people who stepped into Skylight to not have that feeling of emptiness in a social network. 6:25 And I'm not saying that this is right for everyone to do, I'm just saying that there's some ways that we can leverage distribution, the tools that exist, um, and I think you should. 6:36 That's why it exists, that's what kind of makes us better than all those other sucky platforms. 6:41 Um, use it if you can and if you— what is it called— if you want to brainstorm on how you might be able to leverage what exists in the protocol to like make it easier to market your thing. 6:55 I honestly wouldn't even, I mean you can come talk to me, but I would highly recommend talking to my co-founder Reed. 7:00 I swear this is like his bread and butter. 7:02 He's really good at thinking about how you can get your idea in front of more faces or like utilize what exists for people to see what you're building. 7:14 Most folks won't get it first that we are actively working together. 7:19 This is something I constantly have to explain to people. 7:22 I had an advisor emailing me kind of freaking out. 7:26 They're like, "What are you gonna do about Devine?" And I remember I was like laughing. 7:31 I was like, "Cheer them on. 7:33 That's so cool what Ravel is doing. 7:36 I'm so freaking excited that we now, all of us, have access to Vine's old videos." I mean, I had to remind my advisor, this is OpenSocial. 7:46 This is the whole point of what we're doing. 7:48 I, on Skylight, now have direct access to showcasing Vine's videos. 7:53 And it's not just because, you know, Devine exists. 7:56 Like, Rabble legitimately built a tool to make it easier for us to port over those videos. 8:01 And that's what OpenSocial is all about. 8:04 Yeah. 8:08 So I really, really hope, you know, he gets out there and he kills it, and he already is. 8:14 And I hope that for everyone who's building on this protocol, because we have really big fish to fry, and it's almost certainly not each other. 8:23 There are a ton of badasses in this space. 8:25 Find them, collaborate with them, cheer them on, because if they take down the monster in the room, then we all get to win, right? 8:33 Hey, and that's his real yacht, by the way. 8:36 Shout out to Marky Mark. 8:38 OK. 8:42 Tell your story. 8:43 Marketing is absolutely a part of what you're building. 8:46 Storytelling is a part of what you will have to do. 8:49 We want folks using these things, so we have to make sure we build community around them. 8:55 And I know you guys know that. 8:57 Building in public really worked well for us if you're up for it. 9:01 Obviously, can't recommend short-form video more. 9:03 Make sure to post it on Skylight. 9:06 We made a video before we ever wrote a line of code, and maybe your medium isn't video, but highly recommend trying to figure it out. 9:14 And to that point, if you find a format that works, then roll that thing till the cows come home. 9:20 You see this video? 9:21 I have made this style of video where I point at these guys who suck so many times. 9:25 I've gotten over a million views with this style of video. 9:29 It took me a lot of different tries to find this format, but if you try something that works, stick to it. 9:37 And yes, we have to show up where people are. 9:40 Yes, I am on these platforms, and yes, I am telling the people on these platforms about OpenSocial and telling them to get over here. 9:46 And please join me. 9:49 Don't leave me alone. 9:50 I want to be able to tag the things that you built on these platforms when I inevitably make a video about them, okay? 9:56 So join me over here. 9:58 I know that's shocking to say at an OpenSocial conference, but— some language that helped us. 10:03 Decentralized never worked. 10:05 Never say the word app protocol, unfortunately. 10:09 OpenSocial, emphasizing creator ownership, that did work. 10:14 Please put your app logo on the internet so when I make a video about you, I know I can have access to your logo. 10:21 Not just on your blue sky. 10:23 Like page, okay? 10:24 'Cause it's really hard to screenshot those. 10:26 That's a personal request by me. 10:29 Okay, that's it. 10:31 Thanks, you guys. 10:32 Woo!