Everything Craig Cannon said on any show that made the record, most notable first. Each card names its show and opens the statement there.
Cannon: OpenAI uses the Cold War race-to-build moral justification
“I mean, it's the same one that OpenAI is using right now. Yeah. And most AI researchers.”
Cannon: VR combined with chemicals could surpass physical reality
“I mean, VR plus some kind of, like, chemical combination, I could absolutely see being better than real life. For almost everyone. If not everyone.”
Cannon: Early-stage startups should ignore brand building entirely
“There's this third thing called brand, which is kind of this like vague, misty thing about like how people perceive you. You really don't need to worry about this early on.”
Craig Cannon notes investors assume startups are dying when updates slow down
“When those updates start to get a little bit slower, a little infrequent, they're just assumed to be dead.”
Cannon: Avoid content marketing until user churn is low
“You really want to be doing this after you've made something people want, and you're confident that they're not churning. Because if they're churning, you're just wasting a ton of energy and a ton of money.”
Cannon: Remarketing to website visitors is the cheapest Facebook ad audience
“What this is going to do is capture who's hitting your site, who's logged into Facebook, so then you can later remarket at them, and this is going to be much cheaper than any other kind of Facebook audience that you can create.”
Cannon: Y Combinator has relied on content marketing since its inception
“YC, Although PG and Jessica, Robert and Trevor might not say this, has been doing, like, content marketing since the very beginning. Even before. So with Paul's essays, and then Hacker News was built, and then those essays do really well, and that's the exact …”
Cannon: Anonymous high-volume headline brainstorming eliminates status bias
“And so what this is really good at is, first of all, it removes biases for, like, who's the CEO, who's the new guy, or whatever, or girl, whatever. And it also gets you past the point of the, like, those first three ideas Which are kind of the same from everyo…”
Cannon: Content quality should dictate publishing cadence, not arbitrary schedules
“It could take you three days to write a blog post, and, you know, maybe in the early days you can't do that, but you should let that amount of time dictate how often you can put stuff out, rather than go the other way and be like, shit, I have to get a thing o…”
Cannon: Startup content competes with Elon Musk smoking weed on Joe Rogan
“Most people think that they're competing with people that make the same stuff that they make. That is a hundred percent not true. You are competing with this. Elon Musk smoking weed on Joe Rogan's podcast. I haven't checked the views, but I think it's probably…”
Cannon: Text interviews are a low-effort content cheat code for startups
“Text interviews are like a total cheat code for young companies, because they require way less production than video or audio. The internet really likes it for like SEO, for scraping and it makes the person look really smart.”
Cannon: Authentic engagement on Hacker News and Product Hunt remains undervalued
“Just getting good at engaging with your community on HN and Product Hunt is really undervalued.”
Cannon: Publishing physical books helps startups stand out because competitors won't
“This is super powerful in terms of, like, your influencer status, but really, really hard to track. So this would be later down the road, but it is effective, and because it's hard, most people aren't going to do it, so you can really stand out.”
Cannon: Early startups should host blogs on their own site, not Medium
“People connect with other humans, and so you, like having that little extra touch is nice, but I would say overall like you're gonna want it on your own site for SEO. Yeah, I like, I kind of disagree with people putting stuff on Medium and stuff early on.”
Cannon: Raising VC for a $10M-ceiling business is the wrong game
“If you're gonna take venture money, And the max you can do is ten million a year or something, then like you're kind of, you picked the wrong game.”
Cannon: Tech employees suffer intense, unnecessary risk aversion despite safety nets
“Like I've noticed like a pretty intense risk aversion among a lot of tech employees because they're like, oh, I'm in a sweet spot right now. Not realizing that they could leave and go join another big tech company if their thing never worked out.”
Cannon: Zapier's early content mistakenly targeted Hacker News over actual customers
“In the early days, what they were doing Was making things that Hacker News wanted, but it turns out that Hacker News and their customers was not a 100% overlap. So they were spending all this energy generating this content to get people through the door, but t…”
Cannon: Zapier famously paid employees to relocate away from the Bay Area
“Zapier is a great example of one of those companies. They were paying people to delocate from the bay last year, which was interesting.”
Casey Neistat uses a Gmail filter to trash 'pick your brain' emails
“We did a podcast with Casey Neistat about a month ago and He literally has pick your brain as a Gmail filter to trash.”
Craig Cannon reveals the YC podcast gets 10,000 to 100,000 downloads
“So our podcasts do, like, 10 to a 100,000 downloads per episode.”
Cannon: Annual Podcast Ad Revenue Is Roughly $250 Million
“I mean, I think it's two hundred and fifty million a year now in like ad revenues, which is like tiny considering how much people talk about podcasting.”
Cannon: Podcasters must transcribe audio because Google cannot index it
“I would also say to podcasters, definitely transcribe your stuff, because Google is not friendly to audio, and you want that, like, index stuff right there.”
Cannon: Managing people well provides more leverage than top-tier technical skills
“Managing people well is insanely high leverage. Like, way higher leverage than even if I was the best programmer in the world.”
Cannon: Baidu focuses explicitly on products for over 100M users
“They explicitly are focusing on things for over a hundred million people”