Key Traits of Shareable Viral Videos

I've generated 1B+ views from short-form videos.. And I've noticed a pattern: The videos that explode share two specific metrics that stand out from the rest. It's not likes, or comments, or follower count. The two numbers that actually predict virality are: The first metric that matters most for a viral video is Average View Duration (AVD). This measures how long the average person watches your video before swiping. So it's not about your total watch time. But it’s about holding viewers’ attention right from the first few seconds. When AVD is high, the algorithm recognizes your content is engaging viewers. And it rewards you with more distribution. I've seen this with podcast clips. When viewers stick around for the full clip, that video almost always outperforms others. The second crucial metric is the number of shares. When someone shares your video, they're essentially saying "this was so good, so interesting that I had to send it to someone else." That's valuable for the algorithms. Shares signal to platforms that your content is compelling enough to be recommended to others. This creates a powerful feedback loop: More shares → more views → more algorithm favor → even more distribution. When analyzing video performance, I look closely at where viewers drop off. And I can pinpoint exact moments when people lose interest. For example… If viewers abandon a video at a specific point, I'll check what happened there, often it's something confusing or unclear. This data gives me a chance to learn and improve. I've literally re-edited videos, fixed the problem spots, reposted them, and seen dramatically better results. It's an iterative process that gets better with each attempt. The key takeaway for any creator… Your hook (first 3 seconds) must clearly frame what viewers will get from spending time with your content. If they're confused about the premise, you've already lost them, and your AVD will suffer accordingly. If you're creating short-form content, focus on these two metrics: - Average View Duration (AVD) - Number of shares These are the real indicators that your content is resonating and that the algorithm will reward you. Likes are nice, but retention and sharing drive real growth.

I spent 10 hours analyzing why people share content. Here's what I learned... Virality isn’t random. It’s engineered. The data says most people get it completely wrong. Here’s what actually makes content spread: The 5 Science-Backed Reasons People Share Content: 1. People share what boosts their status. (If sharing your post makes them look smarter, funnier, or “in the know,” it spreads.) 2. Strong emotions = more shares. (Anger, awe, anxiety, excitement. “Mildly interesting” = death.) 3. Practical content travels. (If it saves time, money, or effort, people hit share.) 4. Stories > facts. (We forget statistics. We remember narratives.) 5. Timely content wins. (People share what feels urgent and relevant today.) How to Use This to Go Viral: Write something screenshot-worthy. (Make it so valuable people save it for later.) Infuse high-arousal emotions. (If your post doesn’t make people feel something, it dies.) Tie it to something happening NOW. (Trends, breaking news, industry shifts—attach your content to the moment.) The best content doesn’t just get read. It gets shared. (or saved) Are you posting forgettable content… or are you creating ideas that spread?

MrBeast’s onboarding docs leaked and wow, I never realized how intense producing best-in-class Youtube videos was. It’s a 36-page memo but here's 4 points useful for startups and Inari (YC S23): 1️⃣ Extreme focus: make the best YOUTUBE videos. The memo starts off clarifying that the team is creating the best Youtube videos, period. Not funny videos. Not good looking or Hollywood videos. Just be the best in Youtube. Clarifying that goal means they can form processes to win that niche. His example: producing Youtube videos means they can ship engaging videos weekly instead of once or twice a year like in Hollywood. 2️⃣ Being the best requires operational excellence. The memo feels oddly reminiscent of PG’s founder mode essay and startup best practices: be high agency and obsessive about details, eliminate bottlenecks, performance manage anyone holding you up (especially external dependencies), and keep open lines of comms. - “I want less excuses in this company. Take ownership and don’t give your project a chance to fail. Check. In. Daily. Leave. No. Room. For. Error.” - “If multiple people are responsible for the same thing, then that’s a problem and needs to be fixed immediately.” - “Don’t take anything at face value, always dig.” 3️⃣ Anatomy of a viral Youtube video. “Hook people at the start of the video, transition them to an amazing story that they are invested in, have no edull moments, and then have a satisfying payoff at the end of the video with an abrupt ending.” Titles are important for getting someone to click. Add intrigue and make it feel extreme. If a user clicks a video called “World’s Largest Bouncy Castle” but doesn’t see that, they’ll feel lied to and click off since it didn’t hit expectations. The first minute is the most important since it has the most loss. This is why we freak out so much about the first minute and go above and beyond to make it the best we freakin can. Match expectations of the clickbait and front-load the interesting. Minutes 3-6 are the next most important parts where you plan all the most exciting, interesting, and simple content. The goal is to make them fall in love with the story, the people, and the overall video itself. If we can get them to watch the 1st half, there’s a very high chance they’ll watch to the end. Once you have someone for 6 mins, they’re invested and will continue watching without realizing. Typically the not as good content should be in the back half. Every video needs to "wow". Anytime we do something no other creator can do, that separates us and makes our videos special. It changes how they see us and it makes them watch more videos and engage more. 4️⃣ Use “consultants” as cheat codes. The memo pushes for using consultants (or experts) so they don’t reinvent the wheel. Someone has already done what you’re trying to do and likely spent years perfecting it. Find them. Pay them. Use them. Why waste time rediscovering what someone else learned previously?