An actionable guide: what to follow, what to mute, how to train the algorithm. Includes a 3-Day Feed Audit Challenge.
10 min read
You didn't end up watching four hours of red-pill content because you went looking for it. You started somewhere reasonable — fitness, maybe, or self-improvement — and got walked, step by imperceptible step, into something much darker. That's not weakness. That's the design.
Social media platforms make money when you stay on the platform. Full stop. They've discovered, through billions of data points, that certain emotional states are extremely effective at holding your attention: anger, anxiety, inadequacy, and outrage top the list. The algorithm isn't optimizing for your wellbeing. It's optimizing for your eyeballs.
Once you understand that, the question changes. It's no longer "why do I watch this stuff?" It's: who's training whom? Right now, the algorithm is training you. This guide is about training it back.
Why Your Feed is Making You Angrier
Here's the mechanism, spelled out plainly. You engage with content. The algorithm notes what you engage with and serves you more. But not just "more of the same" — it serves you the highest-performing version of that content. And the highest-performing version is usually the most extreme one.
Mild fitness content doesn't out-perform content that claims "modern men have been made weak." Reasonable dating advice doesn't out-perform content that tells you women are the enemy. Nuanced takes don't out-perform outrage. Every recommendation nudges slightly further. By the time you notice where you've ended up, you've been walking for months.
The radicalization pipeline isn't a sudden jump. It's dozens of incremental steps, each one only slightly more extreme than the last, each one feeling like discovery rather than manipulation. The feeling that you're "finally seeing the truth" is part of the design.
The other problem: rage-bait is designed to feel important. You watch it, you get angry, you comment about how wrong it is — and every single one of those actions tells the algorithm "this content is interesting." You've just voted for more of it. Even your contempt is profitable.
What to Follow: The Five Categories Worth Your Time
Most people never think intentionally about what they follow. They accumulate accounts the way they accumulate stuff — a bit at a time, until the drawer won't close. Start here instead.
Skill-building content. Tutorials, how-tos, people teaching you how to do something specific. You should leave knowing more than you arrived.
Nuanced analysis. People who think carefully, acknowledge complexity, and don't promise you a simple villain. Rare and worth protecting.
Genuine community. People sharing real experience — not performing it. Smaller accounts doing actual things, not influencers performing a lifestyle.
Creativity that inspires. Art, music, craft, writing that makes you want to make something — not just feel inadequate by comparison.
Humor that doesn't punch down. Comedy that's actually clever. Humor that finds something funny about ideas or situations, not contempt for people.
Long-form perspective. Video essays, thoughtful interviews, anything that takes more than 60 seconds to say. Depth is underrated.
How to train the algorithm toward this: watch these videos all the way through. Like them. Save them. Comment something real. Every completed view and genuine engagement is a signal: more of this. The algorithm is dumb — it just gives you more of what you interact with. Use that.
What to Mute (or Unfollow): The Red Flags
You don't owe any account your attention. Here are the patterns worth recognizing.
Alpha/beta/sigma hierarchy content — unfollow. If the content organizes the world into dominance rankings and tells you to compete to rise, it's selling your insecurity back to you. The "alpha" framework was based on retracted research from captive wolves. It's not science. It's marketing.
Dating content that treats people as targets — unfollow. If the framing is "how to manipulate attraction," "overcome resistance," or "game female psychology," you're being trained to see other humans as obstacles. This will damage your capacity for actual connection.
Outrage bait about strangers — unfollow. A screenshot of someone on the internet being wrong. A video of someone behaving badly. Culture war content designed to make you furious at a group you've never met. None of this is information. It's just a spike of righteous anger — which the algorithm loves and you'll pay for later.
Aspirational content that reliably makes you feel worse — mute. There's a difference between inspiration and inadequacy. Some accounts leave you wanting to build something. Others leave you feeling like your life is a failure by comparison. Notice the difference and act accordingly.
Simple answers to complex problems — unfollow. "Here's why you're failing with women." "Here's what's wrong with men today." "Here's the one thing holding you back." Anyone claiming a simple cause for a complex problem is almost certainly selling something — your dependency, your anger, or a $197 course.
Content that makes you feel contemptuous of large groups — unfollow. If you've been consuming something that consistently makes you feel superior to "normies," "betas," or any generalized group — and you'd be embarrassed to explain to someone you respect how much time you spend feeling that way — that's a signal.
The Litmus Test
Ask yourself this: would I say this out loud to someone I actually respect? If you'd be embarrassed to voice these opinions to a mentor, a family member, a friend whose judgment you trust — that gap between what you're consuming and what you'd admit to is the gap between who you're becoming and who you want to be.
How to Train the Algorithm: Four Rules
01. Starve what harms you. Don't comment, even in disagreement. Don't share with mockery. Don't hate-watch. Any engagement — positive or negative — is a vote for more. The only way to tell the algorithm you don't want something is to not engage at all.
02. Feed what serves you. Watch good content all the way through. Like and save it deliberately. Comment something genuine. Completing a video is a strong signal. It costs you nothing extra.
03. Use the controls. "Not interested," "don't recommend this channel," "hide" — these exist for a reason. Use them aggressively. They're not just a courtesy; they're direct retraining.
04. Set time boundaries. The algorithm can only influence you to the extent you give it access. No phone for the first hour after waking. No social media after 9pm. Boundaries on access are the simplest intervention available.
Be patient. The algorithm has been learning your behavior for months or years. It takes consistent counter-training to shift it — typically 2 to 4 weeks of deliberate behavior change before the recommendations visibly move. Keep going.
The Comparison Problem
One thing the "what to follow" lists don't capture: it's not just who you follow. It's how you're consuming.
There's a particular mode of scrolling that isn't really about learning or connecting. It's about checking — checking where you rank, how you compare, whether you're ahead or behind some invisible standard. That mode is almost always corrosive, regardless of what you're looking at.
Comparison mode looks like: measuring your insides against other people's outsides, using content to confirm you're behind, scrolling to find evidence you're not enough, leaving feeling worse than when you arrived.
Growth mode looks like: watching to learn something specific, following people doing work you want to understand, feeling energized or curious when you're done, leaving with something you can use.
The feed itself is only half the problem. The other half is the mindset you bring to it. The best-curated feed in the world won't help you if you're using it to punish yourself.
THE 3-DAY FEED AUDIT CHALLENGE
Day 1 — The Baseline (See it clearly)
Track every platform you open — note it, the time, and how long you were on it. No judgment. Just data.
After each session, rate how you feel: Energized / Informed / Numb / Anxious / Angry / Inferior. Write it down. One word is fine.
At the end of the day, ask: what specific content am I most likely to lie about if someone asked me what I watched today?
Note the gap between total time you thought you spent on your phone vs. what your screen time report says. Most people are wrong by 30–50%.
Day 2 — The Unfollow Massacre (Do the surgery)
Go through your following list on your most-used platform. Set a 10-minute timer. For each account ask: does this make me more capable, more curious, or more connected? Or does it make me angrier, more anxious, more contemptuous?
Unfollow at least 10 accounts that reliably leave you worse off. If you hesitated before unfollowing someone, that's probably a sign you should.
For every account you unfollow, find one to replace it — something skill-based, creativity-focused, or genuinely informative. Engage with that new account for real.
Use the "not interested" button aggressively on any recommended content that looks like rage bait, outrage fuel, or hierarchy content. Don't watch it first. Just hide it.
Set one new time boundary: no social media for the first 30 minutes after you wake up. Put your phone in another room to make this easier to keep.
Day 3 — The Reassessment (Notice what shifts)
Do the same check-in as Day 1: track platforms, time, and how you feel after each session. Compare to Day 1's results. What's different?
Notice the cravings. Did you find yourself wanting to go back to the accounts you unfollowed? What were you actually looking for — information, or a familiar emotional hit?
Identify the one change from Day 2 that felt most impactful. What do you want to keep doing beyond this challenge?
Write down one specific thing you want to use your reclaimed attention for — a skill to build, a project to start, a person to actually call. Make it concrete.
Make one new rule for yourself going forward. Not a resolution — a specific, behavioral rule. ("I don't watch videos that use the word 'alpha' in the title." "I don't open TikTok before I've eaten breakfast.") Small and specific beats vague and aspirational.
The Real Goal isn't a Perfect Feed
It's worth being clear about what you're actually after here. You're not trying to build a feed so curated, so virtuous, so optimized that it makes you into a better person by osmosis. That's just another form of passive consumption.
What you're actually trying to do is shift the balance. More signal, less noise. More capability, less dysregulation. More curiosity, less contempt. Even a partial shift changes what you think about, what you believe, and — over time — who you're becoming.
The algorithm isn't your enemy. It's a mirror — a warped one, optimized for someone else's profit — but still a mirror. What it's been reflecting back at you says something about where you've been looking. The question now is whether you want to keep training it, or whether you want to train it differently.
You have more control than the platform wants you to believe. Your attention is the most valuable resource you own. Spending it deliberately — on content that builds you up rather than wires you for anger — is one of the highest-leverage things you can do for the person you're trying to become.
Practice prompt: Spend ten minutes scrolling your main feed right now — not to consume it, but to audit it. After each piece of content, ask yourself: did I seek this out, or was I led here? Does this leave me more capable or just more activated? What would I have to do differently to build a feed that actually serves who I'm trying to become?