I've been running a true crime channel for 18 months. Six thousand-plus subs, 25 long form videos, and just over 300K total views. Each upload is a writing, filming, and editing slog that takes me three to four weeks, so quality is high and the turnaround is slow. At this size, the negativity is ramping up on a curve that matches the view counts. It's not just constructive, it's pointed and personal. I've tried to separate signal from noise, but the recent comments about storytelling or audio get more vicious the bigger the numbers get. Shouldn't surprises be proportional to reach? Everyone says scale brings hate, but I want a playbook for handling it, not just a shrug.

What works for me so far is forcing a filter between the comment and my ego. I wrote a pointless little classifier to catch the worst offenders. It's not ML, just a regex pattern. Here's the stripped version that hides anything that matches explicitly derogatory terms.

import re

nasty_terms = ["stupid", "garbage", "fake", "idiot", "donkey"]
TOXIC = re.compile(r"(" + "|".join(nasty_terms) + r")", re.IGNORECASE)

def should_hide(comment):
 return bool(TOXIC.search(comment))

The filter catches maybe 70% of the trash. For the rest, I use a manual rule: if a comment mentions a specific fix or resource, I read it. If it only mocks or attacks, I hide it and never reply. I've heard the "thanks for the algorithm boost" clapback, but replying to toxicity is still serving attention to it, even if you're playing 4D chess with the ranking system. Better to starve the beast.

I'm curious from people who've pushed past 50K subs: does the volume of hate comments plateau, or does it just get more specific as your audience grows? How do you actually ignore the nastiness without becoming a defensive content machine?

(Also, if someone has a better filter pattern. I'm all ears.)