the viral distribution system

may 2026

i've put 600 million organic views through this system across tiktok, instagram, and youtube. most of it for products other people built. some of it for products i built. a chunk of it for things i now wish i hadn't promoted. the system is the same regardless.

this is what it actually is, with the math the courses skip.

what viral distribution is not

it is not a hook. it is not an algorithm hack. it is not a posting time. it is not "go all in on tiktok." every one of these is a component. none of them is the system.

viral distribution is the practice of producing creative at a rate fast enough, and across a surface area wide enough, that you can run a real selection process on it. the algorithm is your ranker. your job is to feed it enough candidates that the winners aren't accidents.

if you post twice a week, you don't have a distribution system. you have a content schedule. they look the same to a beginner. they are completely different businesses.

the four levers

every distribution system i've ever run, mine or someone else's, comes down to four levers. you can compete on any combination.

volume — how many candidate pieces of creative you can ship per week.
variance — how different from each other those candidates are. high-variance volume is the only kind that works; low-variance volume teaches the algorithm you're spam.
leverage — how many independent surfaces (accounts, platforms, formats) you can ship the winners across once they're identified.
iteration speed — how fast you can take a winner and produce 20 variants of it before the audience moves on.

most operators compete on volume alone. that's why they plateau. the operators who win pick two of the four and out-execute on both.

what i compete on is leverage (200+ faceless accounts, 50+ physical iphones) and iteration speed (a creative pipeline that can turn a winner into 20 variants in 48 hours). i don't out-volume the competition. i out-leverage them.

the hook layer

the hook is not the system. but it is the thing the system rises and falls on, so it matters.

a working hook in 2026 does three things in the first frame:

shows you are about to break a pattern the viewer recognizes.
promises a payoff that is concrete enough to be testable.
costs the viewer almost nothing to find out if you're lying.

most "viral hook" frameworks teach the first one and skip the other two. that's why the videos get scrolled. a pattern break with no testable payoff is a magic trick. a pattern break with a testable payoff is a hook.

i keep a working bank of about 600 hooks tagged by emotional register and pattern-break type. every new product gets matched against the bank, not written from scratch. writing hooks from scratch is a beginner mistake — there are about 40 fundamental hook structures and you should know them cold.

posting cadence — the actual numbers

across all platforms i've worked, the cadence that has consistently produced compound returns is:

3–5 pieces of creative per account per day for the first 30 days of an account's life.
1–2 pieces per account per day after the account has 5k+ followers.
kill any account that isn't averaging 1k views per post by day 60.

this cadence is brutal. it's also the only one that's worked. operators who try to "post less, post better" out of fatigue or vibes are running a content business, not a distribution one.

the multi-account math

the math people skip:

if you post 1 video per day from 1 account and your hit rate is 1 in 30, you find a winner every 30 days.

if you post 1 video per day from 200 accounts at the same hit rate, you find a winner every 4 hours.

the difference between these two operations is not 200x. it's the difference between a hobby and an industrial process. the algorithm doesn't care which one you are. the audience doesn't care. only the operator who's tired of waiting cares.

this is why i run 200+ accounts. not because each account is valuable on its own. because the rate of winner discovery across the network is the only metric that matters.

the honest math on revenue

people quote "600m views" like it's a revenue figure. it's not. the conversion from organic views to revenue runs roughly:

brand awareness — pennies per thousand views. if your only kpi is awareness, this is a slow business.
direct response (link in bio, app install) — typically $0.50–$3 per thousand views, depending on category.
owned audience capture (newsletter, dm funnel) — typically $1–$5 per thousand views over time, because the audience compounds.

the only model i've seen produce real businesses on top of organic distribution is the third one. you use the views to build an owned audience. then you sell the audience things forever. paid ads can't compete with that, because their unit economics get worse as you scale and yours get better.

i did this for marketeros — 5,000+ subscribers built almost entirely from organic distribution. it's the most boring playbook in the world. it works.

what kills distribution systems

three things, in order of how often they happen:

the operator falls in love with one platform. the platform changes the algorithm, the operator's whole business halves overnight.
the operator confuses creative quality with creative discipline. discipline is the volume + variance loop. quality is what the algorithm rewards once discipline has produced enough candidates.
the operator builds the system around themselves. if you can't go on vacation for a week without the engine stalling, the engine isn't an engine. it's a job.

what i'd build if i was starting today

shorter version: i'd build exactly what i'm building.

the creative engine. (ai ugc + human editorial control.)
the distribution engine. (200+ faceless accounts across 50+ physical iphones.)
the owned-audience engine. (newsletter, community, direct contact.)

the three layers are independent businesses with independent moats. a lot of operators build one and call it a system. it isn't. the system is all three running in series, with the output of each measured weekly.

if you're trying to build any of these layers and want to compare notes, email me. i answer everything.