Social Media Engagement Cascade Simulator

Campaign Setup

Content quality (hook strength) 60

Targeting precision 40

Ad seed reach (people) 5

Feed algorithm decay / hop 30%

Ready.

Engagement Metrics

Reach

Likes

Comments

How it works

The audience is a 200-node social graph (a "small-world" network: each person is linked to their 4 nearest neighbours plus a handful of random long-range shortcuts, like real follower graphs). Each person has a hidden interest score i ∈ (0,1) describing how well your niche matches them.

A post spreads using the Independent Cascade Model, the standard information-diffusion model in network science:

seed set S₀ = weighted sample of "reach" people,
    weight(person) = interest^(1 + 3·precision)

for each newly-exposed person p at hop h:
  affinity  = interest(p)^(1 + 2·precision)
  engage_p  = quality · (0.25 + 0.75·affinity) · (1 − decay)^h
  if rand() < engage_p:  p engages (may like / comment / share)
  if p shares: every unexposed neighbour of p becomes exposed at hop h+1

Content quality scales every engagement probability directly — better hooks convert more of everyone who sees the post.
Targeting precision sharpens how strongly interest score determines both who is seeded first and who engages — low precision spreads the ad budget thin and evenly, high precision concentrates it on the audience most likely to respond.
Seed reach is the initial paid/organic impression count before any sharing happens — the cascade's generation zero.
Feed decay shrinks engagement probability by (1 − decay) at every hop, modelling how a feed algorithm de-prioritises older re-shares.

This is the same mechanism marketing teams model with the Bass diffusion equation and platforms use internally to forecast a post's total reach before it plateaus.

Not reached Exposed (pending) Engaged Saw, no engagement

⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.