Social Media Post Virality Model
Social media is a considered a best platform to share the
content with friends with common interests. That is why businesses these days
focus on building social media pages and try to build strong presence with
their customers or prospects.
Social media post can be of variety of types. It can be
engaging, informative, creating awareness, alerting, or emotional. These posts
serves specific purposes and that is why their impact is different. Social
media algorithms are made in such a ways that they detect the signals that
requires more attention, have tendencies of seeking attention of other users,
are addictive that user can come again engage with your posts and this is how
the virality of post works.
Mathematical and graphical representation of virality of social media posts
Consider virality of post(virality score) to be a function
of variables. If the variables of the social media posts are functions of
likes, comments, views, shares, clicks with respect to time. Then virality
score is the sum of definite integral functions of them with respect to time.
The curve of the social media viral post is incremental and
rising while that of non-viral is decreasing curve and diminishing. The place
where these curve meet represent the average post. the maxima of the curve
represents the peak performance when virality score is highest, and minima of
the function represents that the virality score is minimum. The area of the
curve shows ,3 dimensional volume of the curve shows
To derive a general integral formula for virality as
a function of time, we consider that virality arises from the accumulation of
user interactions, weighted by their respective influences on the platform’s
algorithm. Let the virality score 
at time 
be defined as Continuous-Time Virality Function V(t) which is expressed as:-
V(t) = ∫0t [ αL(Ï„) + βS(Ï„) + γC(Ï„) + δR(Ï„) + ϵVw(Ï„) + ζK(Ï„) ] · e-λ(t-Ï„) dÏ„
Variable Definitions
- L(Ï„) = Number of Likes at time Ï„
- S(Ï„) = Number of Shares at time Ï„
- C(Ï„) = Number of Comments at time Ï„
- R(Ï„) = Number of Reactions (Love, Wow, Angry, etc.) at time Ï„
- Vw(Ï„) = Number of Views at time Ï„
- K(Ï„) = Number of Clicks (Profile Clicks, Link Clicks, etc.) at time Ï„
Weighting Coefficients
α, β, γ, δ, ϵ, ζ represent platform-specific importance weights.
- α = Like Weight
- β = Share Weight
- γ = Comment Weight
- δ = Reaction Weight
- ϵ = View Weight
- ζ = Click Weight
Typically, Shares (β) and Comments (γ) carry more importance than Likes (α).
Attention Decay Factor
e-λ(t-τ)
This term models the natural decline in audience attention over time. Older interactions contribute less to overall virality.
λ > 0 represents the attention decay constant.
Practical Discrete-Time Model
V(t) = Στ=0t [ αLÏ„ + βSÏ„ + γCÏ„ + δRÏ„ + ϵVwÏ„ + ζKÏ„ ] · e-λ(t-Ï„)
Virality Threshold Condition
A post becomes viral when:
dV(t)/dt > 0 and V(t) > Vcritical
where Vcritical is a platform-specific virality threshold.
Compact Matrix Representation
V(t) = ∫0t w · X(Ï„) e-λ(t-Ï„) dÏ„
w = [ α, β, γ, δ, ϵ, ζ ]
X(Ï„) = [ L, S, C, R, Vw, K ]T
0 Comments