PUBLIC METHODOLOGY
How TrendBoss Analyses Viral Facebook Posts
The exact TrendBoss method for metric coverage, creator medians, outlier lift, qualification thresholds, evidence confidence and source verification.
DIRECT ANSWER
TrendBoss analyses a Facebook post by comparing its measured engagement with the median of captured metric-backed posts from the same creator. A post qualifies for Viral Discovery from 1.25 times that median. Five measured posts create an early estimate; ten create the normal evidence state. Missing metrics remain explicit.
The method, step by step
1. Preserve the source
Every captured post keeps its original URL, creator identity, capture time and the visible fields available on the page.
2. Separate measured from source-only
A source remains useful even when Facebook does not expose consistent metrics. It is excluded from numerical scoring until evidence is available.
3. Calculate engagement
TrendBoss uses the visible, normalised engagement signals available for the captured post. Components can include reactions, comments and shares.
4. Build the creator median
Measured posts are grouped by creator. The median reduces the influence of a single unusually large result.
5. Calculate outlier lift
Post engagement is divided by the creator median. A 2.0× result means twice the creator’s captured median.
6. Apply evidence thresholds
Five measured posts create an early estimate. Ten measured posts create the normal baseline state.
7. Qualify discovery results
A measured post enters Viral Discovery at 1.25× or more of its creator median, with its evidence state visible.
8. Verify before interpreting
The researcher opens the original post and studies the hook, topic, structure, media and offer in context.
The calculation
Example: a post with 1,300 engagements against a creator median of 258 produces a 5.04× lift.
Why creator-relative analysis matters
Raw leaderboards favour accounts with large audiences. A creator-relative baseline asks whether the result was unusual within that creator’s own captured history. It can surface a smaller absolute post that represents a stronger departure from normal performance.
Known limitations
- Captured history may not represent the creator’s complete publishing history.
- Publicly visible metrics can change after capture.
- Facebook can expose different fields across routes and account types.
- Engagement does not prove business impact, audience quality or causal creative factors.
- A relative outlier is evidence for investigation, not a prediction of future virality.
TrendBoss is designed to make uncertainty visible. The product does not create missing metrics, promise a causal explanation or guarantee a future result.