The Mavrck platform calculates estimated impressions using an algorithm that was developed based on a study done by Stanford University and Facebook's data science team called Quantifying the Invisible Audience in Social Networks.
This algorithm is used because many social network APIs do not reveal how many impressions an individual's post receives from their friends.
The impression algorithm estimates impressions based on an individual's friend/follower count and the amount of engagement an individual's post receives. That's because the more engagements a post earns, the more a social network exposes the post to additional friends and followers.
- If a post does not have ENGAGEMENTS, then IMPRESSIONS equals FRIEND COUNT x 29%
- If a post does have ENGAGEMENTS, then IMPRESSIONS equals (FRIEND COUNT x 35%) + ENGAGEMENT MULTIPLIER
The ENGAGEMENT MULTIPLIER equals (ENGAGEMENTS x AVERAGE FRIEND COUNT BY AGE BRACKET x 29%). We apply the AVERAGE FRIEND COUNT BY AGE BRACKET factor because we have determined that the average friend count is correlated with age and that a majority of an individual's friends are of a similar age.