1.7.2 Beginner Tutorial - Measuring Influence with Relevance and Affinity

High Relevance helps identify ideal influencers and other strategic business opportunities. But important signals can be read from Relevance scores of varying levels. Low Relevance is a good indicator that a group of people don’t share many specific interests in common. However, you should still pay attention to the types of interests that appear in the top results to see if you can find any common threads. For example, you can still look for a broader interest theme by considering the top results collectively.

 

Relevance scores are scaled out of one hundred, but that doesn’t mean they should be interpreted as a direct percentage of audience share (that is, a 27/100 score doesn’t necessarily mean that 27% of the audience share that interest). The table below provides direction on how to interpret when a Relevance score is high or low:

 

Score

Rating

100-50

Very High

49-30

High

29-15

Medium

14-10

Low

9-0 

Very Low

 

Juxtapose Relevance and Affinity to make the smartest data-driven business decisions. Use Affinity to understand whether an interest is significant in context to a network. Use Relevance to understand which interests are most significant to a particular group of people in context to the audience. The information below provides direction on how to analyze Relevance and Affinity comprehensively:

 

High Relevance + High Affinity

Not only is the interest significant to the audience or cluster’s overall interest pattern, but it’s also particularly unique to them on the network at large. This is an ideal indicator for a niche influencer.

 

High Relevance + Low Affinity

The interest is significant to the audience or cluster, but also widely popular on the network.

 

Low Relevance + High Affinity

The interest may not be highly significant when compared to the overall interest pattern, but has a higher interest rate amongst the group when compared to the network at large. If selected as an influencer for a targeted audience or cluster, you may be taking a shot in the dark. 

 

Low Relevance + Low Affinity

It’s likely that the interest doesn’t hold much sway over the audience or cluster.