All scores are calculated per record for the current search period and then accumulated. The tile you see on the scores dashboard is the average of all record scores, and every breakdown in the table is the average of all records that have that particular contact/account/tag/agent/etc. at any point in the life of the record.
A score is (a) the number of times a metric or tag (aka “signal”) occurs, (b) dictated by whether to count per signal, per message, or per record, (c) multiplied by its weight. The combination of these factors is called the "load".
A “points”-style score has a starting value (often null) and goes up or down as we detect signals in the conversation. Scores with positive and negative attributes tend to have a default range of -100 to 100. For a score like Product Friction or Risk of Bad CSAT, it is bound by a 0 to 100 range, as the score itself represents negative activity and only counts up.
Linear scores add up as you might expect, and can optionally be bound to min and max values.
Sigmoid scores behave logarithmically near the min and max in order to approach but never exceed those values while maintaining separation between values. The result is that sigmoid scores can be compressed near the min or max values, and thus are not exactly the result of adding up the loads.
Updated about 1 year ago