Aggregated sentiment score
An aggregated sentiment score rolls the tone of every AI answer about your brand into one number, usually the share that's positive.
An aggregated sentiment score is one number that sums up the tone of the AI answers about your brand. Each mention of you is marked positive, neutral or negative, and the score rolls them up, most often as the share of mentions that are positive.
It's a quick way to see whether the platforms speak well of you, and whether that's getting better or worse.
How it's usually worked out
Tools differ, so always check the definition. A common one takes the positive mentions and divides them by all the mentions that have a tone.
For example, say 140 answers name Fernhill Coffee this week. 60 are positive, 70 are neutral and 10 are negative. Counted as positive out of everything, that's 43%. Counted as positive out of positive and negative only, it's 86%. Same answers, two very different scores.
What a single number hides
A score can hide what you need to know. 86% sounds great, until you see the 10 negative answers all say your coffee arrives stale. That's one fixable problem, and it gets lost in the average.
Tone is also judged by a model, which can read the same sentence two ways. That makes the trend more useful than the exact figure. If the score drops, read the answers behind it before you act.
Sentiment in AnswerOrbit
We don't make a sentiment score. Every mention is marked positive, neutral or negative, and we show you how many of each there are, by brand or by platform, day by day. What's said lists the things the answers say about you, with their tone and the pages behind them. See the Sentiment tab in Perception.
That way, a few negative answers about one thing stay visible, and you can go straight to the pages causing them.