Brand sentiment analysis that scores the video not just the caption
Know where your brand stands, understand what moved it, and see a shift while it is still forming. Clayface builds brand sentiment from what a video shows as well as what is said and written, with every score carrying the posts behind it.
What brand sentiment analysis measures
Sentiment coverage
Brand sentiment scored across every platform and layer
Written posts, comments, spoken audio, on-screen text and the video itself all feed one score for the period, across 25+ platforms.
Sentiment drivers
Sentiment score broken down by theme
Each score breaks into the themes behind it, ranked by how far each moved the number. Shipping, fit and price stay separate rather than averaged.
Sentiment shifts
Brand sentiment tracking week by week
Movement is measured against the period before, so a decline shows in the days it starts rather than the month it is reported. The posts that began it come attached.
475M+
creators database
25+
Social media platforms covered
5B+
Signals processed monthly
Forum & Private Group Coverage for a Powersports Brand
A leading powersports brand tracks twenty motorcycle forums and private Facebook groups alongside social, so owner conversation about accessories and fit surfaces in the same view as everything else.
“We named the Facebook groups on a Tuesday and had them running by Thursday. Everywhere else that has been a month long exercise that does not always end up working."
Know Which Ambassador Content Moves the Brand
A leading beauty brand analyzes every ambassador post as full video, so screen time, spoken claims and on screen products are measured rather than guessed from captions.
“It goes deeper on tone, voice and performance than anything else we evaluated, and it breaks a single video down rather than reading the caption around it.”
Sentiment analysis from video audio and on-screen text
Negative sentiment weighted by reach and intensity
Brand perception measured the same way every month
Why Our Customers Believe Us
“Our sentiment score had been flat for two quarters. Clayface broke it into themes and it turned out two things were moving hard in opposite directions and cancelling each other out. We had been reporting nothing happening.”
“Half the posts we had filed as positive were creators being polite in the caption and critical in the video. Seeing that gap changed how much we trusted the old numbers.”
“We used to find out sentiment had dropped when the monthly deck was built. Now we see it in the week it starts and we can read the posts that caused it.”
See what moved your brand sentiment
Frequently asked questions
Most sentiment tools classify written text: captions, comments, reviews and forum posts. That covers the conversation accurately on text-first platforms and misses most of it on video-first ones, where the opinion is spoken rather than typed. Clayface classifies spoken audio and on-screen text alongside the written layers, so a video carries a score based on what was said in it.
Brand sentiment is the balance of positive, negative and neutral opinion expressed about a brand across public conversation, usually expressed as a single score over a period. It is a measure of what people say, which is why it is only as good as the share of the conversation the measurement can actually reach.
It is used as one by most consumer brands, normally reported monthly or quarterly alongside share of conversation. It works as a KPI only when the method is fixed, because a score built on a changing source set or a changing model is not comparable to the one before it.
Collect every public mention across the platforms where the category talks, classify each one as positive, negative or neutral, then aggregate to a score for the period. The two things that decide whether the number is trustworthy are how much of the conversation was captured and whether the classification read the whole post or only its text.
The comparison holds only when both brands are measured on the same platforms, the same signal layers and the same model. Clayface runs a named rival set through the identical pipeline as your own brand, so the difference between two scores reflects the conversation rather than the method.