How do I measure audience sentiment from social media comments?

A single sentiment score hides more than it shows. Here is how to measure what an audience thinks from its comments in a way you can trust: context first, then categories, then comparison.

The short answer.

To measure audience sentiment from comments, first remove spam and off-topic comments, read each comment in the context of its video, classify it by what it is about and how strongly the person feels, then report sentiment by topic, not as one overall score. Noodle4 classifies comments by intent, sarcasm, relevance, spam, toxicity and product or brand issues and praise, and lets you open the comments behind each number.

Last updated 19 October 2026.

By the numbers

How a brand responds in public shapes how people see it.

64%: say it matters that brands respond publicly on social media (Sprout Social, Q2 2026)
51%: would consider buying from a brand within months of a crisis being properly addressed (Sprout Social, Q2 2026)

Figures are as reported by each source. Sample sizes and dates are listed under Sources.

Why one score is not enough

“72% positive” sounds precise. It may include a lot of spam, miss sarcasm and treat “ok” as the same as “I love it”. It also hides the question that matters: positive about what? A video can have positive comments about the creator and negative ones about the product.

How to measure it properly

Remove noise. Spam, bots, giveaways and off-topic replies.
Read in context. Each comment with its video.
Classify what it is about. Product, brand, creator, price, delivery.
Classify how it feels. Positive, negative, mixed, and how strongly.
Separate who is speaking. An advocate, an existing customer or a potential customer says different things.
Report by topic. Sentiment for each product and issue, not one number.

Sarcasm and context

Sarcasm is where scores go wrong. “Great, another one that leaks” is negative, with a positive word in it. A model that reads the comment alone is easily fooled. A model that has the video and the surrounding comments is much less so.

Noodle4 classifies comments using the video as context. It can tell whether a commenter is an advocate, an existing customer or a potential customer, spot sarcasm, separate relevant positive and negative comments from spam, toxicity and hate speech, and find brand and product issues or praise even when the comment never mentions them.

Reading the numbers

Look at the base. Fifty comments is not an audience.
Compare like with like. The same product, platform and time window.
Watch the change. A move from last month matters more than the level.
Open examples. Read ten from each category.

Reporting it

A useful sentiment report has the same few sections every time: volume and base, sentiment by topic, what changed, the top issues with examples, and what to do. Keep the examples linked, so a reader can check them.

What Noodle4 includes

Noodle4 is a full social intelligence platform, not only a video tool. It reviews text as well as video and images, and it includes:

Text, video and image review. Reads captions, on-screen text, spoken words, images and comments, so the whole piece of content is understood.
Taxonomies. Organise content and comments into your own categories, so results match how your team thinks.
Visual-level SKU tracking. Spot products and SKUs as they appear in videos and images, not only when they are named in text.
Classifications. Every piece of content and every comment is classified by topic, type and relevance.
Anomaly detection. Flags unusual spikes and shifts, so you see what is trending before it spreads.
Slack and Slack Connect. Reviews, risks and reports go to your team’s channels, and to shared channels with partners.
MCP. Ask about your creators, campaigns and social intelligence from Claude, Cursor or any MCP client.
API. Send content in and get structured results back from your own systems.

That covers social listening, brand monitoring, influencer screening and monitoring, content review and reporting in one place. It is priced on usage, not seats, from £99 a month.

Sources

Checked 19 October 2026. Figures are quoted as the sources report them, with the survey and sample size where given.

Sprout Social: Q2 2026 Pulse Survey, crisis response (18 Jun 2026). 2,250 social media users in the US, UK and Australia, fielded 14 to 20 May 2026. Sprout Social sells social media software.

How we know this

These notes come from building Noodle4’s review and listening pipelines, not from a published academic sample. Cost curves, intent splits, and ASR collisions are from internal R&D and client-shaped tests. They will not match every category. We include them because the failure mode is structural — not because one campaign is universal.

Alex Gray-Johnson is Co-Founder at noodle4, with 8 years in AI classification models and marketing.

Frequently asked questions

Is sentiment scoring accurate?

It depends on the model and the content. Test any tool on comments you have read yourself.

Can Noodle4 tell sarcasm from praise?

Yes. It classifies comments using the video as context, including sarcasm.

Can I measure sentiment on one product?

Yes. Use your taxonomy and product-level detection.