Share of voice, sentiment and topics, your brand against the whole category
Note: This agent was renamed from AI Consumer Research to Consumer Insights, and the responding side, Brand Moderator (Review Queue, Rules Engine, Agent Playbook, Action History) and the AI Researcher: is now its own top-level agent. Consumer Insights focuses on listening, and was restructured again: it is now Social / Sub-Brands / Mentions on the
/consumer-insights/*route. The old Overview and Channels pages are folded into Social Intelligence, where Channels is a tab alongside Share of Voice, Sentiment and Topics.
Problem we solve
A brand has thousands of conversations happening about it every day across Instagram, Amazon, Flipkart, YouTube, review sites, and forums. The brand team reads roughly none of it directly. They get a Brandwatch summary once a month, an NPS deck once a quarter, and individual angry tweets escalated by the social team. By the time a real issue surfaces, it's already a Reddit thread.
Consumer Insights is the listening layer. It aggregates everything customers are saying about you, scores sentiment, breaks it down by sub-brand and channel, and points you at the themes that matter. When something needs a reply, the Brand Moderator agent drafts it so the social team isn't drowning.
Value & insights we provide
- Share of category voice and a category rank: not just your own mention volume, but where you sit against every tracked competitor, computed across the connected sources (and the source list is shown, so the number is auditable)
- A single sentiment score you can defend, computed across every channel where your brand is mentioned
- Topic spikes, estimated reach and net sentiment as headline tiles, with export
- Sub-brand-level NPS so you can see which line is loved and which one customers are quietly hating (e.g. Bata Kids' School Shoes at -33 NPS)
- Per-channel health: which platform is declining, which is at-risk, which is strong, with sentiment, volume, and growth side by side
- Top themes customers are talking about per channel, so you know whether the conversation is "quality" or "shipping" or "fit"
- An AI-drafted response queue so the social team can clear a hundred comments in an hour, with each draft showing confidence and the rule it followed
- A Mentions stream so any aggregate can be drilled into the actual posts behind it, including on platforms where you have no official account, since tracking runs on public data
- On-demand deep research across 30+ sources using prebuilt templates, competitive benchmark, crisis deep-dive, campaign impact
Who uses it
The brand manager and insights team live here. The social/community team uses the moderator daily. The CMO uses NPS and sentiment for board materials.
Features in Consumer Insights
- MentionsThe raw stream underneath the charts, every tracked brand mention across channels and conversations, searchable by keyword and filterable by date. This is where you go to read the actual posts behind a sentiment dip, rather than trusting the aggregate.
/consumer-insights/mentions - SocialSocial Intelligence, the main Consumer Insights screen. Four headline tiles (topic spikes, estimated reach, net sentiment, mentions) sit above four tabs that each answer a different question: Share of Voice (where do we rank), Sentiment (how do people feel), Topics (what spiked and why), Channels (where is it happening).
/consumer-insights/social - Sub-BrandsShows you how each sub-brand line is performing in customer perception. Useful when the master brand looks healthy but one sub-brand is silently dragging, a common pattern in multi-line consumer brands.
/consumer-insights/sub-brands