Skip to main content

Listening: analytics

📄️KPI cards with the previous period

KPI cards with the previous period (value, previous, change; change null = new): mentions, posts (posts only: no comments, no reviews), comment_mentions, review_mentions, unique_authors, unique_sources, reactions, comments, shares, views, interactions (= reactions + comments + shares), buzz (= mentions + interactions); interactions_breakdown (share); by_platform (mentions, previous_mentions, change, share, …); daily (mentions, posts, comment_mentions, interactions, buzz…); sentiment, sentiment_buzz, sentiment_by_day (⟨s⟩ and ⟨s⟩_buzz); intent { provider, seeker, general, unknown }, intent_share { provider, seeker, general } (0..1 of the classified mentions), intent_by_day (day + the four counts); media (own|paid|earned); attributes (id, name, type, color, mentions, interactions, sentiment); duration { videos (mentions with a known video length), avg_duration_seconds (null without any), buckets [{ bucket under_15s|15_60s|1_3m|3_10m|over_10m, mentions, share (of videos), interactions, avg_interactions }] }; post_formats [{ format short|reel|live|story|carousel|video|image|text|unknown, mentions, share (of posts), interactions }] (posts only). compare_from/compare_to pick the comparison period (default: the same length right before); attribute_type (attribute|product_line|topic_cluster|tracking_term, comma list) breaks down only the attributes of those types. reviews { reviews, rated (with a rating), avg_rating (2 places, null without any), distribution { 1..5: count }, with_seller_reply, reply_rate (0..1 of reviews, null without any), verified_purchases, by_day [{ day, reviews, avg_rating }], variants [{ variant, reviews, avg_rating }] (top 10), has_review_sources (the project watches a product, shop, place or app) }; by_platform and daily also carry review_mentions

📄️Policy D22), over the distinct authors

Audience estimates, AGGREGATE only (policy D22), over the distinct authors (platform + handle) of the filtered mentions: gender from public display names read as Vietnamese names (middle name Thị/Văn, a curated list of clearly gendered given names; unisex, nickname, brand and page names unknown) and province from the profile location the platform returns (TikTok region, X location, YouTube channel country), normalized to the 34 provinces (former 63 as aliases) or abroad. Pages, news sites, owned accounts and Facebook page posts count in n without an estimate. Any bucket under k = 10 authors is null (shown as ⟨10); when either gender is under k both are. No age, no relationship status. Never returned per author: not in the feed, influencers, leads, creators or any export. Aggregate cached 5 minutes per project + filters