Engagement Forensics
Decompose every like, reply, and bookmark into source, timing, and follow-graph proximity. Spot the difference between organic traction and coordinated amplification — and surface the accounts driving the curve.
How we score it
Analytics Suite · X / Twitter
Forensic analytics for the conversations shaping your brand on X
Move past surface-level likes and impressions. TweetBlocker's analytics layer measures engagement forensics, sentiment drift, thread performance, and competitor signal — so you can read X like a research paper, not a feed.
Six analytics capabilities built for X's noise
Every capability in TweetBlocker is tuned for Twitter/X data — its short-form cadence, reply graph, and quote-tweet topology. We don't recycle generic social dashboards. We read the network the way it actually moves.
Sentiment Drift Analysis
Track tone across replies, quote-tweets, and thread branches in real time. Our classifier distinguishes sarcasm, disagreement, and endorsement — and surfaces when a topic flips from neutral to charged inside a single conversation.
See sentiment reportsReach & Implication Graph
Map first-degree impressions alongside second-order reach through retweets, quote-tweets, and reply chains. Find the bridge accounts whose audiences you don't own yet but are hearing your message anyway.
Read the methodologyThread Performance
Measure drop-off by tweet index, identify the line that lost the room, and reverse-engineer the structure of threads that traveled. We score pacing, hook strength, and CTA placement inside multi-tweet narratives.
Browse thread studiesCompetitor Benchmarking
Side-by-side analytics on the accounts you define as peers: posting cadence, hook vocabulary, share-of-voice, and reply velocity. See who's pulling share inside your category before the quarterly report catches up.
Compare plansExport & Audit Trails
Every report ships with raw evidence — pulled tweet IDs, classifier confidence, and timestamped snapshots. CSV, JSON, and PDF outputs are designed to survive legal review, board scrutiny, and journalism fact-checking alike.
Talk to the team2.4B+Tweets indexed monthly
94.6%Sentiment F1 score
< 90sFirst-pass ingestion
38Languages supported
From raw feed to defendable insight in four steps
Our pipeline is observable end-to-end. Every chart in TweetBlocker can be traced back to the exact tweets, the exact classifier version, and the exact moment we ingested them.
Define the lens
Tell us which accounts, hashtags, lists, or reply graphs matter. We lock in a baseline so future comparisons stay honest. Watchlists can be revised, but never silently rewritten.
Ingest & classify
Tweets are collected via the official X API, then run through our sentiment, sarcasm, and intent classifiers. Every label carries a confidence score you can inspect.
Visualize & annotate
Dashboards surface reach, sentiment, thread pacing, and competitor position in plain editorial layouts. Analysts can pin annotations directly onto the timeline.
Export & defend
Pull a signed report with full evidence chain. The same package you hand to a CMO is the one you can show a regulator, a journalist, or a skeptical board member.
Built by analysts who got tired of dashboards that lie
Most social analytics tools optimize for screenshots, not truth. We optimize for the moment a smart, skeptical person looks at a chart and asks, "show me how you got there."
Methodology, not magic
Every metric in TweetBlocker links to a written methodology page. We publish our classifier versions, our data retention rules, and the limits of what we can and cannot infer from public X data.
Editorial discipline
Our dashboards read like a research brief, not a Vegas slot machine. We show variance, not just winners. A flat line with a clear story beats a glowing number without one.
Privacy by design
We analyze public posts. We do not enrich with private DMs, location trails, or third-party identity graphs. Reports are scoped to the watchlists you define, and export trails are auditable.
What teams use the analytics layer for
From newsrooms triaging breaking stories to brand teams defending share-of-voice, our customers read X the way it actually moves — and act on it before the curve flattens.
"We replaced three separate tools with TweetBlocker's analytics layer. The first week, it caught a coordinated amplification campaign our previous dashboard had been scoring as 'organic growth.' That alone paid for the year."
"The thread performance breakdown changed how our writers structure long posts. We can see exactly which line breaks the scroll, and the evidence trail means editors actually trust the chart."
"The competitor benchmarking view is the only one I've seen on X that shows both reach and sentiment in a way I can defend in a board meeting. The export trail is the part that closes deals."
Next stop
Pair the analytics with the methodology
Numbers without a method are a screenshot. Read the full breakdown of how we score engagement, weight sentiment, and validate reach on X — straight from the team that built it.