How Software Detects What You Want Based on Your Social Media Posts
A system that reads your social media posts to figure out your intent, then automatically serves ads or updates your profile based on how likely you are to actually buy or do something.
Patent Number
US 8521818
Status
Active
Filing Date
June 21, 2012
Grant Date
August 27, 2013
Expiration
~June 2032 (estimated)
Claims
26
Assignee
Solariat Inc
Inventors
Conor McGann, Jeffrey Eric Davitz
Citations
34 forward · 46 backward
What it covers
This patent describes a system that monitors social media posts to identify what a user is trying to accomplish. It breaks text into segments, identifies the 'intention type' (like a complaint or a request), and extracts 'intention topics' using linguistic analysis like n-grams and part-of-speech tagging. The system then scores these intentions based on how 'actionable' they are—meaning how likely the user is to respond to a specific intervention. Finally, it automatically triggers actions, such as showing a targeted advertisement (a 'creative') or updating the user's marketing profile.
What it doesn't cover
- —Does not cover human-to-human manual moderation or customer support responses.
- —Does not cover general sentiment analysis that only labels text as positive or negative without identifying a specific actionable intent.
- —Does not cover systems that rely solely on metadata like location or time without analyzing the actual textual content of the post.
The clever bit
The system doesn't just identify what you are talking about; it calculates an 'actionability score' to decide if it is even worth the effort to show you an ad, preventing wasted marketing spend on low-intent interactions.
Why it matters
This technology is the backbone of modern 'intent-based marketing.' By moving beyond simple keyword matching to understanding the goal behind a user's post, companies can deliver ads that feel more relevant. It represents the shift from passive advertising to proactive, automated engagement in the social media era.
Real-world examples
- 1.Targeted ads appearing after a user asks for product recommendations on Twitter.
- 2.Automated customer service bots that detect a complaint and escalate it to a human agent.
- 3.CRM systems that automatically update lead profiles based on social media activity.
Generated by PatentBrief · Not legal advice · patentbrief.org
US 8521818 · 2026