Stashmetrics was a seed-funded SaaS startup focused on social media analytics for content marketing, running from 2012 to 2017. The core problem it addressed was the difficulty of discovering what content was actually resonating within a specific niche community - as opposed to what was broadly popular - and finding influential accounts within that community.
The two main capabilities were:
- Niche content discovery - given a community of interest, surface the content and topics gaining traction specifically within that group, rather than across the platform as a whole.
- Influencer similarity search - given a small seed set of influential profiles, locate similar accounts using profile text similarity and common list membership, combining graph analysis with vector similarity search.
The platform targeted Twitter, Instagram, Facebook, YouTube, and Tumblr.
The key technical challenge was achieving interactive-speed similarity search across large profile and content datasets with minimal infrastructure budget. We implemented a custom BK-tree index over sparse TF-IDF vectors, built in Go. This reduced profile matching from tens of seconds to tens of milliseconds. This enabled features that would otherwise have required far more expensive infrastructure to deliver at acceptable latency.
Stashmetrics ran for five years before closing due to funding.
Notable technical details
- TF-IDF Analysis - extracting salient terms from profile bios and content to build community topic models
- Vector Search - BK-tree index for fast k-nearest-neighbour search over sparse term vectors
- Task Queue - MongoDB-backed DAG task queue driving the data pipeline