Case Study: Getting Things Done (GTD) achieves searchable, organized access to 100,000+ files and 9TB of content with Repustate's semantic search solution

Semantic Search Solution for Getting Things Done

Getting Things Done (GTD), the productivity-methodology and content provider founded by David Allen, faced the challenge of organizing 20+ years of material—more than 100,000 documents and media files totaling about 9 TB—so practitioners could quickly find relevant guidance. GTD needed transcription, summarization, sentiment analysis, semantic search and content organization but found major cloud offerings expensive, enterprise-focused, and requiring costly third‑party development, so they engaged Repustate to deliver a customized semantic search solution.

Repustate built a pipeline that ingested GTD’s content from Amazon S3, transcoded and transcribed media, applied a custom vocabulary and NER to extract timestamped entities, and enabled concept-based semantic search (e.g., mapping “executive” to related titles and finding the five workflow steps), plus an editing interface that updates metadata on save. The Repustate solution processed 100,000+ items across 9 TB, made related content easily discoverable, eliminated the need for extra development vendors (saving costs), and allowed GTD to deliver targeted knowledge more efficiently with no disruption to their existing systems.

Getting Things Done

David Allen

Getting Things Done