Synththis provides an AI‑driven platform that automatically extracts key themes, generates learning statements, and clusters insights from interview transcripts for service designers and CX researchers. The system runs on a privacy‑first enterprise LLM with hourly data wipes and built‑in anonymisation, ensuring secure handling of sensitive participant information.
Funding
Funding not disclosed
Founders
Product
Problem
Customer experience (CX) research often involves manually reviewing lengthy interview transcripts to identify themes, learning statements, and actionable insights, a process that is time‑consuming, inconsistent, and prone to human bias. Additionally, handling sensitive participant data raises privacy concerns, especially when using generic AI tools that may retain or train on proprietary information.
Solution
Synththis offers an AI‑driven platform that automates the synthesis phase of CX research for designers and CX professionals. The system ingests interview transcripts and extracts precise, consistent key themes aligned with service‑design methodology. It then generates learning statements by applying the “what & why” exercise and attaches relevant customer quotes to preserve the voice of the participant. Across multiple transcripts, the platform clusters themes, learning statements, and quotes to produce actionable insights. All processing runs on a privacy‑first enterprise LLM that does not retain or train on user data, with servers wiped hourly and built‑in anonymisation of participant names, ensuring secure handling of sensitive information.
Target Audience
Primary users are service designers, CX researchers, and product teams who conduct qualitative interview‑based research and need reliable, privacy‑secure synthesis tools.
Features
- Automated extraction of key themes from interview transcripts, calibrated to professional service‑design standards
- Generation of learning statements with contextual “what & why” analysis and linked customer quotes
- Insight synthesis that clusters themes, learnings, and quotes across all transcripts to deliver actionable recommendations
- Enterprise‑grade private LLM that guarantees no data retention or model training on user inputs
- Hourly server wipe and automatic participant name anonymisation for enhanced data privacy
- Full alignment with the Double Diamond service‑design framework, enabling designers to rely on outputs for the entire synthesis phase