The situation
Product teams need to understand how users actually interact with their software — not just what they click, but dwell time, hesitation patterns, scroll depth, and interaction sequences. Most analytics tools provide event counts. Behavioural depth requires a purpose-built layer.
The client was a US-based product company with the domain expertise to define what they needed but without an in-house engineering team to build it. JWS owned the full build: PRD, tech stack selection, architecture, and delivery across the POC-to-MVP arc.
What we built
A full-stack SaaS behaviour analytics platform covering event capture, attribute enrichment, data storage, and visualisation — built from zero to MVP in 6 months.
- 12+ behaviour attributes tracked per interaction event — dwell time, click patterns, scroll depth, hesitation, interaction sequences
- Sub-2-second visualisation rendering — dashboards that feel real-time at enterprise data volumes
- Session replay and user journey mapping
- Funnel analysis with multi-step conversion tracking
- Real-time and historical data views
- Multi-tenant SaaS architecture supporting multiple client accounts
- Tech stack selected and owned entirely by JWS
The outcome
POC delivered in the first phase. Full MVP live within 6 months. Sub-2-second visualisation achieved and maintained at scale. The platform is live with active users and in active iteration.
| Metric | Result |
|---|---|
| Time to MVP | 6 months |
| Visualisation speed | Sub-2-seconds |
| Attributes per interaction | 12+ |
| Architecture | Multi-tenant SaaS |
| Stack ownership | Full (JWS PRD to delivery) |
