A revenue-ready data product in 8 weeks
Objective
Build a modern analytics pipeline to understand user behaviour across features, organisations, and workspaces.
Obstacle
Routine lacked visibility into the complete user journey from onboarding through feature adoption, making it difficult to identify friction points and prioritise.
Outcome
A fully operational analytics platform with centralised product metrics and comprehensive feature tracking across 15+ product areas.
Background
Routine is a next-generation all-in-one work operating system for individuals and teams. Following their development phase, the team recognised the need to move beyond basic event tracking to establish a comprehensive analytics foundation that could scale with their product ambitions.
They partnered with 173tech to build a modern data ecosystem that could untangle the complex relationships between users, organisations, workspaces, and feature usage. The goal: embed product analytics into every decision, from onboarding optimisation to feature prioritisation and organisational growth strategies.
Challenges
Developer Capacity: With no existing analytics infrastructure or dedicated data team, every question required manual SQL queries against raw Segment data. This slowed down product decisions and prevented the team from acting on insights quickly.
Data Fragmentation: Segment events were firing across multiple sources (Application, Controller, API), but the data was not structured for analysis. Critical properties like workspace and organisation were missing from key events, making it impossible to understand how users within organisations and workspaces interacted with the product.
Building Analytics for Unbuilt Features: Routine was actively developing new capabilities while we built the data models to measure them. This meant creating dimension tables, metrics definitions, and dashboard wireframes for features still in development, requiring us to work closely with the product roadmap and anticipate future tracking needs before the code was even written.
Solution
Building the Foundation We kickstarted Routine’s analytics journey with a robust six-week Data Launcher project. The team selected and configured the modern data stack tools for ongoing growth, integrating Segment event data into BigQuery for comprehensive reporting. This foundational work positioned Routine to start building their internal data capabilities.
Data Modelling in dbt: The core of robust, automated analytics is data modelling. This included staging tables which included raw segment data, intermediate tables which then cleaned up this data and production data which fed into our dashboards.
Impact
Designing Analytics For The Future: The unique challenge: we were building analytics for features that did not exist yet. As Routine rapidly developed new capabilities, we designed tracking schemas, data models, and dashboard layouts for product areas still in development. This forward-looking approach meant creating dimension tables and metrics definitions before code was written, working directly from the product roadmap. The result was analytics infrastructure ready to capture insights from day one of each feature launch, dramatically accelerating learning cycles.
Delivering Long-Term Value: We established a scalable dbt project within Routine’s GitHub repository, created four comprehensive dashboard sections in Metabase, and delivered detailed documentation including a Data Dictionary, Segment Requirements specifications, and an Onboarding Flow Analysis. The infrastructure we built continues to serve Routine’s growing data needs and gave them a clear foundation for building internal analytics capabilities. As their co-founder noted: “You created a solid data pipeline which made it much easier for our first analytics hire.”
Creating Value For Routine...
15 product features now tracked with detailed events,
50+ business metrics answering key questions modelled.
And 4 core dashboard providing insights automatically.
Get In Touch
Our friendly team are always on hand to answer questions, troubleshoot problems and point you in the right direction.
