Granular subscription analytics in just a few weeks
Objective
Track complex subscriber lifecycles, multi-level revenue metrics, and promotional campaign effectiveness.
Obstacle
Cinobo lacked the infrastructure to accurately track subscriber behaviour across their complex business model.
Outcome
Comprehensive dashboards revealing subscriber lifecycle patterns, revenue trends, and cohort-based retention metrics.
Background
Cinobo is a leading streaming platform serving audiences across Greece, Cyprus, and Poland. With a growing subscriber base and complex promotional strategies, the team recognised the need to move beyond basic reporting to establish a sophisticated analytics foundation that could untangle the intricacies of subscription commerce.
They partnered with 173tech to build a modern data ecosystem capable of tracking subscribers across multiple dimensions: packages (Basic, Full, Plus, Premium), plans (Monthly, Yearly, 6-month), specific offer IDs with unique commercial terms, and geographic markets. The goal: transform raw transactional data into strategic insights that could drive retention, optimise pricing, and measure the true impact of promotional campaigns.
Challenges
Complex Subscriber States: Subscribers could exist in multiple states simultaneously: trial vs paying, auto-renew enabled (AR_ON) vs cancelled but still active (AR_OFF) vs payment failure grace period (Limbo). The system needed to accurately represent these states at any point in time while tracking transitions between them.
Five-Level Hierarchy Complexity: Cinobo’s business model required tracking every metric across five distinct hierarchy levels: Global (entire subscriber base), Package (subscription tiers), Plan (billing frequencies), Offer ID (specific commercial terms), and Market (geographic regions). Each level needed to maintain consistency while allowing independent analysis.
Sophisticated Promotional Logic: With multiple overlapping promotional campaigns offering various discounts and trial periods, the team needed to attribute revenue correctly, track promotional consumption, and understand when promo codes were active vs expired—complicated by the lack of structured promo duration data.
Solution
Rapid Infrastructure Setup: We worked collaboratively with Cinobo’s engineering team to establish the modern data stack: BigQuery as the data warehouse, dbt for transformation logic hosted in Cinobo’s GitHub repository, and Metabase for visualisation. We coordinated the migration of Airflow pipelines from Postgres to BigQuery, establishing separate datasets for development, CI, and production environments.
Data Modelling in dbt: We built scalable dbt architecture that modelled daily subscription snapshots, lifecycle events, and revenue transactions, enabling Cinobo to analyse subscriber behaviour from first trial through churn and return, all while maintaining accuracy across different subscription tiers and promotional periods.
Tight Collaboration: With only four weeks, we ruthlessly prioritised core functionality over nice-to-have features. Daily check-ins ensured blockers were cleared immediately and decisions were made in hours, not days. When we encountered data quality issues, ambiguous business logic, or technical constraints, we worked with Cinobo’s team to find pragmatic solutions rather than perfect ones, enabling progress while maintaining accuracy.
Impact
React Faster: The analytics infrastructure we built fundamentally transformed how Cinobo operates. Before our engagement, the team spent days manually extracting data from Postgres and reconciling numbers across fragmented reports. Now, daily automated ingestion feeds comprehensive dashboards that refresh overnight, giving leadership immediate visibility into subscriber trends, revenue patterns, and promotional effectiveness.
Untangling Complex Relationships: The multi-level hierarchy architecture proved particularly valuable for Cinobo’s commercial strategy. By tracking subscriber events at user, plan, and offer levels simultaneously, the team can now understand not just aggregate trends but the specific paths subscribers take through their product ecosystem. They can identify which packages drive the highest lifetime value, which promotional offers generate the best long-term retention, and where subscribers are most likely to upgrade or churn.
Ready For Growth: Perhaps most significantly, the foundation we established positions Cinobo for continued growth. The scalable dbt architecture means new metrics can be added as business questions evolve. The daily snapshot model enables sophisticated cohort analysis and predictive modelling that wasn’t previously possible. When Cinobo expands to new markets, the infrastructure scales naturally to accommodate additional currencies and VAT structures.
Creating Value For Routine...
3 classification levels for subscriber events tracking: new, renewal, returning, upgrade, & churn ,
Modelled across 5 hierarchy levels,
Enabling granular subscription analysis in just a few weeks.
Get In Touch
Our friendly team are always on hand to answer questions, troubleshoot problems and point you in the right direction.
