Growth Signals
+34% retention across distributed teams.
34%
Retention Uplift
Retention improved across distributed teams using the platform.
6 mo
From Zero to Live
Designed, built and shipped from scratch in six months.
8
Engineers
A senior-led team delivering product and intelligence together.
Client
HR technology company serving agencies, IT and consultancy firms
Market
HR technology & the future of work
Scale
Distributed teams across multiple organisations
Duration
6 months, built from scratch
Team
8 engineers
Engagement
Managed Delivery
Delivery hub
Zagreb
Capability
Platform & Managed Services
Stack
Python, ML, GPT integration, HR platform APIs
The challenge
Invisible disengagement
Early signs of attrition surfaced only after people had already decided to leave.
Fragmented signals
Engagement data sat across disconnected HR tools with no shared view.
From scratch
A new product had to be designed, built and shipped inside six months.
The approach
Signal aggregation
Engagement signals unified from across HR platform APIs into one model.
Retention modelling
ML that flags teams and individuals trending toward disengagement early.
Guided interventions
GPT-assisted recommendations that turn a signal into a next action.
What we would tell you before starting
Retention signals are only as good as the engagement data feeding them. Where an organisation’s HR tooling recorded little, the model had little to work with. The first weeks were integration and data plumbing, not modelling.
Last reviewed