Analytics · Reforge Foundation
The Data Scaling Framework
Crystal Widjaja's 3-stage data maturity model: Data Informed → Data Driven → Data Led. How to build the analytics, instrumentation, infrastructure, and team that match your company's actual data maturity.
Three stages of data maturity
Crystal Widjaja's data scaling framework distinguishes three stages of organizational data maturity:
- Data Informed. The organization has data, looks at it, and uses it to support decisions — but decisions are still primarily intuition-driven, with data as confirmation or sanity-check.
- Data Driven. Decisions are made from the data. The team has instrumented its key loops and metrics, runs structured experiments, and acts on results rather than gut.
- Data Led. Data shapes strategy, not just tactics. The organization can predict outcomes from leading indicators, runs sophisticated models (MMM, incrementality, LTV cohorting), and integrates data into how products are built.
Four capabilities within each stage
At each stage, four capabilities must be built in tandem:
- Infrastructure. The data warehouse, the instrumentation, the pipelines, the schemas.
- Analytics. The dashboards, the analyses, the experimentation framework, the modeling.
- Operations. The data cadence — who reviews what, how often, what decisions ladder from data.
- Team. The right hires at the right time — generalists early, specialists as scale demands.
Mismatched maturity across these four is the most common failure pattern. A Data-Driven analytics team built on Data-Informed infrastructure produces beautiful dashboards with unreliable data.
RGM experts say
The most common error in scaling data: hiring sophisticated analysts before the infrastructure can support them. A senior analytics hire spends their first six months waiting for clean data instead of producing insights. The fix: build infrastructure first, then add analytics talent that can use it.