From Raw Data to Actionable Insight: Master the 3 Recommendation Stages

Three Stages of Providing Recommendations

1. Give Data – Present raw metrics without interpretation, e.g., “New users 3 million, next‑day retention 65 %, 7‑day retention 17 %”. This describes facts but offers little business value.

2. Give Conclusion – Interpret the data and state a conclusion, such as “The new product’s performance exceeds industry standards.” This adds analysis but still lacks actionable guidance.

3. Give Viewpoint – Combine conclusion with concrete, feasible suggestions based on business understanding, e.g., “Channel F shows high retention and payment rate; increase ad spend on this channel.” This stage delivers actionable insight.

Common Pitfalls for Junior Analysts

Unclear analysis purpose – Simply stacking data without a clear business question leads to meaningless reports.

Useless advice – Generic statements like “reduce churn” without concrete steps provide no help.

Unimplementable suggestions – Recommendations that ignore cost or feasibility, such as “lower prices dramatically,” are unlikely to be adopted.

To avoid these issues, analysts should define clear objectives, dig deeper into data to uncover root causes, and propose specific, cost‑aware actions that align with business KPIs.

Developing a data‑thinking mindset—understanding data acquisition, cleaning, visualization, and linking analysis to business goals—is essential for progressing from basic reporting to strategic insight.