# Analytics That Move Decisions, Not Decks

## The Bottom Line

Most finance dashboards are furniture: pretty, polished, ignored. Charishma builds analytics from the harder question - **detect the problem before it happens.** Her Power BI predictive-maintenance work used statistical modeling and machine learning to surface anomalies as actionable signals, delivering **measurable downtime reduction** - a business outcome, not a dashboard.

## The Industry Default

Dashboards are built to *display* - charts, KPIs, and green numbers that everyone looks at and nobody acts on. Data is pulled, visuals are refreshed, and "analytics" becomes screenshots in a deck. Reporting analysts wait to be asked for a chart rather than volunteering the insight.

## Charishma's Approach

- **Predict, don't describe.** Led a predictive maintenance analytics project using statistical modeling and machine learning - built to detect operational anomalies before they happened.
- **From signal to decision.** Turned raw operational data into signals stakeholders could act on and presented findings directly - not a chart waiting to be interpreted.
- **Measured by outcomes.** Held analytics to a hard standard: if it doesn't change a decision or improve an outcome, it's not done. The result was measurable downtime reduction.

## Why It Matters - So What / Now What

**So what:** A dashboard that describes the past is overhead. Analytics that predict the future is a cost lever - downtime reduction flows straight into capacity, throughput, and margin.

**Now what:** Decision-makers get early-warning signals that let them act before the failure - converting analytics spend into operational savings.

## The Difference In One Line

Builds the dashboard → **builds the decision that reduces downtime.**