Overview of Consumption Insights

Overview Dashboard

• Provides a comprehensive snapshot of total billable usage trends.
• Includes historical data and forecasting for the current and upcoming months.
• Identifies top 5 products and top 5 customers based on billable usage amounts.

Account Insights

• Displays detailed usage trends for each customer account.
• Opportunity Scoring ranks customers based on increasing usage patterns, highlighting up-sell or cross-sell potential.
• Risk Scoring identifies accounts with declining usage, indicating potential churn risks.

Billable Amount Drivers

• Analyzes billable amounts by product or customer to uncover key revenue drivers.
• Provides insights into high-performing products and customers.

Forecasting & Anomaly Detection

• Predicts future revenue and usage trends using historical time series data.
• Identifies anomalies that may require business intervention.
• Enables proactive decision-making to maximize revenue opportunities.

How the Forecasting Algorithm Works

The forecasting process takes your past performance data and intelligently extends it forward. Here are the key elements it considers:

• Historical Data (Timestamps and Values):
  • The algorithm requires a series of historical data points, each associated with a specific date or time.
  • The quality and completeness of this historical data are vital, as the algorithm learns directly from it.
• Forecast Horizon (Number of Periods):
  • You define how far into the future you want the forecast to extend. This is specified as a number of "periods" (e.g., 3 months, 6 quarters, 12 weeks). The algorithm will generate predicted values for each of these future periods.
• Seasonality:
  • Many business metrics exhibit seasonal patterns – predictable fluctuations that occur at regular intervals (e.g., spikes in sales at year-end, dips in usage during holidays).
  • The forecasting algorithm can account for these patterns if identified. You can specify the expected seasonality (e.g., "monthly," "quarterly," "yearly," or "none") to ensure the forecast accurately reflects these recurring trends. If seasonality is present in your data, incorporating it leads to a more accurate prediction.

How Anomaly Detection Works

Zuora's anomaly detection uses sophisticated statistical algorithms to analyze your time-series data (e.g., daily usage, monthly revenue, transaction counts). The core process involves:

• Analyzing Historical Trends: The algorithm first learns what "normal" looks like for your specific data. It examines past values and their timestamps to understand typical patterns, seasonality, and expected fluctuations.
• Identifying Deviations: Once the normal pattern is understood, the algorithm monitors new or existing data points. If a data point falls outside a predefined range of what's considered normal, it's flagged as an anomaly.
• Applying Different Detection Methods: Zuora uses various detection methods to help users identify anomalies. The various methods help users spot both sudden, sharp changes as well as subtle, prolonged unusual behavior.

What makes a data point an anomaly?

An anomaly in Zuora is a data point that deviates significantly from its typical or expected pattern based on historical trends. Our statistical algorithms first analyze your past data to understand its normal behavior, including usual values, fluctuations, and any recurring patterns. They then establish a "normal" range, and any data point that falls outside these learned boundaries—whether it's a sudden spike, an unexpected drop, or a prolonged unusual value—is flagged as an anomaly, indicating it's statistically unusual and warrants your attention.

Why is risk score important?

By understanding your Risk Score, you can proactively identify accounts that might need attention. This allows you to engage with these customers, understand their challenges, and offer solutions to help them continue to succeed with our products.

Why is health score important?

The Health Score helps you identify your most successful and engaged customers. These are often the customers who are great candidates for upsell opportunities, testimonials, or serving as references. By understanding who your "healthy" customers are, you can nurture these relationships and replicate their success across other accounts.

How Root Cause Analysis works

This feature dives deep into your meter data to analyze all the unique values across every dimension associated with a given meter. For example, if your meter data includes dimensions like Sales Channel, Customer Name, Segment, and Country, the Root Cause Analysis will examine the contribution of each individual value within those dimensions.

The results are presented in an intuitive heatmap visualization (as shown in the accompanying image). This heatmap provides two critical pieces of information for each dimensional value:

1. Relative Contribution: It shows how much each specific dimensional value contributed to the overall period-over-period change in your meter. For instance, you can see if a certain Sales Channel was responsible for a large portion of an increase, or if a particular Country drove a significant decrease.
2. Percentage Change: Alongside the contribution, you'll see the percentage change for that specific dimensional value, highlighting the magnitude of its own change.

Identifying Drivers of Change

By using the Root Cause Analysis, you can quickly identify which specific factors are driving the changes in your meter trends. This enables you to:

• Pinpoint Growth Drivers: Understand which products, customer segments, or sales channels are contributing most to consumption growth.
• Diagnose Declines: Quickly identify problematic areas, such as a specific product, region, or customer group, that are causing a drop in consumption.
• Validate Initiatives: See the direct impact of business initiatives on consumption trends by analyzing changes across relevant dimensions.

This powerful visualization and analysis capability provides immediate, actionable insights into the forces shaping your consumption-based revenue.

How Drilldown Analysis works

This feature analyzes your meter data for a selected timeframe and calculates the contribution of every unique value within each of your consumption dimensions (e.g., Sales Channel, Customer Name, Country, Method).

The results are presented in a clear, interactive visualization (as shown in the accompanying image). For each dimensional value, you will see its percentage contribution to the overall total usage within the period you've selected. This allows you to:

• Understand Usage Distribution: See at a glance which segments, products, regions, or other factors represent the largest portions of your total consumption.
• Identify Top Performers: Quickly pinpoint your biggest usage contributors, whether they are specific customers, sales channels, or product offerings.
• Analyze Segment Performance: Break down overall usage by any relevant category to understand the relative strength of different segments.

Gaining Deeper Insights into Your Usage

By utilizing Drilldown Analysis, you gain immediate insights into the structure of your consumption. This enables you to:

• Prioritize Efforts: Focus your resources on the most impactful customer segments or sales channels.
• Optimize Product Strategy: Understand which product configurations or features are driving the most usage.
• Monitor Performance by Segment: Continuously track the contribution of various operational and customer segments to your overall consumption.

This feature provides a comprehensive breakdown, allowing you to drill down into your data and understand the fundamental composition of your usage trends.

Forecasting Revenue

• Predict future billing amounts based on customer usage trends.

• Period over period comparison to identify seasonal variations and align business planning accordingly.

Upsell and Cross-Sell Optimization

• Use opportunity scores to target high-growth accounts.

• Enhance customer engagement strategies based on detailed usage patterns for particular units of measure.

Churn Risk Mitigation

• Leverage risk scoring to identify accounts with declining consumption.
• Take proactive steps to re-engage at-risk customers.