RavenStack SaaS Analytics

Conversion Intelligence • Support Load • Churn & Retention

---

1. Project Overview

RavenStack is a stealth-mode SaaS startup developing AI-driven team collaboration tools.

During its private pilot, RavenStack collected information about:

• Customer accounts
• Sign-ups and trials
• Subscription plans
• MRR and ARR
• Upgrades and downgrades
• Feature usage
• Product errors
• Beta feature usage
• Support tickets
• Customer satisfaction
• Churn events
• Churn reasons
• Refunds

The objective of this project was to transform these disconnected datasets into a decision-oriented SaaS analytics solution.

Instead of simply reporting what happened, the analysis focused on answering:

What drives conversion?

What drives support workload?

What drives churn?

Where should stakeholders take action?

The project combines Python-based exploratory data analysis, hypothesis-driven investigation, data modeling, DAX and Power BI dashboarding to provide an end-to-end business analytics solution.

---

2. Business Problem

RavenStack's leadership needs to understand the health of the product before scaling publicly.

The analysis was structured around three major business objectives:

Objective 1 — Conversion

Identify the customer characteristics and behaviors associated with successful conversion and upgrades.

Objective 2 — Support Load

Understand which customers, industries, plans and product behaviors generate the greatest support demand.

Objective 3 — Churn & Retention

Identify why customers leave, which segments are most vulnerable, and what signals could be used for proactive retention.

---

3. Key Business Questions

Conversion

• Which industries have the strongest upgrade rates?
• Which referral sources generate higher-quality customers?
• Do trial customers engage differently from non-trial customers?
• Does early feature adoption relate to upgrades?
• Do beta users convert differently?
• Which customer segments are more likely to become Enterprise customers?
• Which segments generate the highest revenue after conversion?

Support Load

• Which industries generate the most support tickets?
• Which plan tiers create the greatest support workload?
• Do beta users require more support?
• Which product areas generate the most support activity?
• Which ticket priorities take the longest to resolve?
• Which issues are associated with escalations?
• Where can product improvements reduce support workload?

Churn & Retention

• Which industries have the highest churn?
• Which plans have the highest churn?
• What are the most common reasons for churn?
• Are downgrades associated with subsequent churn?
• Does billing frequency differentiate churn?
• Which churn reasons create the greatest financial impact?
• How much recurring revenue is lost through churn?
• Which customer segments should receive proactive retention attention?

---

4. Dataset

The project uses five interconnected datasets.

Dataset	Description
ravenstack_accounts.csv	Customer and account information
ravenstack_subscriptions.csv	Subscription, billing, plan and revenue information
ravenstack_feature_usage.csv	Product feature usage and error information
ravenstack_support_tickets.csv	Customer support interactions
ravenstack_churn_events.csv	Churn events, reasons and financial impact

The datasets are connected through customer and subscription identifiers.

---

5. Data Model

The Power BI data model connects the five datasets to allow customer-level analysis across the entire SaaS lifecycle.

Simplified Model

                     ACCOUNTS
                        |
      +-----------------+------------------+
      |                 |                  |
      ↓                 ↓                  ↓

SUBSCRIPTIONS SUPPORT CHURN EVENTS | ↓ FEATURE USAGE