Ready to turn your data into your next big win? Start with the powerful analytical thinking model 'What-Why-What Next-What To Do' and watch your insights come alive!
In our increasingly data-driven world, turning raw information into meaningful action isn’t just a skill — it’s your business’s superpower! But simply collecting data isn’t enough. You need a clear, structured approach to ask the right questions, interpret results, and make smart decisions.
That’s where the analytical and strategic thinking model comes in. In this guide, we’ll break down the five essential frameworks — descriptive, diagnostic, predictive, prescriptive, and real-time analytics — with practical examples and tips for when and how to use them. Let’s get started!
🧩 What is the Analytical Thinking Model?
While it's not a formal, named "framework" in the same way as AARRR or HEART (which are specific sets of metrics), it is a fundamental and widely used analytical process model that underpins data analysis and decision-making in product management.
Key point! It's a process, not a list of metrics: Unlike AARRR or HEART, which define what metrics to track, the Analytical Thinking Model describes how to think about and use those metrics. It's the "why" and "what's next" of data analysis.
It’s commonly referred to as the Four Types of Analytics Framework (sometimes also called the Analytics Maturity Model).
It provides the context for the other frameworks (What-Why-What Next-What To Do), which are like a roadmap for your data journey. It helps you:
- Define what questions to ask.
- Organize and prepare your data.
- Choose the right tools and methods.
- Interpret your findings and turn them into action.
Without a framework, analysis can feel like wandering in circles. With one, you go from data overload to clarity and impactful insights.
Let’s explore it in detail and show how this model aligns with and enhances the other frameworks.
🗂️ 1. Descriptive Analytics: “What Happened?”
✅ When to use it:
Use descriptive analytics when you need a clear snapshot of your business’s past and present. It’s perfect for performance reviews, routine reporting, and trend spotting. Descriptive analytics helps you establish a solid starting point for deeper analysis.
🎯 Focus: Objectivity and clarity on the fact of change.
💡 Business questions it responds to:
- How many sales did we make last quarter?
- What was our customer churn last year?
- How many new customers/patients did we acquire last month?
- Which product category had the highest revenue?
👉 Example: DAU (daily active users) dropped by 15% last week. Sales dropped 15% last quarter.
🔧 How to use it:
Gather historical data, clean and organize it, then visualize it using charts, dashboards, or simple reports. Tools like Excel, Python, Tableau, or Power BI make this easy and accessible.
Aligns with frameworks:
- KPI Trees (to surface changes)
- Vanity Metrics Framework (to ensure you’re not misled by surface-level stats)
- OMTM / NSM (monitoring your core metric)
- AARRR or HEART (to identify the metrics that have changed)
👍 Pros: Easy to implement, straightforward, and provides clear summaries.
👎 Cons: Doesn’t explain why things happened or what might happen next.
📌 Learn more about Descriptive Analytics
🔍 2. Diagnostic Analytics: “Why Did It Happen?”
✅ When to use it:
Turn to diagnostic analytics when you notice unexpected trends or changes and want to know the reasons behind them. It’s great for investigating anomalies, understanding business drivers, and validating assumptions. Use it to add context and depth to your descriptive data.
🎯 Focus: Root cause analysis using structured diagnostics.
💡 Business questions it responds to:
- Why did our sales drop in Q2, even though marketing spending increased?
- What caused the spike in customer complaints?
- Why did Product A outperform Product B in a certain region?
👉 Example: Drop in DAU (daily active users) correlates with a failed push notification update.
🔧 How to use it:
Drill down into your data using segmentation, correlation analysis, and root cause investigation. Use pivot tables, advanced queries, and analytical tools that allow deep dives, like SQL, Python, or advanced BI software, to visualize the findings.
Aligns with frameworks:
- KPI Trees (break down the metric into sub-components)
- AARRR (e.g., activation or retention stage breakdown)
- HEART Framework (was task success or happiness impacted?)
- GAME Framework (did actions lead to expected measurements?)
👍 Pros: Reveals hidden relationships and clarifies causes.
👎 Cons: Can be time-consuming and often needs more advanced skills.
🔮 3. Predictive Analytics: “What Will Happen?”
✅ When to use it:
Use predictive analytics to make informed decisions about the future. It's essential for forecasting sales, predicting customer behavior, and planning resources. By leveraging historical data, statistical models, and machine learning algorithms, predictive analytics provides a reliable estimate of future outcomes, helping you stay ahead of the market.
🎯 Focus: Anticipation and risk assessment.
💡 Business questions it responds to:
- How many new subscribers will we gain next month?
- What is our projected sales revenue for the next quarter?
- Which customers are likely to churn soon?
- What’s the risk of supply chain delays next quarter?
👉 Example: Use regression models to forecast demand or customer behavior.
🔧 How to use it:
Train statistical or machine learning models on historical data. Use tools like Python, R, or software like SAS and scikit-learn to build forecasts and run probability scenarios.
Aligns with frameworks:
- OKRs (predict goal attainment)
- GIST (evaluate thresholds and signals)
- SMART Metrics (Is the forecast measurable and time-bound?)
- OMTM (predict impact on the current key metric)
👍 Pros: Provides foresight for better planning and proactive actions.
👎 Cons: Requires high-quality data and expertise; predictions are probabilities, not guarantees.
📌 Explore Predictive Analytics
🚦 4. Prescriptive Analytics: “What Should We Do?”
✅ When to use it:
This is the most advanced type of analytics. Prescriptive frameworks not only forecast future events but also suggest actions to take to optimize outcomes or mitigate risks. They tell you, " How Can We Make It Happen?". It’s perfect for resource allocation, pricing strategies, and operational optimization. This framework adds maximum value when you’re ready to automate or optimize complex decisions.
🎯 Focus: Actionability and ownership.
💡 Business questions it responds to:
- What’s the optimal pricing to maximize profit?
- How should we allocate marketing budgets for the highest ROI?
- What’s the best delivery route to reduce costs?
- What marketing channels should we invest in to acquire the most valuable customers?
👉 Example: Use AI tools to suggest personalized product recommendations or dynamic pricing. Roll back the notification update and run a controlled A/B test.
🔧 How to use it:
Combine predictive models with optimization algorithms and simulations. Leverage AI tools that generate recommended actions. Tools like IBM Decision Optimization or Google OR-Tools can help.
Aligns with frameworks:
- ICE Scoring (prioritize fixes by impact, confidence, effort)
- OKRs (adjust or set new key results)
- CLEAR Framework (ensure action is collaborative, appreciable, and refinable)
- SMART Criteria (make sure the action is specific and time-bound)
👍 Pros: Turns insights into direct, actionable strategies.
👎 Cons: Complex to build, requires advanced tools and skilled teams.
To enhance the Analytics Maturity Framework (What-Why-What Next-What To Do), I always recommend adding Real-Time Analytics where appropriate.
Real-Time Analytics is not just about speed — it’s about reducing the latency between data, insight, and action. When layered onto the 4-question model, it transforms your analytics from historical review to live operational intelligence.
⏱️ 5. Real-Time Analytics: What’s Happening Right Now?
✅ When to use it:
While the four pillars focus on the type of insight, the real-time analytics framework is designed to process and analyze data as soon as it's generated, delivering insights with minimal delay, often in milliseconds or seconds, or in near real-time in minutes. When immediate action or near-instant insights are critical. This is crucial for dynamic environments where events unfold rapidly.
🎯 Focus: Trigger automated actions or rapid human response.
💡 Business questions it responds to:
- Is there a fraudulent transaction happening right now?
- What is happening in the emergency department now?
- How many ambulances are heading to our hospital, and what is the status of the patient in them?
- What product should we recommend to a customer as they browse our website?
- Which ad is driving clicks this minute?
👉 Example: Flags “What Happened?” instantly. Banks leverage it to detect suspicious transactions instantly. Hospitals adopt real-time analytics to enhance patient care and manage operations.
🔧 How to use it:
Set up streaming data pipelines with tools like Apache Kafka, AWS Kinesis, or Spark Streaming. It often involves building sophisticated data pipelines that capture, process, and deliver data continuously, as well as creating real-time dashboards and alerts to detect anomalies or trigger immediate action.
Aligns with frameworks:
- KPI Trees (live drill-down from high-level drop to root node (e.g., payment failure rate spiking now))
- OMTM / NSM (e.g., monitor your One Metric That Matters in real time with alerts)
- AARRR Framework (detect real-time funnel drop-offs (e.g., sign-up completion halved after deploy))
- GIST (use live signals to evaluate thresholds and trigger goal adjustments)
- OKRs (track progress toward objectives with live data — no lag)
👍 Pros: Provides instant insights and supports quick reactions.
👎 Cons: Technically demanding and needs robust infrastructure.
🧭 How to Pick the Right Framework
The truth is, there’s no single “best” data analytics framework — the right choice depends on your goals, the type of data you have, and the resources your organization can commit. Keep these guiding principles in mind:
1. Define Clear Objectives: Start with the end in mind. Are you trying to understand past performance, predict future trends, or automate smart decisions? Clear goals will guide your choice.
2. Know Your Data: Assess what kind of data you have — is it structured or unstructured? Large or small? This shapes which tools, techniques, and frameworks will work best.
3. Plan for Growth: As your business expands, so will your data. Choose a framework that’s flexible and scalable to meet future demands.
4. Protect Data Quality: Great insights come from great data. Make sure your data is clean, accurate, and well-managed. Strong data governance is non-negotiable.
5. Think Iteratively: Data analytics isn’t a one-and-done project. Keep refining your frameworks as your business evolves and new tools become available.
When you choose and apply the right framework thoughtfully, you turn raw data into a powerful strategic asset that drives smart, confident decisions.
🤝 Wrapping Up
A well-chosen analytics framework helps you transform raw data into decisions that matter. Whether you’re analyzing yesterday’s numbers or reacting to events unfolding this very second, there’s a framework ready to guide you.
✨ Remember: It’s not about having more data. It’s about asking better questions.