Learning Objectives
By the end of this chapter, you will be able to:
- Define Prescriptive Analytics and the key question it answers.
- Explain how prescriptive analytics builds on predictive analytics.
- Describe the main techniques used in prescriptive analytics, such as optimization and simulation.
- Provide business examples of prescriptive analytics, including dynamic pricing and next best action.
Prescriptive Analytics: From Insight to Action
Prescriptive Analytics is the most advanced form of business analytics. It goes a step beyond predicting the future by recommending specific actions that a business should take to achieve a desired outcome. It is focused on providing advice and guiding decision-making.
Figure 1: Prescriptive Analytics Overview
If descriptive analytics tells you what happened and predictive analytics tells you what will happen, prescriptive analytics tells you what you should do about it.
flowchart TB
PRED["🔮 Predictions\nfrom Predictive\nAnalytics"]
PRED --> TECH
subgraph TECH["Prescriptive Techniques"]
OPT["⚙️ Optimization\nFind Best Solution"]
SIM["🎮 Simulation\nTest Scenarios"]
RULES["📋 Rule-Based\nSystems"]
AB["🔬 A/B Testing\nExperiment"]
end
TECH --> REC["🎯 Recommendations\n'Do This!'"]
REC --> ANS["❓ 'What Should We Do?'"]
style REC fill:#2e7d32,color:#fff
style ANS fill:#1565c0,color:#fff
Figure 2: Prescriptive Analytics Flow
The Question It Answers: “What should we do?”
Prescriptive analytics aims to answer questions like:
- What is the best marketing offer to send to this specific customer to maximize the chance of a sale?
- What is the optimal price for our product right now to maximize revenue?
- What is the most efficient route for our delivery trucks to take today?
- Given our production constraints, which products should we prioritize making this week?
How It Works
Prescriptive analytics builds on the foundation of predictive analytics. It takes the predictions about future outcomes and combines them with business rules, constraints, and optimization techniques to recommend the best course of action from a set of alternatives. It essentially simulates the likely outcome of each possible action and recommends the one that will lead to the best result according to the business’s goals (e.g., maximizing profit, minimizing cost, or increasing market share).
Key Techniques and Tools
-
Optimization: This is a core technique in prescriptive analytics. It involves finding the best possible solution from a set of alternatives under a given set of constraints. For example, finding the optimal production schedule to meet demand while minimizing manufacturing costs.
-
Simulation: This involves creating a computer model of a real-world system (like a supply chain or a factory floor) to see how it would behave under different conditions or in response to different actions. This allows businesses to test out decisions in a virtual environment before implementing them in the real world.
-
Rule-Based Systems: These systems use a set of predefined business rules (e.g., “IF a customer has been with us for more than 5 years AND their recent activity has dropped by 50%, THEN offer them a loyalty discount”). These rules are often combined with predictions to automate decision-making.
-
A/B Testing: A randomized experiment with two variants (A and B). It is a form of prescriptive analytics where the system tests different actions on different segments of users to see which one performs better and then recommends the winning action.
Business Examples
-
Dynamic Pricing: A ride-sharing app like Uber uses prescriptive analytics to constantly adjust its prices. It predicts future demand and driver supply and then recommends a price (surge pricing) that will balance the two, maximizing the number of completed rides.
-
Next Best Action (NBA): This is a common application in marketing and customer service. A prescriptive model analyzes a customer’s profile and recent behavior and then recommends the “next best action” for a salesperson or service agent to take. This could be a specific product to recommend, a particular marketing offer to make, or a proactive support call.
-
Supply Chain Optimization: A logistics company can use prescriptive analytics to determine the most efficient delivery routes for its fleet of trucks. The model would take into account predictions about traffic, weather, and delivery windows, and then recommend the optimal route for each truck to minimize fuel costs and delivery times.
Prescriptive analytics represents the future of data-driven decision-making, moving from simply providing insights to actively guiding and automating business decisions.
Summary
Prescriptive Analytics is the final frontier of business analytics, answering the question, “What should we do?” By combining the forecasts from predictive analytics with optimization and simulation techniques, it recommends specific actions to help a business achieve its goals. From the dynamic pricing of an airline ticket to the “next best action” recommended to a salesperson, prescriptive analytics is turning data-driven insights into automated, optimized decisions.
Key Takeaways
- Prescriptive analytics recommends actions to achieve a desired outcome.
- It builds on predictive analytics by adding optimization and simulation.
- Key applications include dynamic pricing, next best action, and supply chain optimization.
- It represents a shift from supporting decisions to automating them.
Discussion Questions
- What is the relationship between predictive and prescriptive analytics?
- Google Maps recommending the fastest route based on current traffic is an example of prescriptive analytics. What data does it likely use, and what is the “action” it is recommending?
- What are the potential ethical concerns of using prescriptive analytics, for example, in setting prices for different customers?
